University of Stuttgart
Institute of Geodesy
Assessing the performance of different
classification methods
to detect inland surface water extent
Bachelor Thesis
Geodäsie und Geoinformatik
University of Stuttgart
Alexander Walton
Stuttgart, August 2015
Supervisors: Prof. Dr.-Ing. Nico Sneeuw
University of Stuttgart
Dr.-Ing. Mohammad J. Tourian
University of Stuttgart
M.Sc. Omid Elmi
University of Stuttgart

Erklärung der Urheberschaft
Ich erkläre hiermit an Eides statt, dass ich die vorliegende Arbeit ohne Hilfe Dritter und ohne
Benutzung anderer als der angegebenen Hilfsmittel angefertigt habe; die aus fremden Quellen
direkt oder indirekt übernommenen Gedanken sind als solche kenntlich gemacht. Die Arbeit
wurde bisher in gleicher oder ähnlicher Form in keiner anderen Prüfungsbehörde vorgelegt
und auch noch nicht veröffentlicht.
Ort, Datum Unterschrift
III

Abstract
In recent decades, political as well as environmental conflicts about the Earth’s water resources
became a significant issue with constantly growing importance all over the world. These issues
include flooding as well as drying shrinkage of seas and rivers. In order to estimate the dimen-
sion of these impacts, reliable frequent observation of the surface water over a long period of
time is essential.
By making use of the special spectral reflectance properties of water, especially in higher spec-
tral ranges, it is possible to distinguish between water and other surface materials and create
thematic maps using different classification methods. These methods can be based on super-
vised algorithms which make use of training data to classify an image. On the other hand they
can be automatized computing algorithms which assign pixels to a class without any prior
knowledge. The latter are referred as unsupervised methods.
Assessing the performance of these various methods will be task of this thesis. Here in this
study four satellite images of Landsat 7 are selected for our case study. The images show
the region around the Po River in northern Italy. The valley alongside the river is one of the
strongest economic and agricultural region in Italy but it also suffers from regular flooding,
especially in the delta region around Ferrara close to the Adriatic Sea.
The different classification methods are implemented in the software ENVI which is commonly
used by remote sensing professionals to process and analyse geospatial imagery.
To make a statement about the accuracy of the single classification methods every classifier
undergoes a few essential steps: A binary water mask is created in which the river width is
measured at two selected gauge spots. At these spots precise in-field measurements were ap-
plied regularly in recent years and the results serve as reference values for a comparison in
between the classification methods.
In general the supervised methods performed better than the unsupervised methods. The best
performance is approached by the mahalanobis distance classification which is based on prob-
ability statistics with consistent covariances. This way the pixels can be classified by calculating
their minimum euclidean distance in spectral space.
The results of this study are limited by the low resolution of 30 m which does not leave vast
room for interpretation. More reliable results could be achieved by measuring effective widths
alongside the river and apply an area-wise comparison. However approximate estimates about
the performance of the different classification methods and good first impressions of the advan-
tages and disadvantages are given.

VII
Contents
1 Introduction 1
1.1 Energy Sources and Wavelength Ranges . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Spectral reflectance properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Satellite Imagery Missions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.4 Landsat 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.4.1 Primary Instrument: ETM+ . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4.2 Landsat 8 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.5 Band sensitivities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.5.1 Display of remote sensing data . . . . . . . . . . . . . . . . . . . . . . . . . 8
1.6 Vegetation Indices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.7 Case Study and Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
1.8 Outline of the Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2 Classification Methods 13
2.1 Supervised classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1.1 Maximum Likelihood Classification . . . . . . . . . . . . . . . . . . . . . . 14
2.1.2 Minimum Distance Classification . . . . . . . . . . . . . . . . . . . . . . . . 15
2.1.3 Parallelepiped Classification . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.1.4 Mahalanobis Distance Classification . . . . . . . . . . . . . . . . . . . . . . 18
2.2 Unsupervised classification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.1 Clustering Criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.2.2 K-Means Clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
2.2.3 Isodata Clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3 Results 21
3.1 First Image: Spring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.1.1 Training Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.1.2 Classified Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.1.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.2 Second Image: Summer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.2.1 Training Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.2.2 Classified Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.2.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.3 Third Image: Autumn . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.3.1 Training data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.3.2 Classified Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.3.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.4 Fourth Image: Winter . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.4.1 Training Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.4.2 Classified Images . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
3.4.3 Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
VIII
4 Validation 35
4.1 In Situ Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
4.2 Classification Performances . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5 Conclusion and Outlook 43
5.1 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
5.2 Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
IX
List of Figures
1.1 Signal flow in a remote sensing system . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 The electromagnetic spectrum and the transmittance of the earth’s atmosphere . 2
1.3 Spectral signature of soil, vegetation and water with spectral bands of Landsat 7 3
1.4 Different optic and SAR missions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.5 The Landsat 7 satellite . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.6 Bandpass wavelengths for Landsat 8 . . . . . . . . . . . . . . . . . . . . . . . . . . 7
1.7 Displayed combined bands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.8 Location of the Po river . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1 Example of a histogram: water class in all six available bands1 . . . . . . . . . . . 15
2.2 Illustration of the use of thresholds . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.3 Parallelepiped boundaries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.4 A set of two dimensional parallelepipeds . . . . . . . . . . . . . . . . . . . . . . . 17
2.5 Regions of inseparability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
3.1 Training areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.2 Subset of the image recorded in spring as a negative colour image (432) . . . . . . 22
3.3 First image: Spring classification results . . . . . . . . . . . . . . . . . . . . . . . . 23
3.4 Spatial subset of the image recorded in summer . . . . . . . . . . . . . . . . . . . 25
3.5 Second image: Summer classification results . . . . . . . . . . . . . . . . . . . . . 26
3.6 Misclassified area of the summer image . . . . . . . . . . . . . . . . . . . . . . . . 28
3.7 Spatial subset of the autumn image . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
3.9 Errors in the mahalanobis distance classification . . . . . . . . . . . . . . . . . . . 29
3.8 Third image: Autumn classification results . . . . . . . . . . . . . . . . . . . . . . 30
3.10 True colour (321) and NDWI image of the whole swath . . . . . . . . . . . . . . . 31
3.11 NDWI subset with training areas . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
3.12 Fourth image: Winter classification results . . . . . . . . . . . . . . . . . . . . . . . 33
3.13 Error in the coastal region . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.1 Areas for a comparison with an in-field surveillance . . . . . . . . . . . . . . . . . 35
4.2 Water mask example: Gauge spot Sermide . . . . . . . . . . . . . . . . . . . . . . 36
4.3 Profile measurements alongside the Po river . . . . . . . . . . . . . . . . . . . . . 37
4.4 River profiles; water levels through the year . . . . . . . . . . . . . . . . . . . . . . 37
4.5 Classification performances in Sermide . . . . . . . . . . . . . . . . . . . . . . . . 38
4.6 Classification performances in Pontelagoscuro . . . . . . . . . . . . . . . . . . . . 39
4.7 Average percentaged deviation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
4.8 Absolute deviation from all eight measurements . . . . . . . . . . . . . . . . . . . 41

XI
List of Tables
1.1 Summary of sensor properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2 Landsat 7 details . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3 ETM+ Technical Specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.4 ETM+ Bands . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5 Band sensitivities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.1 Class means, ranges and standard deviation in the classification of the fourth
image . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3.1 Regions of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.2 Spring image statistics results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.3 Regions of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.4 Summer image statistics results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
3.5 Autumn image statistics results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
3.6 Winter image statistics results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.1 Profile widths 2001 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37

1Chapter 1
Introduction
1.1 Energy Sources and Wavelength Ranges
In order to construct an image from the Earth’s surface, reflected energy is measured using a
sensor mounted on an aircraft or spacecraft platform. The process from collecting data to the
resolution of an image is shown in a simplified view in Figure 1.1. The most obvious source of
energy would be reflected sunlight, so that the later formed image is in many ways similar to
the view we would have of the Earth’s surface from an aircraft [8]. Such a sensor is referred to
as a passive sensor. Alternatively the upwelling energy could also be radiated from the Earth
itself or provided by an active sensor, such as a laser or radar. In remote sensing wavelength
ranges far beyond the range of the human vision are used. Short-, long-waved infrared and
sometimes even ultraviolet wavelength ranges are established.
Figure 1.1: Signal flow in a remote sensing system [8]
2 Chapter 1 Introduction
As long as there is any source of energy available, almost every wavelength could be used
to measure the characteristics of the Earth’s surface. However, the wavelengths used in re-
mote sensing, particularly when imaging from spacecraft altitudes, are limited to a so called
atmospheric window [8]. Energy at some wavelengths is absorbed by the molecular constituents
in the Earth’s atmosphere, such as ozone, water vapour and carbon dioxide. The atmospheric
window describes an area in which there is little or no atmospheric absorption. As shown in
Figure 1.2 there are various amounts of windows in the visible and infrared regions and radio
wavelengths even possess almost complete transparency of the atmosphere.
Based on these significant absorption properties, optical remote sensing systems record data
from the visible through to near and mid-infrared range, which covers approximately 0.4–
2.5 µm [8].
Figure 1.2: The electromagnetic spectrum and the transmittance of the earth’s atmosphere [8]
1.2 Spectral reflectance properties
Earth surface objects absorb or reflect sunlight in different ways. The reflectance properties
depend on the materials physical and chemical state, which is why most of them have quite
complex absorption characteristics. The differences can be visualised in so called spectral re-
flectance curves as a function of wavelengths. Figure 1.3 shows a typical spectral reflectance
curve for three basic type of Earth features: green vegetation, dry bare soil and clear water.
The spectral reflectance properties of water are characterized by a high to almost total absorp-
tion at near infrared and beyond. Using linear combinations of different bands with their re-
spective wavelength ranges, water bodies or water containing objects can easily be detected
1.3 Satellite Imagery Missions 3
Figure 1.3: Spectral signature of soil, vegetation and water with spectral bands of Landsat 7 [10]
and delineated. We can distinguish between turbid water and clear water because of its higher
reflectance in the visible region and in a similar manner we can detect oil spills or industrial
wasted water or even algae colonies by making use of its high concentration of chlorophyll.
Chlorophyll strongly absorbs light at wavelengths in the blue and red ranges and shows high
reflection properties in the green region, therefore healthy vegetation appears in colours of
green in the human vision. The reflection in infrared ranges varies amongst different plant
species, with which a distinction between the species can be accomplished [10].
1.3 Satellite Imagery Missions
Optical and synthetic aperture radar (SAR) satellite imagery missions are capable of detecting
inland surface water extent [1]. The Landsat Multispectral Scanner System (MSS) launched in
1972 began the modern era of land remote sensing from space. It was equipped with a 80 m
spatial resolution sensor with 4 spectral bands, each about 0.1 µm wide. Today operational
satellite systems sample nearly all available parts of the electromagnetic spectrum with dozens
of spectral bands and spatial resolutions of better than 1 m [9]. The essential differences in
the several satellite imagery missions are characterized by the number and coverage of the
spectral bands, the spatial resolution, the coverage area and the temporal resolution. SAR
imaging systems are additionally classified according to the combination of frequency bands
but since this study is limited to optical images the SAR missions will not be of any further
interest. Figure 1.4 presents different optic and SAR missions.
The first remote sensing satellite capable of water body and wetland monitoring was the ad-
vanced very high resolution radiometer (AVHRR) and it is available since 1980. Its pixel size of
approximately 1 x 1 km has been improved over the year to reach spatial resolutions in range
of metres. For example one of the most recent satellite missions is the Landsat 8 mission with a
4 Chapter 1 Introduction
spatial resolution of 30 m. MODIS is capable of scanning the Earth’s surface in only two days
with a spatial resolution of 250 m [1].
Figure 1.4: Different optic and SAR missions [1]
Table 1.1 gives us a brief summary of various sensor properties. If the resolution would be
the most important parameter, Worldview and Geoeye would be the best choices available.
However, these sensors also differ in their number of spectral bands and the spectral range.
Airborne sensors provide the widest spectral range, followed by Landsat, SPOT and IRS. Air-
borne sensors suffer from range and time, therefore Landsat is the optimal choice if the spectral
and operational ranges are considered [11]. Landsat 7 data shows a great availability and is free
of charge. For those reasons Landsat 7 imagery is used in this study. Therefore, here we discuss
further about the properties of Landsat 7.
Table 1.1: Summary of sensor properties
Sensor Spatial resolution [m] Spectral range [µm] Temporal resolution [days]
Landsat 7 15 0.45–2.35 16
SPOT 2.5 0.50–1.75 5
IRS 5 0.50–1.70 5
Ikonos 1 0.45–0.85 3
Quickbird 0.61 0.45–0.90 3
Worldview 0.46 0.40–1.04 1.1
ALOS 2.5 0.42–0.89 2
Geoeye 0.41 0.45–0.90 3
Airborne 1–25 0.42–14.00 N/A
1.4 Landsat 7 5
1.4 Landsat 7
Landsat 7 is an Earth observing satellite launched in 1999. Back then no other system could
match the Landsat’s combination of synoptic coverage, high spatial resolution, spectral range
and radiometric calibration. Table 1.2 provides detailed information about the orbital parame-
ters. Unlike its predecessors, Landsat 7 is still active today. The very first Landsat satellite was
launched in 1972 and terminated in 1978. Landsat 5 was terminated in 2013 after a duration of
almost 30 years. Landsat 6 failed to reach the orbit in 1993 so a lot of hope was build on the
Landsat 7 mission.
Figure 1.5: The Landsat 7 satellite in the cleanroom prior
to launch [4]
The system was supposed to insure conti-
nuity of Thematic Type data into the next
century, but on May 31, 2003 the Scan Line
Corrector (SLC) in the sensor failed. The
purpose of the SLC was to compensate for-
ward motion (along-track) of the spacecraft
so that the resulting scans are aligned par-
allel to each other. This was permitted by
a pair of small mirrors that rotate about an
axis in tandem with the motion of the main
scan mirror. Without an operation of SLC the
scanner traces a zig-zac pattern resulting in
duplicated areas in the image. Nevertheless
the satellite is still capable of acquiring use-
ful image data with the SLC turned off since
the distortions are most pronounced along
the edge and gradually diminished towards
the center of the scene. An estimated 22% of
any given scene is lost because of the SLC fail-
ure [13].
Landsat 7 has catalogued the world’s land
mass into 57.784 scenes, each 183 km wide by
170 km long. It produces approximately 3.8
gigabytes of data for each scene [5].
Table 1.2: Landsat 7 details
Launch date April 15, 1999
Sensor ETM+
Altitude 705 km
Inclination 98.2 ◦
Orbit polar, sun-synchronous
Equational Crossing Time 10 AM
Period of Revolution 99 minutes→≈14.5 orbits/day
Repeat Coverage 16 days
Dimensions 4.04 m long, 2.74 m diameter
6 Chapter 1 Introduction
1.4.1 Primary Instrument: ETM+
The primary instrument on Landsat 7 is the Enhanced Thematic Mapper Plus (ETM+). In
comparison with the already successful Thematic Mapper (TM) instruments on Landsat 4 and
5 the ETM+ includes new features that make it a more versatile and efficient instrument for
global change studies, land cover monitoring and assessment and large area mapping.
Primarily these new features are
• a panchromatic band with 15 m spatial resolution
• an on-board, full aperture, 5% absolute radiometric calibration
• a thermal infrared channel with 60 m spatial resolution
The ETM+ uses a fixed "whisk-broom", eight-band, multispectral scanner to procure 532 high-
resolution images of the Earth’s surface per day. It detects spectrally-filtered radiation in visi-
ble and near-infrared (VNIR), short-wavelength infrared (SWIR) and long-wavelength infrared
(LWIR, "thermal infrared). Table 1.4 shows the resolution and wavelength range of every single
spectral band in detail. The high spatial resolution of 15 m without losing any information can
be achieved by panchromatic image sharpening with the help of the eighth band [5], [6].
Table 1.3: ETM+ Technical Specifications
Sensor type opto-mechanical
Spatial Resolution 30 m (60 m - thermal, 15 m - pan)
Spectral Range 0.45–12.5 µm
Number of bands 8
Temporal Resolution 16 days
Image size 183 km x 170 km
Table 1.4: ETM+ Bands
Band Number µm Resolution Light spectrum
1 0.45–0.515 30 m Blue
2 0.525–0.605 30 m Green
3 0.63–0.69 30 m Red
4 0.75–0.90 30 m Near Infrared
5 1.55–1.75 30 m Short-wave Infrared
6 10.4–12.5 60 m Long-wave Infrared
7 2.09–2.35 30 m Short-wave Infrared
8 0.52–0.90 15 m Panchromatic
1.4 Landsat 7 7
1.4.2 Landsat 8
On February 11, 2013 Landsat 8 joined the earth observation satellites. It utilizes a two-sensor
payload, the Operational Land Imager (OLI) and the Thermal InfraRed Sensor (TIRS), which
will collect image data on a total of 11 spectral bands reaching from 0.43 to 12.50 µm. The ad-
ditional three bands are a deep blue band for coastal and aerosol studies, a shortwave infrared
band for cirrus detection and a quality assessment band [12]. A comparison of the spectral
bands along with the atmospheric transmission is given in Figure 1.6.
Figure 1.6: Bandpass wavelengths for Landsat 8 OLI and TIRS sensor, compared to Landsat 7 ETM+ sensor [12]
8 Chapter 1 Introduction
1.5 Band sensitivities
Table 1.5 describes how the single bands of Landsat 7 interact to particular materials and how
they could be combined to create a first impression of the area intended for closer inspection.
Table 1.5: Band sensitivities [7], [8]
Band Sensitivity / Characteristics Application
1 Penetration of water bodies Detection of cultural features
Soil and vegetation differences Analysis of land use
2 Reflectance peak of healthy vegetation Separate vegetation from soil
Sensitive to water turbidity
3 Strong absorption of chlorophyll Discrimination of vegetation
Soil and urban areas highlighted
4 High reflectance of chlorophyll Distinguish vegetation varieties
Strong absorption of water Soil-crop and land-water contrast
5 Sensitive to moisture content Crop draught and plant healthiness studies
6 Sensitive to the Earth’s radiation Measurements of surface temperature
7 Very high water absorption Separation between land and water
Very high soil reflectance Hydrothermal alteration in rocks
1.5.1 Display of remote sensing data
Satellite images are not photographs but pictorial representations of measured data [10]. Com-
bination of certain bands represented in red, green and blue (RGB) now lead to images with
more information content compared to solely panchromatic images. In each band, grayscales
are assigned respectively to the intensity of the electromagnetic radiation and saved digitally
in pixels. Through assigning the three fundamental colours (red, green, blue) to three differ-
ent bands satellite image composites are produced. Figure 1.7 demonstrates how various band
combinations are shown by the Landsat satellite [3].
Figure 1.7a shows a "True colour" image. This band combination is used to represent an image
in natural colour and therefore best approaches the appearance of the human visual system.
Also this solution provides the highest penetration of clear water.
Combination of bands 4,3 and 2 (Figure 1.7b) will result in a "False colour" image. Band 4 shows
the reflectance peak of chlorophyll, therefore vegetation appears in tones of red. Clear water
appears in dark blue to black, thus land-water detection is way easier in this combination.
Another example is the combination of the bands 4,5 and 3 in Figure 1.7c. Band 5 makes use
of the water absorption capabilities of water, thus enabling detection of thin water layers. Also
variations in soil and rocks can be detected. When a crop has a relative lower moisture content,
the reflection from band 5 will be relatively higher, meaning more contribution of green and
thus resulting in a more orange colour.
In Figure 1.7d healthy vegetation appears in bright green, therefore this combination can be
referred as "natural-like". Band 7 highlights moisture content and is increasingly sensitive to
1.5 Band sensitivities 9
(a) "True colour"
321: combination of red(3) - green(2) - blue(1)
(b) "False colour"
432: combination of VNIR(4) - red(3) - green(2)
(c)
453: combination of VNIR(4) - SWIR(5) - red(3)
(d)
742: combination of SWIR(7) - VNIR(4) -
green(2)
Figure 1.7: Different combined bands. Depicted is an oceanic region in Italy with urban and fallow as well as
vegetated and forestal areas.
10 Chapter 1 Introduction
the emissive radiation which enables access to detect heat sources. This combination is often
used in desert regions or for geological and wetland studies.
1.6 Vegetation Indices
Other than assigning the three fundamental colours to different bands, mathematical opera-
tions applied to the raw data can extract more information. Addition, subtraction and division
of the brightness of two or more bands are the most common. For example the Normalized Dif-
ference Vegetation Index (NDVI) is used to identify the health status of plants, depict phenologi-
cal changes and other remote sensing measurements [8]. The NDVI is calculated as follows
NDVI =
(∅nir −∅red)
(∅nir +∅red)
(1.1)
in which ∅nir is the near infrared brightness of a pixel and ∅red is its visible red brightness.
Every Pixel is now represented in values between −1 (no vegetation) and +1 (abundant veg-
etation). For the different classifications in chapter 2 and 3 this index will be used. Two other
indices sometimes used are the Normalized Difference Water Index (NDWI), which gives an indi-
cation of soil moisture
NDWI =
(∅green −∅nir)
(∅green +∅nir)
(1.2)
Equivalent ∅green is now the green wavelength brightness of a pixel [8]. In this index, water
features values between 0 and +1. As a third example, the Land Surface Water Index is used to
monitor liquid water content in vegetation and soil [8]
LSWI =
(∅nir −∅swir)
(∅nir +∅swir)
(1.3)
1.7 Case Study and Data 11
1.7 Case Study and Data
The study area for this thesis is the Po river in Italy. As visualized in Figure 1.8 it flows 682 km
eastward across northern Italy from a spring in the Cottian Alps through a delta projecting into
the Adriatic Sea. Very large discharge of the Po leads to heavy flooding in the whole area and
consequently into many economic problems. But still the valley is one of the main areas for
industry and agriculture thus it appears that reliable frequent observation of the water extent
is essential.
Figure 1.8: Location of the Po river; Scenery in a negative colour image (band 432)
A total of four Landsat 7 satellite images are used in this thesis. One for each season of the year
2001. This way different growing seasons and weather influences like clouds or light shade
effects will be considered in the single classifications.
1.8 Outline of the Thesis
The physical backgrounds that are required for satellite imagery have been introduced, the
interpretation of raw image data will be the task of Chapter 2. There is no single, correct method
for analysing digital images, therefore it is important to know the pros and cons of all available
options. Various techniques to classify an image and produce maps of land cover, land type
and land use will be described in detail in this section. Chapter 3 will illustrate the results in
comparison with an in-field surveillance. A concluding discussion will be given in Chapter
4.

13
Chapter 2
Classification Methods
Digital image classification is a mapping from the spectral measurements acquired by a remote
sensing instrument to a set of labels that represent land cover features. Land cover could be
agricultural, urban, forested or water-covered types of features. In order to provide a unique
label, or theme, for each pixel in the image, different computer processing algorithms are used.
This process is called thematic mapping and the produced map therefore is referred to as a the-
matic map. Knowing the size of a pixel in the corresponding ground metres, accurate estimates
of the area can be calculated [8].
The classification techniques in remote sensing can be categorized in two groups:
• Supervised classification
• Unsupervised classification
2.1 Supervised classification
The major task of the supervised methods is to segment the spectral domain into regions of
interest to a particular application. These techniques are mostly used for quantitative analysis
and are divided into soft classification methods and hard classification methods. The difference
in these two methods lays in the way they identify and describe the regions in spectral space.
Some methods seek a segmentation based on probability distributions while others use simple
geometric partition. Hard classification generates firm boundaries between classes whereas soft
classification sometimes leads to spatially overlapped regions [8].
Irrespective of the method chosen, it requires training pixels whose class labels are known and
which can be used to form training data for the desired classifier. A training set can be estab-
lished using image products formed from the data such as a RGB-image from Section 1.5.1 or
the NDVI-resolution from Section 1.6. In spectral domain the training pixels will enclose a
common region, the so called training field.
The training data can now be used to assess the parameters (signature) of a particular classifier
and later on label every pixel to one of the predefined classes.
In general the supervised classification follows these essential steps [10]:
• Definition of the different land cover classes
• Classification of suitable training areas
• Employing of the actual classification with a particular algorithm
14 Chapter 2 Classification Methods
• Producing a thematic map
• Assessing the accuracy of the final product using a labelled testing data set
2.1.1 Maximum Likelihood Classification
The maximum likelihood classification is the most common algorithm used in remote sensing.
It uses a probability density function based on Bayes’ classification rule, meaning that each pixel
is assigned to the class that has the highest probability (maximum likelihood). Each pixel is
labelled according to the decision rule
x ∈ ωi if p(x|ωi)p(ωi) > p(x|ωj)p(ωj) for all j 6= i (2.1)
The classes are represented by ωi, i = 1...M where M is the number of classes. x is a column
vector with the brightness values for the pixel in each measurement band, therefore p(x|ωi)
describes the chance of finding a pixel at position x in spectral space from each class ωi. p(ωi)
is referred as prior probability and describes the probability that pixels from class ωi appear
anywhere in the image.
p(x) can be removed, since it is not class dependant. This leads to the discriminant function
gi(x) = ln p(x|ωi)p(ωi) = ln p(x|ωi) + ln p(ωi) (2.2)
Substitution of (2.2) into (2.1) gives a new decision rule
x ∈ ωi if gi(x) > gj(x) for all j 6= i (2.3)
Figure 2.1 shows that the classes of pixels in spectral space are normally distributed (Gaussian
distribution) which leads to the Gaussian maximum likelihood classifier1
gi(x) = − ln |Ci| − (x−mi)TC−1i (x−mi) (2.4)
where mi and Ci are the mean vector and covariance matrix of the data in class ωi.
If the highest probability of a pixel falls below a specified value1 a threshold can be set and
the pixel won’t be classified in any of the available classes. This circumstance will guarantee a
certain degree of accuracy and is illustrated in Figure 2.2.
1For more information and the mathematical derivation see John A. Richards, Remote Sensing Digital Image Analysis,
5th ed., Springer Science and Business Media, 2012, p. 250 ff.
2.1 Supervised classification 15
Figure 2.1: Example of a histogram: water class in all six available bands1
2.1.2 Minimum Distance Classification
The accuracy of the maximum likelihood classification (2.1.1) method depends on the num-
ber of training pixels used for each class. If this sufficient number is not possible, inaccurate
valuation of the covariance matrix will lead to poor classifications. The minimum distance clas-
sification method depends only on the mean vector for each spectral class and does not make
use of the covariance information. The training data is used to calculate class means and each
pixel is then classified in the class of the nearest mean.
x ∈ ωi if d(x, mi)2 < d(x, mj)2 for all j 6= i (2.5)
in which mi, i = 1...M are the means of the M classes, which are obtained from training data.
Similar to (2.3) the discriminant function of the minimum distance classifier is
x ∈ ωi if gi(x) > gj(x) for all j 6= i
with
gi(x) = 2mi · x−mi ·mi (2.6)
Table 2.1 shows an example for the class means in the image recorded in winter. Additionally
the range in which the pixels are assigned to the classes is shown. The classification is employed
on basis of the NDWI image, therefore the pixels have values between -1 (vegetation) and +1
(water).
1Band 6 is actually Landsat Band 7
16 Chapter 2 Classification Methods
Figure 2.2: Illustration of the use of thresholds to avoid questionable classification decisions [8]
Table 2.1: Class means, ranges and standard deviation in the classification of the fourth image
Class Minimum Maximum Mean STD
Vegetation -0.5861 -0.2078 -0.2722 0.0607
Dry soil -0.2074 -0.0894 -0.1283 0.0301
Urban -0.0894 -0.0425 -0.0668 0.0125
Wet soil -0.0424 0.2221 -0.0003 0.0539
Water 0.2224 0.7287 0.4393 0.0974
Analogical to the thresholds in the maximum likelihood classifier, such can also be applied
to the minimum distance classifier. They are usually specified in terms of a number of stan-
dard deviations from the class means. This classification technique is advantageous when the
number of training pixels is limited or if linear separation of the classes is suspected [8].
2.1 Supervised classification 17
2.1.3 Parallelepiped Classification
The parallelepiped classifier uses a simple decision rule, that is to find the upper and lower
brightness values in each spectral dimension. Making use of the histograms of the individual
spectral components in the available training data would be the most obvious solution to find
these boundaries, as shown in Figure 2.3. For each class a multidimensional box or parallelepiped
is formed (Figure 2.4). If a unknown pixel lies in between this box, it is assigned to that class.
Figure 2.3: Setting the parallelepiped boundaries by inspecting class histograms in each band
Figure 2.4: A set of two dimensional parallelepipeds
Figure 2.5: Classification of correlated data showing
regions of inseparability
For correlated data some parallelepipeds can overlap, as illustrated in Figure 2.5. The classi-
fication software ENVI assigns pixels within this inseparable region to the last class matched.
Another disadvantage of this classifier are gaps between the parallelepipeds. These areas will
be designated as unclassified. Alike with the minimum distance classifier, there is no provi-
sion for prior probability of class membership with the parallelepiped rule, leaving the sim-
ple and fast computing algorithm as the only advantage in the parallelepiped classification
method [2], [8].
18 Chapter 2 Classification Methods
2.1.4 Mahalanobis Distance Classification
The initial point of the mahalanobis distance classification is the discriminant function of the
maximum likelihood classifier (2.4). If the sign is reversed, the function can be considered as a
distance squared measure because the quadric entry has those dimensions and the other term
is a constant [8]. We can define
d(x, mi)2 = ln |Ci|+ (x−mi)TC−1i (x−mi) (2.7)
and look for the smallest d(x, mi) to classify unknown pixels, similar to the minimum distance
algorithm. Considered now that all class covariances are equal and given by Ci = C, which
could be for example a class average, (2.7) can be reduced to
d(x, mi)2 = (x−mi)TC−1i (x−mi) (2.8)
This squared distance is called the mahalanobis distance. The additional simplification C = σ2I
would result in the minimum Euclidean distance classifier of Section 2.1.2. Compared to the
maximum likelihood classification method, this technique is faster and yet retains a degree of
direction sensitivity via the covariance matrix C [8].
2.2 Unsupervised classification
Unsupervised classification is a method by which each pixel is assigned to a class without any
prior knowledge of the existence or names of those classes. This process of grouping is called
clustering. The user identifies the number of clusters to generate and which bands to use but
takes no part in an algorithm’s learning process directly. The produced cluster map now has to
be compared to available reference data to identify the particular classes or clusters.
This method of classification is often useful for determinations prior to detailed analysis by the
supervised methods from Section 2.1 [8].
2.2.1 Clustering Criteria
The task is to find groups of pixels that are somehow similar to each other. This similarity is
measured in the spectral attributes recorded by the sensor used to acquire the data. The most
common form of measurement in clustering procedures is usually a simple distance measure.
Two examples would be the Euclidean distance (2.9) and the city block or Manhattan (2.10) dis-
tance. The Euclidean distance of two measurement vectors x1 and x2 whose similarity is to be
checked is calculated as
d(x1, x2) =̂ ‖x1 − x2‖
= {(x1 − x2) · (x1 − x2)}1/2
=
{
(x1 − x2)T(x1 − x2)
}1/2
=
{
N
∑
n=1
(x1n − x2n)2
}1/2 (2.9)
2.2 Unsupervised classification 19
with N =̂ number of spectral components. The city block (L1) distance is given by
dL1(x1, x2) =
N
∑
n=1
|x1n − xx2n| (2.10)
The city block is way faster to compute at the cost of quality in spectral similarity within a given
distance [8].
2.2.2 K-Means Clustering
The k-means clustering method is one of the most common approaches used in image analysis
applications. It aims to partition all available measurement vectors into a preassigned number
of clusters. After assigning an initial set of cluster centres the algorithm alternates between two
steps:
- Reassign each pixel vector to the cluster with the closest mean using an appropriate dis-
tance metric between the pixel and cluster means. Euclidean distance is commonly used.
- Calculate new cluster means from each group.
These steps repeat until there is no significant change in the value of each pixel vector [8].
2.2.3 Isodata Clustering
The Isodata clustering uses the same algorithms as the k-means but with certain refinements
such as splitting, merging and deleting clusters.
• Merging clusters: Sometimes two or more clusters are so close together that they repre-
sent unnecessary information content, in which they should be merged.
• Deleting clusters: Sometimes statistical distributions of certain pixels are important and
must be available to generate reliable mean and covariance estimates, as for example in
the maximum likelihood classification (see Section 2.1.1).
• Splitting clusters: Clusters with improper shapes in the spectral space can be split in
two by pre-specifying a standard deviation in each spectral band beyond which a cluster
should be halved [8].

21
Chapter 3
Results
For each of the four images, one for each season of the year, a spatial subset to show the training
areas and the results of the six classifiers was defined. Along with that more information about
the arrangement of the training areas and the statistic results for each classification will be
presented in different tables and figures.
3.1 First Image: Spring
3.1.1 Training Data
As an example for training areas, Figure 3.1 shows a segment of the first Image to be classified
created with the software ENVI. Five land cover classes (Regions of interest; ROI) were defined:
Water (blue), vegetation (green), dry soil (yellow), wet soil (orange) and urban area (cyan).
(a) True colour (321) (b) NDVI image with training areas
Figure 3.1: Image segment to be classified; consisting of different land cover classes
The amount of training pixels for each class and their related colour is shown in Table 3.1. The
training pixels for the water class were created by using a band threshold with values from −1
22 Chapter 3 Results
Table 3.1: Regions of interest with their corresponding amount of pixels and colours
ROI Name Colour Pixels Polygons1 Points
Water Blue 232,253 0/0 232,253
Vegetation Green 8,167 28/8,167 0
Dry soil Yellow 20,467 24/11,104 9,363
Wet soil Orange 29,169 26/12,266 16,903
Urban Cyan 5,816 9/5,816 0
to −0.1 from a NDVI image as a region of interest. Every pixel in between these values is now
used as a training pixel. The other training areas were generated with polygons, rectangles,
ellipses or expanded single points in the RGB image.
3.1.2 Classified Images
Figure 3.2 shows a spatial subset of the full image as a negative colour image. This subset
includes the five land cover classes as well as critical sectors in which the classification is am-
biguous. The results of the six classifiers are depicted in Figure 3.3.
Figure 3.2: Subset of the season 1 image as a negative colour image (432). The image shows vegetated area as well
as bare soil. The urban area in the center was the critical sector to be classified and shows different results in
each of the classifiers.
1Amount of polygons and how many training pixels they contain in total; remaining pixels are created from
expanding single pixels or from band thresholds
3.1 First Image: Spring 23
(a) Maximum Likelihood Classifier (b) Minimum Distance Classifier
(c) Parallelepiped Classifier (d) Mahalanobis Distance Classifier
(e) K-Means Clustering (f) Isodata Clustering
Figure 3.3: First image: Spring classification results
24 Chapter 3 Results
Most images were generated without any thresholds or limitations. Only in the minimum dis-
tance classifier (Figure 3.3b) a maximum standard deviation from the mean limitation of the
factor 2 in the urban class has been applied. Otherwise too many pixels would have been clas-
sified to the this class. For proper distribution of the pixels in the parallelepiped classification
(Figure 3.3c) a couple of thresholds have been used. A factor of 3 in the water class, 2 in vege-
tation, 1.3 in bare and 2.5 in wet soil and the urban class was stretched by the factor 7 from the
mean value. This way only 0.27 % of the pixels were left as unclassified.
The properties in the K-means algorithm (Figure 3.3e) was limited to 10 iterations or less than
3 % of the pixels change their class. Also no maximum standard deviation or distance error
from the mean in each class were applied. Same properties apply to the Isodata clustering
(Figure 3.3f) but additionally the algorithm could chose from five up to seven classes. It still
classified the image into five classes. On the first sight we see no difference in the unsupervised
classified images but they can primarily be separated by the number of pixels in the wet soil
and the urban area class.
3.1.3 Discussion
A precise notation of the percentage of pixels in each class for the single classified images is
given in Table 3.2.
Table 3.2: Spring image statistics results
Classifier Unclassified Water Vegetation Dry soil Wet soil Urban
Maximum likelihood 0 % 6.12 % 23.22 % 33.51 % 13.84 % 23.32 %
Minimum distance 0 % 6.47 % 18.51 % 29.92 % 20.84 % 24.26 %
Parallelepiped 0.27 % 5.77 % 19.28 % 25.44 % 28.76 % 20.48 %
Mahalanobis distance 0 % 6.78 % 16.55 % 29.17 % 29.54 % 17.96 %
K-Means clustering 0 % 6.61 % 20.74 % 34.95 % 25.43 % 12.27 %
Isodata clustering 0 % 6.66 % 19.51 % 34.67 % 28.34 % 10.81 %
As we can see, the amount of water varies between 5.77 % and 6.78 % or in area units: be-
tween 207.9405 km2 and 244.2861 km2 which makes a difference of 36.3456 km2. The maximum
likelihood (Figure 3.3a) and the mahalanobis distance (Figure 3.3d) classifier outperform as far
as the urban area is concerned while the unsupervised classifiers couldn’t distinguish between
soil and urban area but they still remained very accurate in the river boundaries and the oceanic
region.
3.2 Second Image: Summer
3.2.1 Training Data
The second image takes place in early summer of the year 2001. The whole area suffers from
high cloud coverage which can be seen in Figure 3.4a. It is almost impossible to get accurate
classification results in the regions right beneath the clouds and in the cloud-shadowed areas.
3.2 Second Image: Summer 25
(a) True colour (321) (b) Natural like image (752); training areas
Figure 3.4: Spatial subset of the image recorded in summer; The whole image suffers from high cloud coverage and
shaded areas
With help of the false colour image (432) and the NDVI two additional classes to examine these
areas were defined. The other classes will be retained but a discrimination in different soil
features was not necessary anymore, which gives us a total of seven land cover classes. Like in
the season 1 image the training pixels for the water class were generated with help of the NDVI
image but this time in form of polygons. The best examination of the vegetated areas could be
provided with a 742 band combination and the same goes for soil features. The total amount
of pixels for each region of interest is depicted in Table 3.3 and a small subset is pictured in
Figure 3.4b.
Table 3.3: Regions of interest with their corresponding amount of pixels and colours
ROI Name Colour Pixels Polygons1
Water Blue 9,964 25/9,964
Vegetation Green 9,111 33/9,111
Soil Yellow 5,301 19/5,301
Urban Cyan 2,888 12/2,888
Cloud White 2,620 12/2,620
Shadow Blue 1,107 12/1,107
3.2.2 Classified Images
The performance of the six classifiers and how they classified the cloud-covered areas are
shown in Figure 3.5 in the same manner as in the previous section. Again a small subset which
includes all critical sectors of the whole image will represent the results for this image.
1Amount of polygons and how many training pixels they contain in total; remaining pixels are created from
expanding single pixels or from band thresholds
26 Chapter 3 Results
(a) Maximum Likelihood Classifier (b) Minimum Distance Classifier
(c) Parallelepiped Classifier (d) Mahalanobis Distance Classifier
(e) K-Means Clustering (f) Isodata Clustering
Figure 3.5: Second image: Summer classification results
3.2 Second Image: Summer 27
Again as little thresholds as possible were applied to classify as much pixels as possible. Some
distribution corrections set in the parallelepiped (Figure 3.5c) and the mahalanobis distance
classification (Figure 3.5d). Wihtout thresholds in the first case, almost the whole image would
have been classified to the vegetation class and in the mahalanobis distance classification huge
errors in the water and shadow classes made a usage of thresholds inevitable.
The properties for the K-means classification (Figure 3.5e) were the same as in the image
recorded in spring, but this time with a limitation to six classes. In the Isodata clustering (Fig-
ure 3.5f) a minimum of six and a maximum of eight classes were chosen, otherwise the algo-
rithm would have set the number of classes down to five and merged the water and shadow
class together. This way the number of classes equals the same in all pictures.
3.2.3 Discussion
Table 3.4 shows the distribution of pixels in the single classes similar to Table 3.2 from spring.
If we consider that cloud-covered and shadows area is lost data and add these two classes to
the unclassified pixels we will get a certain percentage of lost data for each of the classifiers.
Table 3.4: Summer image statistics results
Classifier Unclass. Water Vegetation Soil Urban Cloud Shadow
Max. likelihood 0 % 5.61 % 45.04 % 6.46 % 35.09 % 3.41 % 4.40 %
Minimum distance 0 % 5.71 % 34.93 % 7.80 % 35.01 % 4.18 % 12.37 %
Parallelepiped 5.21 % 6.54 % 41.16 % 14.77 % 25.36 % 3.25 % 3.70 %
Mahalanobis dist. 0 % 8.18 % 41.23 % 15.25 % 28.31 % 2.86 % 4.17 %
K-Means clustering 0 % 8.64 % 16.67 % 16.00 % 36.14 % 5.46 % 17.11 %
Isodata clustering 0 % 7.51 % 19.85 % 16.65 % 36.36 % 5.46 % 14.18 %
This way the mahalanobis classifier (Figure 3.5d) would show the best classification as far as the
data loss in the whole scenery is concerned. The most classified pixels, especially in the cloud
class, look very accurate, however this classifier has difficulties in separating the shadowed
areas from the water areas. Also a misclassification in the oceanic region appeared.
The maximum likelihood (Figure 3.5a) along with the minimum distance classifier (Figure 3.5b)
performed pretty well in the shadowed and cloud-covered areas and did not classify any sig-
nificant areas in a wrong way. Only cropped fields with high moisture content were classified
differently into the vegetation or accordingly in the shadow class. The parallelepiped classifier
(Figure 3.5c) assigned these pixels into the water class and since this class is the one of highest
interest, a misclassification is very disadvantageous. Additionally this classifier misinterpreted
some cloud shadows as an area of water content. This circumstance of misclassification in this
image is shown in Figure 3.6.
In the classification images provided by the unsupervised methods similar effects appeared.
The K-means classifier (Figure3.5e) shows the same errors as the parallelepiped classifier but
in an even bigger impact which explains the 8.64 % water content in the whole scenery. The
Isodata classifier (Figure 3.5f) eliminated these errors in the shadowed areas but remains them
in areas of high water content. Still this classification can be referred as the most accurate in
this season.
28 Chapter 3 Results
(a) Maximum likelihood (b) Minimum distance (c) Parallelepiped
Figure 3.6: Misclassified area of the summer image
3.3 Third Image: Autumn
3.3.1 Training data
The image of the third season takes place in late summer and shows some brightness issues in
the southern part of the image as seen in Figure 3.7a. The same land cover classes as in season
1 were defined: Water, vegetation dry and wet soil and an urban class.
(a) True colour (321); brightness issues (b) NDVI image with training areas
Figure 3.7: Spatial subset of the autumn image; An area in the south suffers from brightness issues
As seen in Figure 3.7b the brightness issues are completely diminished in the NDVI image.
Therefore this index would operate perfectly as a base for our training data. The colours and
the approach to the training areas remained the same as in the previous sections.
3.3 Third Image: Autumn 29
3.3.2 Classified Images
The classification results for the third image are presented in Figure 3.8.
No thresholds were applied to the maximum likelihood classification (Figure 3.8a) and only
small limits in the wet soil and urban class were set in the minimum distance image (Fig-
ure 3.8b). Same goes for the mahalanobis distance classifier (Figure 3.8d). The water class in
the parallelepiped classification (Figure 3.8c) was highlighted while the vegetation and dry soil
class has been narrowed down.
Adding additional classes in the unsupervised classifications was of no avail. The algorithm
just classified soils and rocks into more and more classes of different moisture content. Thus
the properties in the K-means (Figure 3.8e) and the Isodata classifier (Figure 3.8f) remained the
same as in the first season.
3.3.3 Discussion
To illustrate how the classifiers performed in an image affected by brightness issues Table 3.5
gives us a short overview of the assignment of pixels in each class in the same way as in the
previous sections.
Table 3.5: Autumn image statistics results
Classifier Unclassified Water Vegetation Dry soil Wet soil Urban
Maximum likelihood 0 % 5.21 % 13.02 % 21.87 % 22.24 % 37.66 %
Minimum distance 0 % 7.86 % 13.20 % 23.87 % 19.09 % 35.98 %
Parallelepiped 2.06 % 6.38 % 14.66 % 24.47 % 29.04 % 23.40 %
Mahalanobis distance 0 % 10.96 % 14.14 % 29.35 % 11.20 % 34.34 %
K-Means clustering 0 % 6.27 % 15.38 % 20.08 % 24.37 % 33.93 %
Isodata clustering 0 % 6.31 % 14.08 % 19.18 % 26.23 % 34.20 %
If we take a closer look at Figure 3.8 the supervised classifiers assigned most of the brightened
area to the urban class. An exception is the minimum distance classifier. The unsupervised
classifiers led to better results in this case.
Figure 3.9: Errors in the mahalanobis distance clas-
sification
The maximum likelihood and the mahalanobis
distance classifier were not able to generate firm
water boundaries and show a significant amount
of misclassified pixels in the oceanic region and
in the separation between wet soil and water (see
Figure 3.9. These errors could not be corrected by
applying thresholds or excluding single spectral
bands from the classification.
Even though the minimum distance and the par-
allelepiped classifier had some issues in the detec-
tion of thin water layers their water boundaries
around the river seem to be pretty accurate and
low in misclassified pixels. The images created by
30 Chapter 3 Results
(a) Maximum Likelihood Classifier (b) Minimum Distance Classifier
(c) Parallelepiped Classifier (d) Mahalanobis Distance Classifier
(e) K-Means Clustering (f) Isodata Clustering
Figure 3.8: Third image: Autumn classification results
3.4 Fourth Image: Winter 31
the unsupervised methods show a high similarity to the minimum distance image. Only a lit-
tle area affected by the brightness issues was classified in a wrong way in the unsupervised
classification methods.
3.4 Fourth Image: Winter
3.4.1 Training Data
The fourth and last image was recorded at the end of the year 2001. As seen in Figure 3.10a
the true colour RGB implementation won’t be of any practical use. The brightness values in
the swath wary widely through the whole spectrum by what the image almost appears as a
black and white image. The NDVI image does not help us either but the Normalized Difference
Water Index (Figure 3.10b) in collaboration with some false colour images will lead to reasonable
training areas (Figure 3.11). Also the histogram of the NDWI image has been balanced to a
minor extent.
(a) True colour image (321) of the whole swath
(b) Normalized Difference Water Index (NDWI)
Figure 3.10: True colour (321) and NDWI image of the whole swath
3.4.2 Classified Images
The maximum likelihood classification image in Figure 3.12a was computed based on the
whole spectral subset, which means that all possible spectral bands were used (band 1 to 5
and band 7). Also no thresholds were applied. In the parallelepiped (Figure 3.12c) and the ma-
halanobis distance classifier (Figure 3.12d) only band 5 and 7 were used to guarantee as much
32 Chapter 3 Results
Figure 3.11: NDWI subset with training areas
atmospheric penetration as possible. Otherwise those classifiers would have totally misclassi-
fied the dimmed and the brightened area in the bottom right of the image. Limitations to all
classes in the parallelepiped classification were applied but apart from that the water class was
stretched. This way a various amount was pixels were left as unclassified but at least the water
area in the dimmed area shows reliable results.
The minimum distance classifier (Figure 3.12b) along with the unsupervised methods (Fig-
ures 3.12e and 3.12f) were classified based on the NDWI image. Irrespective of the settings the
usual image did not lead to any accurate results.
3.4.3 Discussion
Table 3.6 will show us if there will be any significant difference in classification images com-
puted from the NDWI or the usual image on different spectral subsets.
Table 3.6: Winter image statistics results
Classifier Unclassified Water Vegetation Dry soil Wet soil Urban
Maximum likelihood 0 % 5.42 % 9.87 % 51.70 % 18.96 % 14.04 %
Minimum distance 0 % 4.71 % 4.80 % 35.07 % 13.90 % 41.52 %
Parallelepiped 5.86 % 5.74 % 14.32 % 27.41 % 35.56 % 11.12 %
Mahalanobis distance 0 % 7.58 % 9.10 % 47.07 % 17.68 % 18.57 %
K-Means clustering 0 % 4.67 % 6.42 % 27.34 % 8.03 % 53.54 %
Isodata clustering 0 % 4.60 % 4.71 % 24.38 % 6.56 % 59.71 %
3.4 Fourth Image: Winter 33
(a) Maximum Likelihood Classifier (b) Minimum Distance Classifier
(c) Parallelepiped Classifier (d) Mahalanobis Distance Classifier
(e) K-Means Clustering (f) Isodata Clustering
Figure 3.12: Fourth image: Winter classification results
34 Chapter 3 Results
Figure 3.13: Error in the coastal region; represented
in the parallelepiped classification
The maximum likelihood classifier (Figure 3.12a)
showed a great performance in the urban and
vegetation classes but assigned the dimmed area
almost completely into the wet soil class. The
mahalanobis distance classifier (Figure 3.12d) on
the other hand assigned these pixels to the ur-
ban class and overall looks very pixelated without
any fixed structure. In the image created by the
parallelepiped classifier (Figure 3.12c) many pix-
els were left as unclassified especially in the vege-
tated areas in the dimmed part of the image.
The minimum distance classification along with the unsupervised classifiers did not really clas-
sified an urban class. It is more a leftover class for farmland and urban features. It is evident
that this similarity comes from the NDWI image as a basis for the classification.
An error in the coastal region appeared in almost every classified image. A subset from this area
is represented in Figure 3.13. The parallelepiped classifier left the pixels around the coast as
unclassified while the maximum likelihood and the mahalanobis distance classifiers assigned
them to the urban and the vegetation class. Since this area should contain sand or turbid water,
the water or the wet soil class would be the most obvious classification. The minimum distance
and the unsupervised classifiers applied to this solution.
35
Chapter 4
Validation
To give precise differentials in the distance between the river widths calculated from the classi-
fications and the results from in situ measurements two spots alongside the river were chosen
as seen in Figure 4.1. At these two spots, located in Sermide and Pontelagoscuro, a direct com-
parison is performed.
(a) Whole swath; locations of Sermide and Pontelagoscuro
(b) Sermide (c) Pontelagoscuro
Figure 4.1: Areas for a comparison with an in-field surveillance
In order to measure the exact river widths in the classified images the images are now con-
verted to a binary water-mask. As an example the water-mask from the spring image Isodata
classification is shown in Figure 4.2a. Since there is no significant difference in the single masks,
the display of all masked image is not necessary. Combination of the mask and other images
with simple mathematical operators can improve the visual results. In Figure 4.2b a simple
36 Chapter 4 Validation
multiplication of the values from the NDVI image and the binary water mask has been ap-
plied.
(a) Isodata water mask (b) Water mask multiplied with the NDVI image;
River profile to be measured
Figure 4.2: Water mask example: Gauge spot Sermide
With help of the measurement tool in ENVI precise measurements at desired locations (Red line
in Figure 4.2b) can now be applied. The results from each classification can now be compared
to an in-field surveillance to assess the performance of the different classifiers.
4.1 In Situ Results
Between 1995 and 2011 profile measurements alongside the river have been taken regularly.
The location of these profiles are shown in Figure 4.3 and as mentioned earlier two of these
profiles are chosen for closer investigation. They are depicted in Figure 4.4. The river widths at
these two spots for all four seasons of the year 2001 are depicted in Table 4.1. They will serve
as reference values for the following performance analysis. We can see that the river widths
decline constantly throughout the year and if we look at past and future years from the date
this measurements were taken this circle will repeat every year. The rivers profile in Sermide
shows an abrupt slope together a levee (see Figure 4.4a), probably due to heavy flooding in this
area, by what the river width does not differ very much throughout the year.
4.1 In Situ Results 37
Figure 4.3: Profile measurements alongside the Po river; Gauge spots Sermide and Pontelagoscuro
(a) Sermide (b) Pontelagoscuro
Figure 4.4: River profiles; water levels through the year
Table 4.1: Profile widths 2001
Season Sermide Pontelagoscuro
1 489.80 m 263.18 m
2 488.75 m 261.66 m
3 485.98 m 257.92 m
4 483.01 m 253.19 m
38 Chapter 4 Validation
4.2 Classification Performances
The single classifiers are now represented in a bar chart with a color assigned to each of them
(see Figure 4.5 and Figure 4.6). The red line shows the reference value from the in-field mea-
surement.
Gauge Spot Sermide
(a) Spring (b) Summer
(c) Autumn (d) Winter
Figure 4.5: Classification performances in Sermide
4.2 Classification Performances 39
Gauge Spot Pontelagoscuro
(a) Spring (b) Summer
(c) Autumn (d) Winter
Figure 4.6: Classification performances in Pontelagoscuro
Figure 4.7 shows the overall mean deviation from both measurement spots and all four seasons
in percent and in Figure 4.8 the deviations were summarized to give an absolute value in the
difference from the in-field measurement in metre.
In almost every case the classifiers assigned too little pixels to the river class resulting in much
narrower widths compared to the in-field measurement. The Maximum likelihood along with
the parallelepiped classifier show less performance. Not only they showed many misclassified
pixels but also the profile measurements alongside the river were the least accurate. It has to be
mentioned that the maximum likelihood classifier showed some very unlucky misclassification
in Pontelagoscuro (Figure 4.6c) which is the main reason why it was so outperformed by the
other classifiers in such a big manner. A few errors in the water class appeared but it was able
to identify disruptive elements like clouds and shadows sharply.
40 Chapter 4 Validation
Figure 4.7: Average percentaged deviation
The parallelepiped classifier left many pixels as unclassified but still created very firm bound-
aries in between the classes and it was the only classifier which was able to clearly identify
bridges and other small objects alongside the river. The unsupervised classification techniques
performed slightly better than the parallelepiped classifier. They both performed equally well,
there was no significant difference in the subset where the profile measurements were applied.
In some other areas the images classified by Isodata clustering looked like an eroded version
of the ones created with the k-means classification meaning the river was thinned out and as
a result the boundaries appeared more accurate. When it comes to disruptive elements like
shadows or brightness issues the unsupervised techniques did not perform very good. Same
goes for the urban areas.
Close behind the accuracy of the remaining two classifiers it was advantageous to just measure
the profile in a visual way with the help of the NDVI image. Of course it is very difficult to
distinguish between water and land in a greyscale image with a spatial resolution of 30 m but
still this method led to acceptable results.
The second best performance was approached by the minimum distance classification. It per-
formed very good in cases of thin layers of water and also created very firm boundaries be-
tween the single classes. Like the unsupervised methods it had some issues in identifying
urban areas but overall the misclassified areas are kept within a limit.
Among the classification methods used in this thesis we can see clearly that the mahalanobis
distance classifier outperforms as far as the river widths are concerned. It often looks very
4.2 Classification Performances 41
Figure 4.8: Absolute deviation from all eight measurements
pixelated and also a few errors in the coastal region and in distinguishing between wet soil
and water appeared. However, it is by far the best classification method in the urban area and
it also assigned pixels beneath clouds and in shadowed areas more accurate than any other
classifier.

43
Chapter 5
Conclusion and Outlook
5.1 Conclusion
In this study the unsupervised clustering algorithms were outperformed by the supervised
classifiers. But still these methods are suitable for quick and reliable results and create very firm
boundaries in between the classes. Same goes for the comparatively simple algorithms like the
parallelepiped and the maximum likelihood classification. Also a brief visual measurement
in the NDVI or in a RGB image gives a good first overview of the local conditions. The best
result is achieved by the minimum distance and the mahalanobis distance classifier. What both
have in common is how they assign unknown pixels to a class: they both look for the smallest
distance in spectral space.
However, it has to be noticed that these results only apply to the two gauge spots used for
this study. A general statement about the mahalanobis distance classification being the best
method among the six classifiers can not be made. Depending on different environmental cir-
cumstances other classifiers may lead to better results. On individual applications one should
always test different classification methods to see which one gives the most reliable results.
5.2 Outlook
One of the most crucial points for reliable remote sensing image analysis is the spatial reso-
lution. Landsats relatively low resolution of 30 m implemented on a 300 m wide river profile
does not leave vast room for interpretation. To guarantee more reliable results measurements
on water surface areas instead of river profile widths could be applied. Area-wise comparison
between the classification results and these so called effective river widths may lead to better esti-
mates but such an arrangement would require good testing grounds alongside the observation
area.
The extent of this bachelor thesis could not cover all issues of this topic. However, despite the
low resolution, good first impressions of the advantages and disadvantages of a wide range of
classification methods were given. It is a good starting point for further studies in this area of
research.
XIII
Bibliography
[1] Elmi, O. [2015], The role of multispectral image transformations in change detection, Mas-
ter’s thesis, University of Stuttgart.
[2] EXELIS [2015], ‘classification’.
URL: http://www.exelisvis.com/docs/classification.html
[3] GDSC [2015], ‘band combinations’.
URL: http://gdsc.nlr.nl/gdsc/en/information/earth_observation/band_combinations
[4] Landsat 7 Picture [n.d.].
URL: http://www.lockheedmartin.com/content/dam/lockheed/data/space/photo/landsat-7.jpg
[5] NASA [2015a], ‘Nasa landsat science’.
URL: http://landsat.gsfc.nasa.gov/
[6] NASA [2015b], ‘The Landsat 7 summary’.
URL: http://geo.arc.nasa.gov/sge/landsat/l7.html
[7] Resolutions of Remote Sensing [n.d.].
URL: http://www.edc.uri.edu/nrs/classes/nrs409/rs/lectures/howremotesensonwork.pdf
[8] Richards, J. A. [2012], Remote Sensing Digital Image Analysis - An Introduction, 5. edn,
Springer Science and Business Media.
[9] Schowengerdt, R. A. [2007], Remote sensing: models and methods for image processing, Aca-
demic Press, Oxford.
[10] Siegmund, A., Kollar, I. and Siegmund, A. [2015], ‘Introduction to Remote Sensing’.
URL: http://www.seos-project.eu/modules/remotesensing/remotesensing-c00-p02.html
[11] Unsalan, C. and Boyer, K. L. [2011], Multispectral Satellite Image Understanding - From Land
Classification to Building and Road Detection, Springer Science and Business Media, Berlin
Heidelberg.
[12] U.S. Geological Survey [2015a], ‘Landsat 8’.
URL: http://landsat.usgs.gov/landsat8.php
[13] U.S. Geological Survey [2015b], ‘Slc-off products: Background’.
URL: http://landsat.usgs.gov/products_slcoffbackground.php