Anda di halaman 1dari 60

CHAPTER - 1 Remote Sensing (RS)

www.geospatialcareers.net http://www.doleta.gov/business/ http://www.gis.com/careers/geospatial_career.html www.careervoyages.gov

CHAPTER - 1 Remote Sensing (RS)


Learning Objectives By the end of this chapter students should be able to:
Understand Spectral Reflectance Signature of various earth features ean about !igital "mage #rocessing Understand the $isual "nterpretation of Satellite data %now &erial #hoto and "ts "nterpretation earn about &dvanced Remote Sensing 'echnologies Understand the &dvantages and Benefits of RS

1.1 Intro !ction (e have learnt in class )", that Remote sensing *RS+ is the observation of an ob,ect, surface or phenomenon through the use of a variety of recording devices that are wireless, or not in physical or intimate contact with the ob,ect. &n aircraft, spacecraft, satellite or ship may be used for this purpose and e-uipped with recording devices such as camera, laser, radar, sonar etc. Remote sensing deals with inventory, monitoring and assessment of natural resources through analysis of data obtained from remote sensing platform.

Remote Sensing measures energy such as ultra.violet, infrared, microwave,which that cannot be reached by human vision. Remote sensing data has a uni-ue advantage of multidisciplinary application. 'he same RS data can be used by researchers / wor/ers in different disciplines such as geology, forestry, land use, agriculture, hydrology etc. "t offers wide regional coverage and good spectral resolution. Sun is the principal source of 01 energy 0arth receives only 2/34 millionth of total solar energy 567 of solar energy reaching the earth falls in the visible region of 01

0lectro 1agnetic Spectrum 'he basic principle involved in remote sensing is that different ob,ects reflects or emits radiations in different wavelengths and intensities depending 9b,ects properties. 01R serves as the main communication lin/ between the sensor and the ob,ect. &ll ob,ect / matter that has a temperature higher than absolute :ero 4; emits 01R continuously.

Stages in Remote Sensing & source of electromagnetic energy 'ransmission of 0nergy from the source to the surface of earth "nteraction with the intervening atmospheres "nteraction of 01R with the earth8s surface 'ransmission of 0nergy from the surface to the remote sensor Sensor !ata output

!ata transmission , #rocessing and &nalysis 'he intensity of the emitted radiation is depends upon the composition and temperature of the body. & blac/body is an ideal body that absorbs all radiation incidents on it without any reflection. "t represents a continuous spectral emission curve, in contrast to natural bodies that emit only at separate spectral bands. 'emperature plays great role on the intensity of

blac/body emitted radiation. 'his relationship is called (ien8s displacement

aw.

aw

represent as: < ma= > & / ' where < ma= is the wavelength *cm+ where highest radiation occurs. & is constant *> 4.?@ cm %+ and ' is the temperature *%+ of the ob,ect. Using this law it can estimate the temperature of ob,ects by measuring the wavelength of pea/ radiation. 'he above figure shows spectral

Spectral distribution of energy by a blac/ body distribution of energy radiated from blac/ bodies of various temperatures such as sun, incandescent lamp, fire and 0arth. Aor the Sun < ma= occurs at 4.5B Cm, which measures the temperature of the Sun appro=. as 6444 % similarly for the earth, the ambient temperature is D444 % and < ma= occurs at @.ECm. 'he ambient temperature of fire is 3444 % and for incandescent lamp it is 5444 %.

Blac/ body is an ideal body which absorbs all radiation without any reflectance. (hereas white body completes reflects incident radiation and absorbs nothing

1ost useful regions of the 01R are visible, "nfra red and thermal and microwave for carrying out RS activities. 'he human eye can detect energy in the visible portion of the electromagnetic spectrum. #hotographic cameras are sensitive to broader range of wavelength ranges from 4.D Cm G 4.@ Cm, the near ultraviolet to the near infrared. 'hermal scanners operate in the thermal infrared portion of the spectrum. 1ultispectral scanners operate over a broad range of wavelengths from ultraviolet to thermal infrared. #assive microwave and active radar systems operate in microwave portion of the electromagnetic spectrum. "n case of dust, cloud, fog all wavelengths are e-ually scattered so they are loo/s white. Fatural uneven *rough+ surface scattered in multiple directions. Shorter wavelengths are scattered more than longer wave lengths. 'his type of scattering is seen more in ultraviolet and blue. 'hat is way the s/y would appear blue otherwise it would appear as dar/ space. 'he Sun is primary source of energy. (hen these solar rays arrive at the 0arth, the atmosphere absorbs or bac/scatters a part of them and transmits the rest. (hen rays stri/ing the land, ocean surface and atmospheric targets, such as air, moisture, and clouds the incoming radiation go through various modes of energy.interaction response for e=ample transmission, absorption, reflection and scattering as shown in both below figures.

Remote Seining System

Source: HR"S#

Transmittance - 'he radiation penetrates into certain surface materials such as water and if the material is transparent, it transmits 2447 of energy. Absor"tion # "t occurs when 01R passes through an opa-ue mediumI Re$%ectance# Reflection occurs when incident 01R bounces off from a smooth surface Scattering G "t occurs when 01R disbursed in all the direction from a $arious modes of energy "nteraction response of incoming radiation

rough surface. "n this picture the energy interaction with red roof, glass window J blac/ soil are shown. (hen incident energy stri/es on red roof it reflects in red band, so roof appears red. Similarly the blac/ soil absorbs all the energy therefore it appears in blac/ color. (hen the energy passes through glass window it transmits all energy that is why it appears transparent in color.

0nergy "nteraction with (indow, Roof and Soil 1.& S"ectra% Re$%ectance Signat!re 'he reflectance characteristics of the different features of the earth surface are measured by the incident energy that is reflected by the surface. 'his spectral reflectance of natural features are collected and stored by satellite sensors . Spectral reflectance of any ob,ect usually varies according to the wavelength of the 01R. & graph showing the spectral reflectance of an ob,ect for various wavelengths is /nown as a Spectral Reflectance Hurve. "t 'he radiation reflected as a function of the wavelength is called the spectral signature of the surface. helps in selecting the wavelength bands for identifying the ob,ect. Spectral reflectance characteristics are the most important aspect for feature classification in any satellite imagery. 'ypical spectral reflectance curve for soil, vegetation, water are shown in above graph. !etails of spectral behaviors of soil, water and vegetation are discussed below separately. Radiation reflected or emitted from earth surface is converted to signal *!igital Fumbers. !F+ 'he reflectance from a feature depends on the atmospheric conditions, seasons, time of a day, physical and chemical characteristics of the feature Secular reflectance from smooth surfaces e.g. water, paved roads is lower than

Spectral Reflectance curve for (ater, $egetation and Soil

'he graph shows the reflectance of water in blue band, vegetation in green and infrared band where as bare soil shows linear reflectance in visible and infrared bands. "n

below figure vegetation reflects in green band and the red roof reflects the energy in red band the green therefore and red color appears as respectively.

Reflectance from roof and vegetation

Soi% Reflectance characteristics of soil are depending on the various factors: Soil moisture content Sand, silt J clay Homposition Soil te=ture, color J grain si:e Surface roughness 1ineral composition.

& typical reflectance curve various soils is shown in two graphs. Airst graph shows spectral behavior of dry and

for

wet soil. 'he

Reflectance curve based on soil moisture

presence of moisture in soil will decrease the reflectance in the visible and near.infrared regions. !ry soil increases reflectance in visible and near infrared wavelength as compare to wet soil due to less moisture content. Second graph shows reflectance curve for soil based on soil te=ture.

'here is high reflectance in visible and infrared band, for sandy loam as compare to fine sandy loam and very fine sandy load due to its fine te=ture soil. 'he presence of iron o=ide in soil will also decreases the reflectance in visible wavelength. Reflectance curve for soil based on soil te=ture

'egetation Kealthy growing vegetation appears green because of Hhlorophyll content. 'he reflectance from vegetation depends on various factors: eaf pigmentation eaf cell structure eaf moisture Hrown architecture #lant physiology

'he

reflectance of healthy green vegetation is shown in graph above. "n the visible band of 01R the healthy vegetation will have absorption in the blue and red bands because of the Reflectance curve of vegetation based on health presence of chlorophyll. 'his wavelength ranges from 4.53 Lm and 4.63 Lm. "f the plant is turned to yellow it results in less chlorophyll so there will be less absorption in blue and red band. &s we move from visible to near infrared portion of the spectrum which is about 4.E5 Lm to 2.D Lm the plant leave reflects 54 to 347 of the energy incident upon it. "f the moisture is low in leaves of the plants then it reflects more energy in all the bands compared to the leaves with high moisture as shown in below graph.

'he plant stress in graph

Reflectance curve for low and high moisture vegetation

shows, decrease

absorption in F"R as compare to the healthy vegetation. !ue to less moisture, disease, and pest contents the crops reflectance are more. (ater 'he reflectance from water depends on various factors !epth Suspended particles in water Aloating vegetation Sun angle

#ure clear water has a relatively high reflectance in the visible wavelength ranges from 4.5 and 4.6 Lm. 'here is no reflectance in the near.infrared *4.E Lm+ and higher wavelengths. So in infra red "mage shows the water body in dar/ color. Hlear water absorbs little energy in 4.6

Lm. the

"f

water

Reflectance curve for clear la/e and turbid river water

contains of suspended sediments it has higher reflectance in visible as compare to clear water.

Keavy sediments loads *high turbidity+ prevent radiation penetrating water bodies. So it reflects the energy. "n Fear "nfra Red *F"R+ and 1iddle "nfra Red *1"R+ wavelengths water strongly absorbs radiation and very little energy will be reflected or transmitted. 'he presence of algae in water affects the spectral response of the water. 'he chlorophyll contents of algae increases the reflectance of water body in the green band as shown in below figure and reduces the reflectance in red and blue band, therefore water loo/s green.

Reflectance in green band due the presence of algae in water body

Roc) 'he roc/ constitutes of different minerals and te=tural properties. 'he absorption and reflectance are basically depends on mineral composition in roc/s. !ifferent roc/ such as Biotite granites and granite has less water contents therefore, they reflect more in infrared. 'he roc/ which contains more water content will absorb more. 'he roc/s which contents "ron will e=hibit more absorption in ultraviolet and blue band and reflection in infrared band. 'his is due the content of metals such as "ron 1anganese, Hopper, Fic/el and Hhromium. 'he other factors which affect the reflectance of the roc/ areI Fature of the roc/ 'op cover 'opography / shadow Surface roughness

Below curve shows reflectance behavior of various roc/s such as Honcrete, &sphalt, Bare soil, Mravel, and Shingles.

Reflectance curve for various roc/ types

'he reflectance of all the roc/ types is gradually increasing in all wavelengths such as with $isible, F"R, and 1"R. Honcrete is light and bright in color as compare to asphalt. 'herefore it shows higher reflectance as compare to all roc/s. 'he shingles are bluish in color so it reflects more in 4.5 to 4.3 Cm Reso%!tion* Resolution is defined as the ability of the sensor to detect the information at the smallest meaningful element, in terms of distance *spatial+, wavelength band of 01R *spectral+, time *temporal+ and radiation -uantity *radiometric+. 'here are four types of resolution which are listed belowI Spatial, Spectral, Radiometric, 'emporal.

S"atia% Reso%!tion* "t is the minimum element area that the sensor can detect to measure. Spatial resolution is classified into three types. Kigh spatial resolution which covers *4.6 G 5m+, medium spatial resolution covers from 5.D4 m J low spatial resolution covers D4 N 2444 m. 'his resolution element is called #i=el. 'he various e=amples and pictures of spatial resolution are given below.
Source: NRSC Hyderbad

0=ample for #i=el 'he spatial resolution specifies the pi=el si:e of satellite images covering the 0arth Surface. LISS III +an +an (i.S A(i.S LISS III / PA0 LISS I' I1O0OS Cartosat -& SPOT- 2SS & to , ?D.3 m E4.3 m 2BB.D m 36 m 3.B m ?.3 m 2m 4.B m D4 m, B4 m

Spatial resolution of various Satellites D4 m 1ultispectral * andsat . '1+ B4 m 1ultispectral * andsat . 1SS+

2 m #an *"%9F9S+
Source: FRSH Kyderbad

D4 m 1ultispectral * andSat+
Source: FRSH Kyderbad

S"ectra% Reso%!tion* "t refers to the sensing and recording power of the sensor in the different bands of 01R. 'he sensor can observe ob,ect separately in different bands and colors or in one band which is panchromatic * Blac/.(hite+. Aor e=ample, andsat 1SS . E Bands, S#9' G 5 Bands, "RS . 5 Bands.

LISS III
+an s (ave%engt3s 4.3?.4.3@ Cm 4.6?.4.6B Cm 4.EE.4.B6 Cm 2.33.2.E4 Cm 4.3.4.E3 Cm

Band ? Band D Band 5 Band 3 #&F

Spectral Resolution of various Satellites Ra iometric Reso%!tion* "t is determined by the number of discrete levels into which signals may be divided. "t is recorded in !igital Fumber *!F+ for different bands of Satellites. "n another words it is dividing the total range of the signal output of the sensor into a large number of distinct colors to enable distinguish ground features which are differing slightly in a radiance or reflectance. 'he various e=amples are given in below table: Ra iometric Reso%!tion E bit / 2?B levels E bit / 2?B levels 6 bit / 65 levels 24 bit/24?5 levels 24 bit/24?5 levels

Sensor LISS III (i.S PA0 A(i.S Cartosat-1

Co%or 4 .2 ?E Holors 4 . 2?E Holors 4 . 6D Holor 4 .24?D Holor 4 .24?D Holor

Radiometric Resolution of various Satellites

Radiometric resolution describes the ability of a sensor to discriminate very light differences in energy. 'he finer the radiometric resolution of a sensor it is more sensitive to detecting small differences in reflected or emitted energy. Tem"ora% Reso%!tion* 'he ability to collect imagery of same area of the 0arth8s surface at different periods of time is one of the most important elements of RS. 'emporal resolution is also called as the repetitive cycle which is the capability of the satellite to record the same area at the same viewing angle at different periods of time. Spectral characteristics of features may change over time J these changes can be detected by collecting J comparing multi.temporal imagery. Aor e=ample, Spot revisits the same area in ?6 days. 'emporal imagery helps us to identify the change occurred over a period of time. 'he various e=amples are given in below table:

Sensor "RS 2&/2B "RS 2H/2! #an (iAS &(iAS andsat

Revisit Perio (0o o$ a6s) ?? ?5 3 D 3 26 . 2B

'emporal Resolution of various Satellites 1.4 5igita% Image Processing (5IP) 'he spatial data is ac-uired by remote sensing technology in the form of "mages / #hotographs. 'he raw data received from the imaging sensors on the satellite platforms contains inaccuracies. 'o overcome these inaccuracies it needs to undergo several steps of processing. $arious techni-ues are used to enhance -uality of the images. 'his techni-ue is called as !igital "mage processing. !igital processing has the greatest potential for preserving the correct radiometry and the ma=imum resolution of the images. !"# is the collection of algorithms processed by the computer system to enhance the -uality of raw data in order to get enhanced "mages for further process of i"nterpretation and data e=traction.

'he steps involved in !"#. "mage restoration Statistical analysis "mage enhancement "mage classification

Image Restoration "t refers to correct the distorted or degraded image to represent original scene. 'he steps includes in image restorations areI Meometric correction *"mage Rectification+ Radiometric correction Foise removal

"mage restoration processes are designed to identify noise, geometric distortion introduced into the data during scanning, transmission J recording processes. 'he ob,ective is to map the image resemble to original scene.

7eometric Correction 'he geometric errors in image occur due to perspective of the sensor, scanning system, the Meometric distortions include nonsystematic distortion such as variation in spacecraft altitude, velocity J altitude. 'hese distortions must be corrected from ground control points. River &n e=ample for Meometric Horrection

motion of platform and curvature J rotation of the

0arth. Because of these the image e=hibits some sorts of scaling, s/ewing, rotation errors. 'his correction procedure is called as geometric corrections or rectification. Rectification is

the process of assigning the geo locations to the "mage / #hotograph to remove the geometric distortion. 'he figure shows an e=ample of corrected and distorted image. Ra iometric correction 'he reflectance of the ground features varies as the time and location varies. (hile mosic/ing two different images ac-uired at different time and location it is necessary to apply sun elevation and earth sun distance corrections to normali:e the reflectance of the images. 'his process is called radiometric corrections. 0oise Remova% "mage noise is any unwanted disturbance in the image data due to limitation in the sensing instrument or data recording process. Resampling method is used to remove the noise from images. Image En3ancement "mage enhancement techni-ues are applied to improve the -uality of an image. "mage enhancement is applied after correcting geometric and radiometric distortions. 'here are various types of techni-ues available to improve image -uality for father analysis and interpretation. 'he contrast stretch, density slicing, edge enhancement, and spatial filtering are commonly used image enhancement techni-ues. "mage enhancement processes consists of different techni-ues employed in the calibration of image data for the correction or reduction of errors occurring during capture or transmission of the data. "t increases the ability of the analyst to recogni:e features of interest. Histogram Kistogram is a graph showing number of pi=els in an image each different intensity value. an B.bit grayscale image there ?36 different 'hus possible the intensities. the at Aor are

histogram will graphically display the range from 4.?33 numbers showing the distribution of pi=els. Kistogram e-uali:ation is the techni-ue by which the dynamic range of the !F values of an image is increased. Kistogram e-uali:ation assigns the Kistogram

intensity values of pi=els in the input image in such a way that the output image contains a uniform distribution of intensities. "t improves contrast and obtains a uniform histogram. By modifying this histogram it enhance the image which is often called Hontrast stretching or histogram normali:ation. "t is a simple image enhancement techni-ue that improve the contrast in an image by OstretchingP the range of intensity values. 'he e=amples are shown in this below figure. &fter applying the histogram manipulating techni-ues image loo/s more clearer than the raw

9riginal image with Kistogram

1odified image after e-uali:ed histogram Hontrast generally refers to the difference in grey level values in an image. "t other words it is the ratio of the ma=imum intensity to the minimum intensity over an image. arger ratio ma/es easy interpretation of the image. Satellite images lac/ in ade-uate contrast and re-uire contrast improvement. Hontrast enhancement techni-ues e=pand the range of brightness values in an image so that the image can be efficiently displayed.. Kistogram shows the number of pi=els that correspond to each !F inear Hontrast Stretch is the simplest contrast stretch algorithm. 'he grey values in the original image and the modified image follow a linear relation. & density number in the low range of the original histogram is assigned to e=tremely blac/ and a value at the high end is assigned to e=tremely white. 'he remaining pi=el values are distributed linearly between these e=tremes. 'he features or details that were unclear on the original image will be clear in the contrast stretched image. inear contrast stretch operation can be represented graphically as shown in below Aigure.

"n Fon.

inear Hontrast 0nhancement Source: nasa.gov

linear contrast

enhancement, the input and output data values follow a non.linear transformation. Fon.linear contrast enhancement is defined by y > f *=+, where = is the input data value and y is the output data value. 'he non.linear contrast enhancement techni-ues have been found to be useful for enhancing the color contrast between the nearby classes and subclasses of a main class. inear contrast which greatly improves the contrast of the original brightness values but there is a loss of contrast at the e=treme high J low end of !F values.

Spatial filtering is another techni-ue of digital processing functions. "t is used to enhance the appearance of an image to derive valuable information. "n the real world there are

boundaries between features which cannot be seen in raw images. 'hese boundaries can be emphasi:ed using several computer algorithms. &lgorithms for this purpose are called QfiltersQ. Spatial filters are designed to highlight or suppress specific features in an image based on their spatial fre-uency. & low . pass filter bloc/s the high spatial fre-uency details. "t is designed to emphasi:e larger, homogeneous areas of similar tone and reduce the smaller detail in an image. ow. pass filter generally serve to smooth the appearance of an image and reduce Rsalt and pepperQ noise as shown in below figure. 'he most commonly used low pass filters methods are mean, median and mode

Raw "mage . Source: nasa.gov

Kigh pass filet are used to emphasi:e fine detail and edges. Below figure shows the e=ample for Kigh pass filter in an "mage. 'his filter enhances details and clear edges and abrupt discontinuities. 'herefore roc/ ,oints, faults, field boundaries, and street patterns are clearly visible as compared to raw image

Raw "mage

Source: nasa.gov

+an ratioing "mage division or spectral ratioing is one of

&n e=ample of Kigh #ass Ailter Source: nasa.gov the

most common techni-ues applied to image data. Ratio "mages are division of !F values of one spectral band by corresponding !F of another band. 'he ratio has been used to identify different features. Aor e=ample the ratio of near.infrared / red is less than 2 it represents water and greater than 2 it represents vegetation. F"R / R images are used to identify vegetated area. "mage ratioing highlights slight variations in the spectral responses of various surface features. Kealthy vegetation reflects strongly in the near infrared portion of the spectrum while absorbing strongly in the visible red. 9ther surface types, such as soil and water, show nearly e-ual reflectance in both the near.infrared and red portions. & ratio image of andsat 1SS Band E *Fear. "nfrared . 4.B to 2.2 mm+ divided by Band 3 *Red . 4.6 to 4.E mm+ would result in ratios greater than 2.4 then it will be easy to identify the vegetation, soil and water. 'his is the most common arithmetic operation applied to images in geological, ecological and agricultural applications of remote sensing.

F"R/R *'15/'1D+, with values ranging from 4 to B, soil to vigorous vegetation, vegetated areas shown in white.

9riginal "mage

0orma%i8e 5i$$erence 'egetation In e9 (05'I) F!$" is used to assess, analy:e and estimate live green vegetation, crop yields, pasture etc. F!$" is directly related various parameters Mround cover #hotosynthetic activity of the plant surface water eaf area inde=.

Several algorithms are used to e=tract such information from remote sensing data, which are referred as vegetation indices. F!$" is a numerical indicator that uses the visible and near. infrared bands of the electromagnetic spectrum. Kealthy vegetation absorb in visible band and reflects in near.infrared band. Unhealthy vegetation reflects in visible band and less in near.infrared band. 9n the other hand bare soils reflect moderately in both the red and infrared band of the electromagnetic spectrum. By /nowing the behavior of plants across the electromagnetic spectrum F!$" information can be derived by using following formula. Fear. infrared and red bands are most sensitive to deriving vegetation information by the sum of near.infrared and red bands. F!$" > *F"R . R0!+ / *F"R S R0!+ (here F"R . NFear "nfrared band R0! . NRed band from visible range 'he F!$" algorithm subtracts the red reflectance values from the near.infrared and divides it. F!$" values are represented as a ratio ranging from .2 to 2. 0=treme negative values represent waterI values around :ero represent bare soil. 'he bigger the difference between the near.infrared and the red reflectance, there will healthy vegetation. "n the below figure the F!$" is calculated in %harif Season in #un,ab from 1ay to Fovember months. 'he dar/ green, yellow color shows less vegetation *1ay, Tune ,9ctober+ and dar/ red color shows healthy vegetation.

F!$" 1ap for #un,ab during %harif Season Source: #RS0 udiyana 'egetation In e9 (P'I) 'he plot of reflectance in the and near infrared bands, of soil, under various degrees of wetness, is the baseline for vegetation inde=. 'he illustration is shown in figure. is measured perpendicularly this soil baseline. #oint & a higher #$" than point B. mathematically this measure represented below #erpendicular vegetation inde=: *soilR - vegR)& / (soi%IR - vegIR)&

Per"en ic!%ar red bare the #$" from has is

Image C%assi$ication Hlassification is a process to categori:e all pi=els in a digital image into one or several land cover classes, or QthemesQ. !igital image classifications uses the spectral information represented by the digital numbers in one or more spectral bands, and classify each individual pi=el based on this spectral information. Hlassification techni-ues are generally applied to the single.date image or series of multi.date images for change detection. Formally, multispectral data are used to perform the classification. Mrouping of the pi=els are "mage classification is a process of automatically categori:ation of all pi=els in an image into different land cover classes or themes based on their spectral information *!F+ present within the data. 'he ob,ective of image classification is to identify different features occurring in an image to identify different land features. 'hese groups are called as classes. Based on !F values we can create the classes li/e $egetation, (ater body, Barren and, Buildup &rea etc. Broadly we can use two approaches for classification: 2. Unsupervised classification ?. Supervised classification "n Unsupervised classification spectral classes are grouped first, based on the numerical information in the data. Hlustering algorithms are used to determine the natural classes in the data. &lgorithms e=amine the un/nown pi=els in an image and group them into a number of Unsupervised classification is popular in industries involved in long term M"S database maintenance. Because it uses clustering procedures which are e=tremely fast. classes. Unsupervised Hlassification method does not utili:e training data as the basis for

classification. Based on the reference data, area /nowledge and e=perience, user compare the classified data and identify the features and name them as shown in the figure. 'here are numerous clustering algorithms that can be used to determine the natural spectral groupings present in data set. 'he most used algorithm is R%.meansU approach also called as "S9!&'& *"nteraction Self.9rgani:ing !ata &nalysis 'echni-ue+. Unsupervised classification is not complete without human involvement.

Unsupervised Hlassification #rocess Supervised classification is more accurate for mapping classes, but it depends heavily on the area /nowledge and s/ills of the analyst. 'he analyst should recogni:e classes in an "mage based on prior /nowledge and assign class names. 'hese are called as training sites. 'he training sites areas representing /nown land cover category that appear homogeneous on the image. "mage processing software categori:es of the reflectance of each class. 'his process is called Qsignature analysisU. 'he entire image is classified based on this signature assigned. 1ost fre-uently used algorithms for supervised classification are parallelepiped, minimum distance and ma=imum li/elihood. 'he basic steps involved in a typical supervised classification procedure are shown in figure.

Steps in Supervised Hlassification !efine training site Aeature selection *signature analysis+ Selection of appropriate classification algorithm #ost classification smoothening &ccuracy assessment Hlassified "mage *Result+ Statistical report generation for different landuse and landcover

"n below figure the training sites are shown and names are assigned such as 'urbid (ater, Raw "mage
Source: FRSH Kyderbad

!eep (ater,

"mage with 'raining Sites


Source: FRSH Kyderbad

Shallow (ater, (et

and, Barren

and, &griculture and Aorest.

Based on these training sites the raw "mage is classified and final output is shown below.

Hlassified "mage Source: FRSH Kyderbad

&n e=ample for supervised classification along with the Statistical report on and use/ and covers *Source : "SR9 Report+

&fter classification the classified map needs to be compared with other map of same area to assess and verify the accuracy. "n areas of comple= terrain, the unsupervised approach is preferable as compare to supervised classification. "n such conditions if the supervised classification is used, the user will have difficulty in selecting training sites because of the variation of spectral response within each class. &s unsupervised classification distinguishes classes based on spectral response, ground truthing re-uirements are reduced for some

e=tent. . Unsupervised classification reveals separate classes of un/nown places without visiting to the site. 1., 'is!a% Inter"retation $or Sate%%ite 5ata "magery obtained by remote sensing is used in various applications. By studying the -ualitative and -uantitative aspects of images one can derive useful information by having /nowledge about the area for further interpretation and analysis. "mage interpretation is the art and science of e=amining image to identify the ob,ects and evaluates their significance. "dentifying features in remotely sensed images are based on followingI Shape Si:e #attern 'e=ture Shadow &ssociation "mage interpretation is defined as e=amining images to identify ob,ects and ,udge their significance.

Tone - one refers to the relative brightness *color+ of ob,ects in an image. 'he term is used for each distinguishable shade from blac/ to white, such as dar/, medium, light gray as shown in picture. 'one is the important element for identifying different features. $ariations in tone also help identification of shape, te=ture, and pattern of ob,ects. Aor e=ample dry sand will appear in light shades of grey, while wet sand appears in dar/ shades of grey. "n above figure road loo/s brighter than the wet field. S3a"e Shape refers to the general form, structure, or outline of individual ob,ects. Shape is a very important element for interpretation. Straight edge shapes represent urban or agricultural fields. Fatural features, such as forest edges, are generally more irregular in shape. Aarm or cropland irrigated by rotating sprin/ler systems 'onal $ariation in an "mage

Aeature indentified based on Shape

Aeatures indentified based on tonal $ariation would appear as circular shapes. 1any geomorphologic shapes are identified, such as sand dunes la/es, volcanic cones etc. Aor e=ample in above figure the tennis court is visible.

Si8e 'he si:e of a feature is also significant element in image interpretation. Si:e of ob,ects in an image is a function of scale. Si:e when considered along with shape and this association, figure large helps buildings in such easy in as identification of the features. Aor e=ample factories or warehouses would suggest commercial property, whereas small Aeature indentified based on Si:e
Source: HHRS

buildings would indicate residential use.

Pattern #attern refers to the spatial arrangement of visibly separate ob,ects. &n orderly repetition of similar tones and te=tures will produce a distinctive and recogni:able pattern. Urban streets with regularly spaced houses are good and orchards with evenly spaced trees as shown in figure are good e=amples of pattern.

Aeature indentified based on #attern

Te9t!re 'e=ture is closely associated with tone. 'his includes smooth, and rough surfaces. Rough surfaces have a spotted tone where the grey levels change abruptly in a small area. Aor e=ample forest canopy shows rough te=ture in an image.

Smooth surfaces have very little tonal variation. Aor e=ample fields or grasslands appear smooth and uniform in image. Aeature indentified based on 'e=ture
Source: HHRS

Aeature indentified based on 'e=ture *Source RRH Forth+ S3a o: Shadow is an important element for "mage provides interpretation. easier of profile "t the ma/es it features. reduce non for identification because

Shadow length can be used for height measurements. Shadows interpretation in that area because of visibility of features. Shadow is useful

Shadow in an "mage

identifying topography and landforms. Aigure shows that the shadows of the tall building which covers the features from other side of the building which cannot be seen due to this information cannot be derived.

Association &ssociation of ob,ects is one of the most important elements of image interpretation. "t ta/es into consideration the relationship between other ob,ects which are near to the target.. Aor e=ample commercial properties be associated with ma,or transportation an routes, whereas in residential areas if we see open space it would be associated with schools, playgrounds, and sports fields. "n above figure, a la/e is associated with a marina, and ad,acent recreational land. 1.Aeria% P3oto an its Inter"retation Aeatures are indentified by their association
Source: HHRS

may

boats,

&erial photography is obtained by using aircraft as platforms. &erial photographs give Bird8s eye view of a large area enabling to view earth surface features which cannot be possible to obtain through ground observation. &erial photograph can record earth features in visible, ultra violet and near infrared in the form of a visible image. (ith proper selection of camera films and flight parameters it is able to record more spatial details on photograph as compare to human eye. (ith proper ground reference data it can obtain accurate measurement of positions, distance, directions, areas, heights, volumes and slope. &erial photography is the permanent record of e=isting condition which can be studied whenever re-uired. Single image can be used by a large number of users for different types of application such as

geology, agriculture, soil, crop study. &erial photos are also useful to study dynamic phenomenon such as floods, wildlife population, traffic, oil spills and forest fires. 'he films used in aerial camera are of two types #anchromatic *B J (+ G "t is cheap and commonly used in #hotogrammetry. Holor. it is easy to interpret but fu::y due to atmospheric scattering

5etermining t3e P3oto sca%e: Scale of the #hoto can be derived by the following formula S>f/K (here: S > Scale of &erial #hotograph f > Aocal length of Hamera *normally 23? mm+ K > Alying height "f 2mm on a photograph represents ?3 meter on the ground then the scale of photograph is represented as 2mm > ?3 m *represents e-uivalents+. "f it is 2/?3444 *it represents fraction+ 2:?3444 *represents the ratio+. Alights are usually scheduled between 24 am to ? pm to have low wind, clear s/y with ma=imum illumination and minimum shadow to obtain clear weather picture. Aebruary to &pril months is optimal period to ac-uire the aerial photographs. &erial #hotographs can be used for flood, vegetation mapping, and soils study 'he various factors affect the -ualities of the aerial photos which are: Hlouds, ha:e, shadows/sun angle, snow !istortion due to tip J tilt, relief distortion, radial distortion Storage and handling can be a problem imited to 4.D . 4.@ m *U$.F"R+ $ertical 9bli-ue . Kigh J ow

&erial #hotographs are classified as follows:

Stereo/D!

'ertica% $ertical photographs are made with the camera a=is directed as vertically as possible. 'he photographs made by single lens frame camera. 'his is most common type of aerial photography used in RS application. 'he vertical air photos with less than D; tilt are considered as vertical photographs. 'o achieve this, it re-uires the special e-uipment to maintain the verticality between the lens a=is and to the ground. $ertical photos used for planning and redesigning industrial sites. Ob%i;!e &n e=ample of $ertical #hoto are the

(hen the aerial photographs are ta/en with an intentional inclination of the camera a=is is called obli-ue photographs. 9bli-ue photographs are easy to understand because it allows identification of structural and topographic variation. "t is easy and cost effective to ac-uire. 'he photographs with more than D; tilt are considered as obli-ue. "t is used to study commercial, residential, industrial, and transportation buildings and infrastructure. 'here are two basic types of obli-ue aerial photography. 2. Kigh angle obli-ue ?. ow angle obli-ue. "n a high angle obli-ue, the apparent hori:on is shown, while in a low angle obli-ue the apparent hori:on is not shown. "n high obli-ue photo the camera inclination angle is from D4 to 64 degrees where in ow obli-ue the inclination angle is 3 to D4 degrees.

&n &n e=ample e=ample of of Kigh ow 9bli-ue #hoto

Stereo 45 Stereo photography is a techni-ue to ma/e two photographs of the same sub,ect, from slightly different positions. 'hese two positions should differ appro=imately about 24 cm. 'hese images are called stereo images and can be viewed using stereoscope or by stereo computer graphics devices. 'he resultant image appears to be D dimensional and depth can be seen. 'his depth can be controlled by increasing or decreasing the distance between photography locations.

Stereo photography involves ta/ing a photograph from two positions, which corresponds to two QeyeQ positions. #hotographs are ta/en generally with some percentage of overlaps. Aor stereoscopic coverage the general overlap is: Aorward 9verlap : 647 *for stereoscopic studies+ ateral 9verlap : ?4.D47 *for edge matching+

Inter"retation o$ Aeria% P3oto 'he interpretation of aerial data is -uite similar with Satellite images. 'he "nterpretation /eys for aerial photographs are listed below Shape Si:e #attern 'e=ture Shadow &ssociation

'he details about these /eys are already discussed in $isual "nterpretation of Satellite data.

&erial #hotograph #anchromatic

1.< A vance Remote Sensing Tec3no%ogies


H6"er S"ectra% Imager6 Kyper spectral sensors collect information as a of PimagesP. 0ach image represents a range of electromagnetic spectrum and is also /nown spectral band. 'hese PimagesP are combined form a three dimensional hyper spectral cube processing and analysis. Kyper spectral cubes generated from airborne sensors li/e the F&S&8s &irborne $isible/"nfrared "maging Spectrometer *&$"R"S+, or from satellites li/e Kyper Spectral "mages with multiple bands
Source: rst.gsfc.nasa.gov

set the as a and for are

F&S&8s Kyperion. Kyper spectral imaging collects and processes information from visible as well as from the ultraviolet to infrared band. 'hese data can be used for application in agriculture, mineralogy, physics, and surveillance. Kyper spectral images are ac-uired at entire spectrum. So it does not re-uire prior /nowledge of sample data, and post.processing allows all available information from the dataset to be used. Kyper spectral allows preparing more accurate models and classification of image. 'he disadvantages of these data are costly J comple=ity. Aast computers, sensitive detectors, and large data storage capacities are needed for analy:ing hyper spectral data. T3erma% Remote Sensing Remote sensors cover two thermal intervals that are D . 3 Cm and B . 25 Cm broad bands. "t allows sensing of thermal emissions from the land, water, ice and the atmosphere. 1any of the meteorological satellites include at least one thermal channel. & thermal band is included on the andsat 'hematic 1apper.

'hermal "mage
Source: rst.gsfc.nasa.gov

2icro:ave Remote Sensing 1icrowave band in 01R radiation is used in remote sensing to

provide useful information about the 0arthPs atmosphere, land and ocean. 1icrowave region in 01R ranges from 4.2 to D4 centimeters is called the. Because of their long wavelengths microwave radiation can penetrate through cloud cover, ha:e and dust. 'he microwave sensing detects of microwave energy under almost all weather and environmental conditions so that data can be collected at day and night both. 1icrowave Remote Sensing is classified in two categories namely: active and passive microwave remote sensing. #assive microwaves Remote sensing uses emitted energy from thermally activated bodies of earth surface. &ctive remote sensing uses manmade device such as R&!&R, S9F&R, idar etc. &ctive microwave remote sensing system provides their own illumination. 'h e

basic principle of the "maging radar is to emit electromagnetic radiation towards the earth surface and record the -uantity and time delay of energy bac/scattered.

Radar is also called as a distance measuring device.

'he e=ample for Radar images ac-uired by 0RS.2 and

andsat.3 on the

same date on the fourth of Tuly 2@@B *Source FRSH.Kyderbad+ Radio !etection and Ranging *Radar+ is an active sensor system. "t consists fundamentally of a transmitter, a receiver, an antenna, and an electronics system to process and record the data. "t generates its own illumination to interact with the surface features. Some features bac/scattered the energy which is collected by radar antenna. 'he bac/scattered signal is measured to discriminate between different features, the time delay transmitted and reflected energy. 'he transmitter generates successive pulses of microwave *&+ at regular intervals which are focused by the antenna into a beam *B+ shown in the figure. 'he radar beam illuminates the surface at a right angle to the motion of the platform. 'he antenna receives a portion of the transmitted energy reflected / bac/scattered from various earth surface ob,ects within the illuminated beam *H+. 'he location of the ob,ect can be determined by measuring the time delay between transmission of a pulse and the reception of the bac/scattered QechoQ from different features. 'hus creates two.dimensional image of the surface. Radar operates in part of the microwave region of the 01S in the fre-uency ranges from 54,444 to D44 megahert: *1K:+ and higher fre-uencies of the broadcast.radio region. Below table shows the fre-uencies and their corresponding wavelengths used in R&!&R. it the R&!&R 1echanism as

S%. 0o 1 & 4 , < > ?

+an 1a 1 1! = C S L P

.re;!enc6 54,444.?6,444 1K: ?6,344.2B,344 1K: 2B,444 G 2?,344 1K: 2?,344.B,444 1K: B,444.5,444 1K: 5,444.?,444 1K: ?,444.2,444 1K: 2,444. D44 1K:

(ave%engt3 Range 4.B.2.2 cm 2.2.2.6E cm 2.6E.?.5 cm ?.5.D.B cm D.B.E.3 cm E.3.23.4 cm 23.4.D4.4 cm D4.4.244.4 cm

'his chart summari:es the above information on Bands in the 1icrowaveregions the 01 spectrum: Source: rst.gsfc.nasa.gov Interaction bet:een 2icro:aves an Eart3@s S!r$ace (hen microwaves stri/e a surface, the proportion of energy scattered bac/ to the sensor depends on many factors:

#hysical factors such as the dielectric constant of the surface materials which also depends strongly on the moisture content Meometric factors such as surface roughness, slopes, orientation of the ob,ects relative to the radar beam directionI 'he types of land cover *soil, vegetation or man.made ob,ects+.

1icrowave fre-uency, polari:ation and incident angle.

1icrowave remote sensing used in various types of application such as meteorology, hydrology, and oceanology. 1eteorologist can use microwave to measure atmospheric profiles and to determine water and o:one content in the atmosphere. Kydrologists use microwaves to measure soil moisture. 9ceanographic applications include mapping sea ice, currents, and detection of pollutants, such as oil slic/s. Lig3t 5etection an Ranging (Li ar) idar is an active remote sensing techni-ue. 'his technology involves the use of pulses of laser light coming to ground the measuring the time of pulse returning to sensor as shown in the figure. 'he return time of each pulse bac/ to the sensor is processed to calculate the variable distance between the sensor and the ob,ect. 'here are three basic generic types of idar Range Ainders !ifferential &bsorption idar *!"& + !oppler idars.

Range finder idars are the simplest idars. 'hey are used to measure the distance from the idar instrument to solid or hard target. !"& is used to measure chemical concentrations such as o:one, water vapor and pollutants in the atmosphere. !oppler idar is used to measure the velocity a target. resolution bare earth idar can create Kigh. !01s with in 24 cm of a

accuracy. idar is also able to map elevations dense "!&R Remote Sensing *Source 0&&R System+ canopy forested or vegetated areas .

!01 prepared by idar !ata 1. > A vantages an +ene$its o$ Remote Sensing (RS) !uring last two decades the satellite remote sensing has been effectively utili:ed in ac-uiring and analy:ing valuable information about the earth resources. Remote sensing dramatically enhanced men8s capabilities for resource e=ploration, mapping and monitoring the earth resources environment on local as well as global in scales. various "t has brought as tremendous change in both -ualitative and -uantities areas such communication, television, radio broadcast, meteorology, agriculture, education, disaster management, environmental monitoring, and natural resource inventory. 'hese are important for strong database for planning, &dvantages of Remote Sensing Synoptic view Repetitive coverage continuous ac-uisition of data Hoverage of inaccessible areas Up.to.date information, accurate and reliable data Vuantifiable data Few information 1ulti disciplinary applications 'ime, manpower saving Vuic/ assessment of resources Serves as a large archive of historical data of has

process and implementation of various programs and pro,ect both at the national and regional levels. Aew benefits of Satellite remote sensing are listed below !ata collected using satellite remote sensing can be used for following purposes

&ssessing and observing vegetation types Honducting soil surveys Harrying out mineral e=ploration 1ap ma/ing to facilitate easy study of information Honstruct thematic maps based on re-uirement #lanning and monitoring water resources Harry out urban planning &ssessing crop yields and other agriculture management &ssessing and managing natural disaster, etc. 'o study the various spatial features in relation to each other and regional features trends, phenomenon delineation of

Overvie: o$ 5igita% Imaging So$t :ares !igital image processing software8s allows the use of algorithms for image processing to perform following processes:

"mage modification #i=eli:ation #ro,ection "mage enhancement inear filtering Hlassification Aeature e=traction #attern recognition #rincipal components analysis "ndependent component analysis and many other spatial analysis and modeling

Some of the popular "mage processing software8s are listed belowI

S%. 0o 1 & 4 ,

So$t:are 0ame Rolta Meomatica 0rdas "magine "mage &nalyst 0F$"

'en ors 0ame Rolta "ndia td 0rdas "nc "ntergraph "'' $isual "nformation Solution

Im"ortant :ebsites to get more in$ormation on Remote Sensing www.isro.org www.iirs.nrsc.gov.in www.india.gov.in www.esriindia.com www.gisdevelopment.net www.en.wi/ipedia.org/wi/i/remote_sensing www.ias.ac.in www.nrsc.gov.in/ www.ccrs.nrcan.gc.ca

Let@s :ra" !" :3at :e covere in t3is c3a"ter Remote sensing is the process of ac-uiring the information about earth features without being into direct contact Remote sensing *RS+ deals with inventory, monitoring and assessment of natural resources through the analysis of data obtained from remote sensing platforms. RS enables continuous ac-uisition and receives up.to.date information. 'he techni-ues also help in saving time and manpower.

'he energy emitted from is called as electromagnetic radiation. & blac/body is an ideal body which absorbs all radiation without any reflection. &ll ob,ect matter that has temperature higher than absolute :ero or .?ED degrees emits 01R continuously Sun is the primary source of energy. 'he incoming radiation goes through various modes of energy interaction, for e=ample, transmission, absorption, reflection and scattering "n case of dust cloud and fog all wavelengths in the visible band are e-ually scatters, that is why all loo/s white in color Uneven *rough+ surface scatters in multiple direction 'he shorter wavelengths are scattered more than longer wavelengths. 'hese types of scattering are seen more in ultra violet and blue. 'hat8s why s/y appears in blue. (hen energy passes through the glass windows it transmits all the appears transparent in color 'he radiation reflected as function of the wavelengths is called as spectral signature of the surface 'he reflectance from a feature depends on the atmospheric condition, season, time of a day, and physical J chemical properties of the feature Smooth surface are low reflectance as compare to rough surface 'he reflectance of the vegetation depends on various factors such as leaf pigmentation, leaf cell structure, moisture, and crown architecture and plant physiology 'he healthy growing vegetation reflects in green band that8s why color appears as green 'he healthy vegetation will have abortion in blue and red band of the chlorophyll 'he reflectance of the soil depends on soil moisture, te=ture, color, grain si:e, sand, silt Jd clay composition and mineral composition. 'he reflectance from water depends on depth, suspended particles, floating vegetation #ure clear water has a relatively high reflectance in the visible wavelength between 4.5 and 4.6 Lm because it absorbs all in color bands because of the presence energy so it

the energy therefore, it appears dar/er

Keavy sediments in water prevents radiation penetration that why it reflects the energy in F"R 'he absorption and reflectance roc/s depend on the various factors such as nature of the roc/, top cover, topography, shadow and surface roughness. Resolution is defined as the ability of the sensor to detect the information at the smallest meaningful element 'here are four types of types in resolution, those are, spatial, spectral, radiometric, and temporal 'he spatial resolution is the minimum element area that the sensor can measure. 'he resolution element is called pi=el. Aor e=ample "SS """ has ?D.3 m resolution, #an 3.B m Spectral resolution is refers to sensing and recording power of the sensor in the different bands of 01R. Aor e=ample andsat G 1SS E bands, "RS. 5 Bands. Radiometric resolutions recorded in digital number of different bands of sensors. Aor e=ample "SS.""" E bit *2?B levels+ color is 4.2?E colors. 'emporal resolution obtains the spatial and spectral data at a certain time interval. Aor e=ample "RS.2&/2B revisits the same at ?? days, where as #an revisits at 3 days. !"# is the collection of algorithm processed by the computer system to enhance the -uality of the raw data for further interpretation and data e=traction 'he steps involved in !"# are "mage restoration, statistical analysis, "mage enhancement and "mage classification "mage restoration includes geometric, radiometric J noise removal Meometric errors in the image is due to sensor, scanning system, motion of the platform, curvature and rotation of the earth Radiometric distortion is corrected by applying sun elevation and earth and sun distance corrections Resampling method is used to remove the noise from the images "mage enhancement techni-ue is used to improve the -uality of the image "mage enhancement techni-ue includes histogram, histogram e-uali:ation, linear J

nonlinear contrast stretch, spatial filtering, band ratioing and F!$"

Kistogram is the graph showing the number of pi=els in an image at Kistogram e-uali:ation is the techni-ue by which the dynamic an image is increased Kistogram manipulation improves appearance of the image Spatial filtering is used to enhance the appearance of the image to valuable and detailed information

different intensity

range of the histogram of

derive

the

Band rationing is used indentify the various earth features such as healthy vegetation, soil, water F!$" is used to assess analy:e and estimate green vegetation, crop yield etc F!$" is calculate as F"R.R / F"RSR F!$" values are represented as a ratio ranging from .2 to S2. represents water, :ero represents bare soil Hlassification is the process of automatic categori:ation of all pi=els different land cover classes or themes. Broadly there are two approaches of classification, unsupervised, and supervised. 'he basic steps involved in supervised classification are defining the training sites, feature selection, selection of classification algorithm, post classification, smoothening, accuracy assessment, final classified image and statistical report generation in an image into Fegative values

$isual "mage interpretation is the art and science of e=amining image to identify the ob,ects and evaluates their significance. 'he identification of the features is based on the shape, si:e, pattern, te=ture, shadow and association &erial photographs give Bird8s 0ye view of a large area enabling to view earth surface features which cannot be possible to obtain through ground observation &erial photographs are categori:ed as vertical, low obli-ue, high obli-ue, and stereo D! 'he photographs are ta/en generally with some percentage of overlaps for stereoscopic coverage Menerally the forward overlaps is about 647 and lateral overlap is about ?4.D47 'here are two types of the films are used in aerial photographs, panchromatic, and color

'he scale of the photo can be calculated by formula s>f / K where s is photo scale and f is the focal length of camera and K is flying height 'he various factors that affects the -uality of the aerial photographs are cloud, ha:e, sun angle, and snow, distortion due to relief, tilt, and radial Alights are usually scheduled between 24 am to ? pm to have low wind, clear s/y with ma=imum illumination and minimum shadow to obtain clear weather picture 'he interpretation /eys for aerial photos are shape, si:e, pattern, te=ture, shadow, and association &dvanced remote sensing technology includes, Kyper spectral, thermal, microwave and idar &dvantage of Remote sensing are Synoptic view, Repetitive coverage, continuous ac-uisition of data, Hoverage of inaccessible areas, Up.to.date information, accurate and reliable data, Vuantifiable data ,Few information, 1ulti disciplinary applications, 'ime, manpower saving, Vuic/ assessment of resources, Serves as a large archive of historical data

A!estion 'e6 S3ort A!estion 2. 0=plain (ien8s displacement law. ?. 'he radiation reflected as a function of wavelength called the or falseW D. Keavy sediments in water transmit the radiation. 'rue or falseW 5. (hy water loo/s greenW 3. (hat is ResolutionW 6. Fame different types of resolution E. ist the steps involved in !"# B. (hat is supervised classificationW @. (hat is unsupervised classificationW 24. ist down basic elements of image interpretation 22. Fame the advanced remote sensing technology. 2?. (hat are the basic "mage "nterpretation /eys for aerial photographsW spectral signature is true

2D. Kow many types of films are used to ta/e photographW 25. Radar is passive sensor systemW 'rue or Aalse 23. Radar operates in which part of electromagnetic regionW 26. (hat is the microwave regionW 2E. $isible and infrared radiation can penetrate through cloud, ha:e and dust. 'rue or falseW 2B. ist down the application of microwave remote sensing 2@. (hich remote sensing system allows the study in all weather conditionW &nd whyW ?4. (hat is full form of idar and R&!&RW ?2. (hat are the three basic types of idarW S3ort A!estions 2. !escribe why dust cloud, and fog loo/ white and s/y loo/s blueW ?. !efine the following a. Spectral Signature b. 'ransmission c. &bsorption d. Reflection e. Scattering D. (hat are the various factors effecting the reflectance of various earth featureW 5. !efine the following with e=amples a. Spatial resolution b. Spectral resolution c. Radiometric resolution d. 'emporal resolution 3. !efine the following with e=amples a. Radiometric correction b. Meometric correction c. "mage restoration d. Foise removal e. "mage enhancement 6. !efine the following with e=amples

a.

ow pass filter

b. Kigh pass Ailet c. Supervised classification d. Unsupervised classification E. (hat is "mage classificationW B. (hat is the linear and nonlinear contrast stretchW @. (hat are the basic elements of "mage "nterpretations /eys for aerial photographsW 24. (hy Stereo D! is usedW (hat is difference between normal and 22. (hy stereoscopic coverage is ta/en with some percentage of 2?. 0=plain about photo scale of aerial photo with formula. 2D. Kow many types of microwave remote sensingW 0=plain Long A!estion 2. !raw a curve for spectral distribution of energy by a blac/ body ?. Solar rays stri/e the land and ocean and atmosphere targets the incoming radiation go through various modes of energy interaction, mechanismW 0=plain with diagram D. !escribe the energy interaction mechanism for glass window, soil and roof and e=plain how the color of this ob,ects effectW 5. 0=plain the about a typical spectral reflectance curve of different features wet soil, dry soil, turbid water, clean water and healthy vegetation, stressed vegetation and roc/ with graphs. 3. (hat is digital image processingW (hy this techni-ue is usedW 0=plain the steps involve in !"# 6. (hat is histogramW 0=plain in detail with various techni-ues E. 0=plain the various enhancement techni-ues to improve the -uality of the imageW B. (hat is the purpose of filteringW @. (hat are advantages and disadvantages of linear, non linear contrast stretchW 24. (hat is the difference between supervised and unsupervised classificationW 22. (hat is F!$"W 0=plain with formulaW (hy it is usedW 2?. (hat are the difference between satellite images and &erial photographsW what are the different interaction D! photoW overlapsW 0=plain.

2D. (hat are advantages of Remote SensingW 25. !efine the following elements used in visual interpretation of aerial and satellite data with e=amples a. Shape b. Si:e c. #attern d. 'e=ture 23. 0=plain benefits and applications of a. Satellite images b. &erial #hotographs 26. !iscuss about the aerial photographsW photographsW 2E. (hat is difference between Kigh and low angle obli-ue photographsW 2B. (hat factors affect the -ualities of aerial photographW 0=plain. 2@. (hat is microwave remote sensing systemW (hy it is usedW ?4. 0=plain about Radar SensorW ?2. 0=plain the fre-uency range of 1icrowave remote sensing system. 0=plain in details about types of aerial e Shadow f &ssociation

Anda mungkin juga menyukai