Unit - 2 Data Products and Image Analysis

This section focuses on the types of data products generated from remote sensing imagery, including multispectral, hyperspectral, and radar data. It discusses the processes involved in image analysis, such as classification, change detection, and feature extraction, which help interpret and extract meaningful information from raw data. Techniques like supervised and unsupervised classification are highlighted, along with the importance of image resolution and pre-processing methods. The applications of these analyses span various fields, including agriculture, urban planning, and environmental monitoring.

1. UNIT 2 – Data Products and Image Analysis A data product is any tool or application that processes data and generates insights. These insights are aimed at helping businesses make better decisions for the future. Stored data can then be sold or consumed by users internally, or by customer organizations which then process the data as needed. There are two types of data used in geoinformatics. One is raster data and the other is vector data. When we talk of remote sensing data, we always mean raster data. In its simplest form, a raster data means data consisting of a matrix of cells or pixels organised into rows and columns (or a grid) where each cell contains a value representing information, such as reflected electromagnetic radiation (EMR), temperature, or height values. Raster data products include digital aerial photographs, imagery from satellites, digital pictures, or even scanned maps.

2. The data from various sensors are presented in a form and format with specified radiometric and geometric accuracy which can be readily used by various application scientists for specific themes of their interest. Remote sensing data can be procured by a number of users for various applications and information extraction, in the form of a ‘data product’. This may be in the form of photographic output for visual processing or in a digital format amenable for further computer processing. There are varieties of remote sensing data which are acquired by different sensors and satellites. Before reaching to users, the data undergo some processing steps. Requirements of users may vary depending upon their interests and project objectives, hence there are various remote sensing data providers/suppliers which prepare variety of data products in different formats.

3. INDEX NUMBERS FOR DATA PRODUCTS All remote sensing data products carry a specific index number. This index number is generated using the satellite path which runs from North Pole to South Pole of Earth. This pole to pole coverage on the Earth for each pass of the satellite is given a specific number, called path number or track number. In case of coverage of India, path numbers range from 88 to 116 from west to east for LISS sensor of IRS. East-West lines (called row) cut across these lines giving a grid like structure. This is how a series of columns and rows are created. Just like each column is given a specific number, each row is also given a specific number. India is covered by rows 45 to 70 from north to south.

4. TYPES OF DATA PRODUCTS The types of remote sensing data products depend on: level of processing output media/scale and area of coverage.

5. Standard Data Products These products are generated using information received directly from the satellites and by applying necessary radiometric and geometric corrections. These products are of various types: Path-Row Based Standard Products: Actually, the data are recorded by the sensors along the track/path of the satellite covering a specific width (across the satellite track/path). The user specifies path and row of the scene, sensor, sub-scene (if any), number of scenes, date of pass of the satellite, band numbers/band combination (for photographic products) and the product code (specified by the data provider) as necessary information to procure the desired products from the provider, for example in India, the National Data Centre (NDC) of National Remote Sensing Centre (NRSC), Hyderabad. Shift Along Track (SAT) Products : In case a user’s area of interest falls between two successive scenes of the same path (one below the other), the data can be supplied by sliding the scene in along-the-track direction. These products are called Shift-Along-Track Products. In this case, the percentage (10% to 90% in multiples of 10%) of shift has to be specified by the user in addition to the other necessary information about both the scenes.

6. Quadrant Products: In IRS series of satellites, LISS III full scene (path/ row based products) data has been divided into 4 nominal quadrants designated by letter A, B, C & D. Each of these quadrants is one full scene of PAN data (B & W data), and is further divided into 9 quadrants. These are given numbers from 1 to 9 which are nothing but PAN subscenes. While placing a request for these products, users need to specify the quadrant number in addition to the details specified for path/row based products. Georeferenced Products: Georeferencing is the process of transforming the remote sensing data or a map to a coordinate and projection system. This means, all the objects/elements in remote sensing data and the map get specific geographic location in terms of longitudes and latitudes. Thus, georeferenced products are north-east oriented products. Basic Stereo Products: A stereo-pair comprises two images of the same area, acquired on different dates or same day; from different angles. These products are used for generating digital elevation maps or 3D visualization of the study area. Cartosat-1 mission provides along track stereo images.

7. Value Added Products When standard products are processed according to specifications and requirements of users, these products get converted into value added products. These products are basically of four types viz. Geocoded products Merged products Ortho product Template registered products.

8. Geocoded Products: These products are also known as georeferenced products (mentioned in one of the preceding paragraphs). Geocoding can be performed with reference to both topographical maps and floating points. Thus, geocoded products are of two types viz. topographical map based products, and floating point (not based on topographical maps) based products. Merged Products: Remote sensing satellites record multiband/multispectral (MSS) data and single-band/panchromatic data. Panchromatic/ mono-band data contains higher spatial resolution and multi-band data have comparatively low in spatial resolution. When MSS data area is merged with panchromatic data, we get MSS data with spatial resolution of panchromatic data. Data can be merged only when both MSS and PAN data have been registered with the same georeferencing system.

9. Ortho Products: Geometrically corrected products, with corrections for displacement caused by tilt and relief, are called Ortho Products. Therefore, ortho images show ground objects in their true planimetric positions, like the positions of objects in a map. The basic inputs required for ortho-image generation are: ( i ) Digital Elevation Model, (ii) Ground Control Points, (iii) Satellite ephemeris (orbit and altitude information), and (iv) Radiometrically corrected satellite data. Template Registered Products: In specific cases, such as the study of crops and their monitoring, data of the same area having similar geometric fidelity and temporally registered are required. Such data sets are called template registered products

10. Derived Products Information extracted from remote sensing data is also provided as useful products by data providers. For example, vegetation index map or sea surface temperature profiles extracted from various remote sensing data products are called derived products. These derived products are generated by further processing /analysing the data, and are readily usable by the user. Based on output media, data products are available on both photographic as well as digital media. Photographic products can be supplied as films or prints. Output products can vary from 1:1 million to 1:50000 or even to 1:25000. The following table gives detailed information on the type of product available for various sensors of IRS series of satellites. The table is reproduced from the content of NRSC website (www.nrsc.gov.in/products.html).

11. Products Based on Output Media/Scale NRSC provides remote sensing products in two types of media: photographic media digital media. Photographic products are supplied on film or paper prints. The scale of these photographic products can range from 1:1M to 1:5000.

12. Based on level of processing-o/p The Levels of Processing Model proposes three distinct levels of processing: shallow, intermediate, and deep. The levels of processing model (Craik & Lockhart, 1972) focuses on the depth of processing involved in memory, and predicts the deeper information is processed, the longer a memory trace will last. Memory is just a by-product of the depth of information processing, and there is no clear distinction between short-term and long-term memory. Levels of processing: The idea that the way information is encoded affects how well it is remembered. The deeper the level of processing, the easier the information is to recall.

13. Shallow Processing 1 . Structural processing (appearance) which is when we encode only the physical qualities of something. E.g. the typeface of a word or how the letters look. 2 . Phonemic processing – which is when we encode its sound. Deep Processing Semantic processing, which happens when we encode the meaning of a word and relate it to similar words with similar meaning. Deep processing involves elaboration rehearsal which involves a more meaningful analysis (e.g. images, thinking, associations, etc.) of information and leads to better recall.

14. Scale The process of representing geographic features on a sheet of paper involves the reduction of these features. The ratio between the reduced depiction on the map and the geographical features in the real world is known as the map scale, that is the ratio of the distance between two points on the map and the corresponding distance on the groun d. Fractional scale: If two points are 1 km apart in the field, they may be represented on the map as separated by some fraction of that distance, say 1 cm. In this instance, the scale is 1 cm to a kilometer . In India, commonly used fractional map scales are 1:1,00,000,00; 1 :250,000, 1 :50,000; 1 :25,000 and 1 :10,000. The method of representing this type of scale is called Representation Fraction (RF) method.

15. Graphic scale : This scale is a line printed on the map and divided into units that are equivalent to some distance such as 1 km or 1 mile. The measured ground distance appears directly on the map in graphical representation. Verbal scale: This is an expression in common speech, such as, "four centimeters to the kilometer ", "an inch to a mile". This common method of expressing a scale has the advantage of being easily understood by most map users. The ratio and map scale are inversely proportional. The small scale maps depict large tracts of lands (such as continents or countries) usually with a limited level of detail and a simple symbology. Large scale maps can depict small areas (such as cities) with a richness of detail and a complex symbology.

16. Area/Coverage A data model for storing geographic features. A coverage stores a set of thematically associated data considered to be a unit. It usually represents a single layer, such as soils, streams, roads, or land use.

17. Data Availability GIS integrates many different kinds of data layers using spatial location. Most data has a geographic component. GIS data includes imagery, features, and basemaps linked to spreadsheets and tables. Data Ordering USGS is a primary source of geographic information system (GIS) data. Our data and information is presented in spatial and geographic formats, including The National Map, Earth Explorer, GloVIS , LandsatLook , and much more. Data Price Depending on the type and quality of services, as well as the size of your project, GIS software can cost anywhere from $600 to $17,000 per year. The most expensive part of a Geographic Information System is the collection and processing of the data. The cost factor underscores the vital importance of creating or accessing the reliable spatial data. Every detail seen or observed may be convertible to a digital form and become an intelligent input to a GIS.

18. Image interpretation The study of land use, land cover or land form from the air has been a focus of interest since the early days of aerial photography. It has been gaining momentum again with the availability of new remote sensing techniques using aircraft and spacecraft as platforms with the capacity for operating outside the visible part of the electromagnetic spectrum through microwaves (radar) and thermal radiation . Thus photo interpretation is now complemented by the interpretation of other kinds of imagery which are different from photography. The limitation of 'photo-interpretation’ has now been changed to broad spectrum of 'image interpretation.

19. Types of Pictoral Data Products The remote sensing data products are available to the users In the form of (a) photographic products such as paper prints, film negatives, diapositives of black and white, and false colour composite (FCC) on a variety of scales (b) digital form as computer compatible tape (CCTs) after necessary corrections S atellite data products can be classified into different types based on satellite and sensor, level of preprocessing and the media. Data products acquired for the specific period can be generated if the data pertaining to the period of interest is available in archives. Depending upon the corrections applied and on the level of processing, data products can be classified as : raw data, partially corrected products, standard products, geocoded products, and precision products.

20. The raw data is radiometrically and geometrically uncorrected data with ancillary information. Standard products are radiometrically and geometrically corrected for systematic errors. Geocoded products are systematically and geometrically corrected products. The systematic corrections are based on the standard survey of India toposheet and rotation of pixels to align to true north and resampled to standard square pixel. Precision products are radiometric and geometric corrections refined with the use of ground control points to achieve greater locational accuracy.

21. Data products can be broadly classified into two types depending upon the output media, as photographic and digital. Photographic products can either be in black and white, or colour. Further they could be either film or paper products, and in films it is possible to have either positive film or negative film. The sizes of photographic products can vary depending on the enlargement needed, and this is specified as 1X, 2X, 4X and so on. The size of film recorders is generally 240 mm and this is the basic master output from which further products are generated. When we say colour photographic products, it generally means false colour composites (plate 3). FCCs are generated by combining the data contained in 3 different spectral bands into one image by assigning blue, green and red colours to the data in three spectral bands respectively during the exposure of a colour negative.

22. Types of product media Different types of photographic products supplied by National Remote Sensing Agency (NRSA) data centre, Govt., of India (NDC) are: Standard B&W and FCC, films. Standard products are available in colour, and black and white in the form of 240 mm films, either as negatives or positives. The photographic products contain certain details annotated on the margins. These are useful for identifying the scene , sensor, date of pass, processing level, band combination, and so on .

23. Process of Image Interpretation Image interpretation or analysis is defined as the "act of examining images for the purpose of identifying objects and Judging their Significance". Interpreters study remotely sensed data and attempt through logical process to detect, identify, classify, measure, and evaluate the significances of physical and cultural objects, their patterns and spatial relationship. The sequence begins with the detection and Identification of Images and later by their measurements. The image interpretation has various aspects which have overlapping functions. These aspects are detection, recognition and identification, analysis, classification, and idealisation. The detection is a process of 'picking out' an object or element from photo or image through interpretation techniques. It may be detection of point or line or a location, such as, agricultural field and a small settlement

24. Recognition and Identification is a process of classification or trying to distinguish an object by its characteristics or patterns which are familiar on the image. Sometimes it is also termed as photo reading of water features, streams, tanks, sands, and so on. Analysis is a process of resolving or separating a set of objects or features having a similar set of characters. In analysis, lines of separation are drawn between a group of objects and the degree of reliability of these lines can also be indicated. Classification is a process of identification and grouping of objects or features resolved by analysis. Idealization is a process of drawing ideal or standard representation from what is actually Identified and interpreted from the image or map.

25. There is no single "right" way to approach the interpretation process. The interpretation process depends upon the type of data product, interpretation equipment and a particular interpretation task undertaken. Before an interpreter undertakes the task of performing visual interpretation, two important issues should be addressed. The first is the definition of classification system or criteria to be used to separate the various categories of features occurring in the images. The second important issue is the selection of minimum mapping unit (MMU) to be applied on the image interpretation. MMU refers to the smallest size areal entity to be mapped as a discrete area.

26. Interpretation of aerial photo In case of Aerial photo the characteristics of the image of earth's surface relates to tone, colour, shadow, texture, shape and size and other associations between features and their surroundings. In addition certain ground characteristics such as land form, vegetation, land use, drainage, erosion and lineaments help in interpreting the image. The following procedure can be adopted for both aerial photo and remot sensed imagery: Preliminary examination Detailed examination Interpretation Compilation

27. Preliminary stage In this stage it is necessary for the interpreter to become familiar with the site and its surrounding area. At this examination, easily Identifiable images are correlated with ground features such as roads, rivers etc. The principle of working from "whole to part" is also followed here from overall viewing of terrain using aerial photos. This is achieved by making a photo-mosaic of the aerial photos (by joining successive and overlapping photos and removing common area). Detailed examination The next stage, the detailed examination is to examine overlapping photos under a stereoscope to see the third dimension. viz: topography of the terrain and its variation. Overlays are then prepared by tracing from the photos or tracing paper showing features, which have been identified.

28. Interpretation stage Many times the stages of detailed examination and interpretation can be combined and can be carried out side by side. Some of the general and common features to be noted from aerial photos, which are common to most of the project are as follows: Topography - hills, valleys, flood plain, lakes, steep and gentle slopes, depression, quarries, embankments, costal' features etc Drainage and drainage patterns-springs, seepage, major rivers, streams, channels, marshy lands, ponds, lakes etc. General geology viz: soil boundaries, soil types Exiting hazards like abandoned mining sites, trenches, buried foundation etc. Site history such as earlier use of presently abandoned buildings, waste land, removal of vegetation etc.

29. Compilation Stage Consists of making maps from aerial photos and showing features of significance as identified by interpretation. Depending upon the topography and tilt of the photos, any existing photogrammetric methods such as rectification may be employed to get a final map with symbols depicting the features of the terrain and other information of the terrain such as land use etc. Three-dimensional interpretation Method The basic requirement is a set of overlapping photographs (as per specification) and a stereoscope to obtain a three-dimensional view of the terrain. Stereoscopic depth perception Stereoscopic vision is binocular vision when the left and right eye are focused on a certain point, the optical axes of the two eyes converge on that point intersecting at an angle called the parallactic angle. The nearer the object the greater the parallactic angle and vice versa.

30. The distances between the two eyes are called the eye base. (be)

31. When the eyes are focussed at point A, the optical axes converge forming parallactic angle ~a’ Similarly, when sighting at B the optical axes converge at B forming parallactic angle ~b’ The brain associates the distance DA and DB with corresponding parallactic angle ~a and ~b' The depth between the object A and B is DB - DA and is perceived as the difference between these two parallactic angles. Thus the basic concept of creating three dimensional or stereoscopic impression of objects is by viewing identical images of objects placed at different distances. The same principle is used for creation of three-dimensional effect by viewing aerial photographs taken from two different exposure stations.

32. The two overlapping photos are laid on a table and viewed in such a way that left eye see only left photo and right eye sees only right photo. The brain judges the heights (3-~ model) of each overlapping identical objects , by associating the depth with their corresponding parallactic angles. The effect is the viewing of a virtual three-dimensional model , which are associated with the continuous changing parallactic angles. The three-dimensional model thus formed is called a stereoscopic model and the overlapping pair of photos is called a stereo pair.

33. Key Elements of Visual Image Interpretation Knowledge of spatial database is important for planning and management activities and it is considered an essential for modeling and understanding the earth as a system. As mentioned frequently, the source of data collection for the creation of such a spatial database is remote sensing system including photographic sensing systems. If the creation of spatial database is based on the remote sensing data analysis, then the visual image interpretation (VIP) techniques and/or digital image processing (DIP) techniques can be employed. The maps information layers namely land use/ land cover, hydrogeomorphology, soils, geology, drainage network, road network, geomorphological units/landforms and other related maps can be derived from satellite data, whereas layers of information like slope, contour, and watershed boundaries are also considered spatial database and can be derived from other existing maps/toposheets.

34. Visual interpretation elements A systematic study .of aerial photographs and satellite imageries usually involves several characteristics of features shown on an image. The following characteristics (elements) are called fundamental picture elements. These elements aid visual interpretation process of aerial photos and/or satellite imagery. ( i ) Tone: Ground objects of different colour reflect the incident radiation differently depending upon the incident wave length, physical and chemical constituents of the objects. The imagery as recorded in remote sensing is in different shades or tones. example, ploughed and cultivated lands record differently from fallow fields. Tone is expressed qualitatively as light, medium and dark. one, therefore, refers to the colour or reflective brightness. Tone along with texture and shadow (as described below) help in Interpretation and hence is a very important key. Differences in moisture content of the soil or rock result in differences in tone.

35. In a black and white photograph dark tone indicates dark bodies, namely, greater moisture contents and grey or white tone reflect the dry soil. The aerial photos with good contrast bring out tonal differences and hence help in better interpretation. Tonal contrast can be enhanced by use of high contrast film, high contrast paper or by specialized image processing techniques such as 'Dodging' or 'Digital Enhancement'. (ii)Texture: Texture is an expression of roughness or smoothness as exhibited by the imagery. It is the rate of change of tonal values. Mathematically it is given as dD /dx where D is the Density and 'x' the distance measured from one arbitrary starting point, and can be measured numerically by the use of microdensitometer. Texture is dependent upon (a) photographic tone (b) shape, (c) size, (d) pattern and scale of the imagery.

36. The texture is a combination of several image characteristics such as tone, shadow, size, shape and pattern etc., and is produced by a mixture of features too small to be seen individually because the texture by definition is the frequency of tonal changes. (iii) Association The relation of a particular feature to its surroundings is an important key to interpretation. Sometimes a single feature by itself may not be distinctive enough to permit its identification. (iv) Shape : Some ground features have typical shapes due to the structure or topography. For example air fields and football stadium easily can be interpreted because of their finite ground shapes and geometry whereas volcanic covers, sand, river terraces, cliffs, gullies can be identified because of their characteristics shape controlled by geology and topography. (vi) Size: The size of an image also helps for its identification whether it is relative or absolute. Sometimes the measurements of height (as by using parallax bar) also gives clues to the nature of the object.

37. (vii) Shadows: Shadows cast by objects are sometimes important clues to their identification and Interpretation. (viii) Site factor or Topographic Location: Relative elevation or specific location of objects can be helpful to identify certain features. For example, sudden appearance or disappearance of vegetation is a good clue to the underlying soil type or drainage conditions. (ix)Pattern: Pattern is the orderly spatial arrangement of geological topographic or vegetation features. This spatial arrangement may be two-dimensional (plan view) or 3-dimensional (space). Geological pattern may be linear or curved. Linear pattern are formed of a very large number of continuous or discontinuous short ticks which when viewed by eye appear to be continuous lines. vegetation pattern may be of the 'Block' type or 'Alignment' type. The 'Alignment' type may be further subdivided into the Linear, Parallel and Curved type.

38. Example of topographic pattern are the typical drainage patterns (controlled and uncontrolled type). The uncontrolled type are those, which are purely governed by topography, i.e., the slops whereas the controlled type are those, which are governed by the underlying geological formations. The well-known drainage patterns are: ( i ) Dendritic, (ii) Trellis, (iii) Annualsr , (iv) Radial, (v) Rectangular, (vi) Parallel Type, (vii) Braided, (viii) Anastomotic, (ix) Asymmetrical, (x) Collinear. Drainage patterns take an important place amongst the various interpretation elements, used as criteria for identification on geological and geomorphological phenomena.

39. Interpretation key Keys that provide useful reference of refresher materials and valuable training aids for novice interpreters are called image interpretation keys. These image interpretation keys are very much useful for the interpretation of complex imageries or photographs. These keys provide a method of organising the information in a consistent manner and provide guidance about the correct identification of features or conditions on the images. Ideally, it consists of two basic parts' ( i ) a collection of annotated or captioned images illustrative of the features or conditions to be identified, and (ii) a graphic or word description that sets forth in some systematic fashion the image recognition characteristics of those features or conditions. There are two types of keys: selective key and elimination key. A selective key is also called reference key which contains numerous example images with supporting text. The interpreter selects one example Image that most nearly resembles the feature or condition found on the Image under study.

40. An elimination key is arranged so that the interpretation proceeds step by step from the general to the specific, and leads to the elimination of all features or conditions except the one being identified. Elimination keys are also called dichotomous keys where the interpreter makes a series of choices between two alternatives and progressively eliminates all but one possible answer. Elimination keys are also called dichotomous keys where the interpreter makes a series of choices between two alternatives and progressively eliminates all but one possible answer. This key gives more positive answers than selective keys. But the elimination key sometimes gives more erroneous answers if the interpreter is forced to make uncertain choice between any two unfamiliar image characteristics The elimination key is used and successfully employed for agricultural studies, and forestry applications. An important point is that, these keys are normally developed and used on a region-by-region and season-by-season basis, that is these keys are dependent upon location and season.

41. Digital image processing If the data are in digital mode, the remote sensing data can be analysed using digital image processing techniques and such a database can be used in raster GIS. In applications where spectral patterns are more informative, it is preferable to analyse digital data rather than pictorial data. Basic Character of Digital Image These values are simply positive integers that result from quantizing the original electrical signal from the sensor in to positive integer values using a process called analog -to-digital (A to 0) signal conversion.

42. Basic character of digital image data (a) original 200 x 200 digital image (b) enlargement showing 20 x 20 of pixels (c) 10 x 10 enlargment (d) digital numbers corresponding to radiance of each pixel The original electrical signal from the sensor is a continuous analog signal shown by the continuous line plotted in the figure. This continuous signal is sampled at a set time interval ( ~ T) and recorded numerically at each sample point (a , b, ... .. , i , k) . The sampling rate for a particular signal is determined by the least to twice the highest frequency presents in the original signal in order to adequately represent the variation in the signal .

43. T he incoming sensor signal in terms of an electrical voltage value ranging between 0 and 2 V. The DN output values are integers ranging from 0 to 255. Accordingly, a sampled voltage of 0.46 recorded by the sensor would be recorded as a DN of 59 . Typically, the DNs constituting a digital image are recorded over such numerical ranges as O to 255, 0 to 511 ,0 to 1023, or higher. A digital image is defined as a matrix of digital numbers (DNs) . Each digital number is the output of the process of analog to digital conversion . The image display subsystem carries out the conversion operation, that of taking a digital quantity of an individual pixel value from an image.

44. The surface of the ground divided into a number of parcels. Each parcel of land can be represented as a pixel (picture element) on the image and each pixel is occupied by a digital number and is called pixel value . This pixel value or digital number shows the radiometric resolution of remote sensing data.

45. Satellite remote sensing data in general and digital data in particular have been used as basic inputs for the inventory and mapping of natural resources of the earth surface like agriculture, soils, forestry, and geology. Space borne remote sensing data suffer from a variety of radiometric and geometric errors caused by satellite motion, the sensor system, earth's rotation and so on. These distortions would diminish the accuracy of the information extracted and reduce the utility of the data. In order to update and compile maps with high accuracy, the satellite digital data have to be manipulated using image processing techniques. The central idea behind digital image processing is that, the digital Image is fed into a computer, one pixel at a time. The computer is programmed to insert these data Into an equation or a series of equations, and then store the results of the computation for each pixel. These results are called look-up-table (LUT) values.

46. Virtually, all the procedures may be grouped into one or more of the following broad types of operations. ( i ) Preprocessing (iii) Image enhancement (v) Image transformation (ii) Image Registration (iv) Image filtering (vi) Image classification

47. Preprocessing The correction of deficiencies and removal of flaws present in the data through some methods are termed as pre-processing methods. This correction model involves the initial processing of raw image data to correct geometric distortions, to calibrate the data radiometrically and to eliminate the noise present in the data. All pre-processing methods are considered under three heads, namely, ( i ) geometric correction methods, (ii) radiometric correction methods, and (iii) atmospheric correction methods. Geometric Correction Methods The transformation of a remotely sensed image into a map with a scale and projection properties is called geometric correction.

48. to transform an image to match a map projection to locate points of interest on map and image to bring adjacent images into registration to overlay temporal sequences of images of the same area, perhaps acquired by different sensors to overlay images and maps within GIS, and to integrate remote sensing data with GIS. To correct sensor data, both internal and external errors must be determined. Internal errors are due to sensor effects, External errors are due to platform perturbations and scene characteristics, The errors of altitude and attitude of the space craft, and scan skewing are due to the platform instability. Departures of the spacecraft from nominal altitude of the landsat produce scale distortion in the sensor data.

49. (ii) Radiometric Correction Methods The primary function of remote sensing data quality evaluation is to monitor the performance of the sensors. The performance of the sensors is continuously monitored by applying radiometric correction models on digital Image data sets. The radiance measured by any given system over a given object is influenced by factors, such as, changes in scene illumination, atmospheric conditions, viewing geometry and instrument response characteristics. (iii) Atmospheric Correction Methods If the sky is clear with no scattering, then the radiance reflected from the earth surface feature in any of the region of the electromagnetic spectrum should be the same. This is the ideal case. In reality, because of the presence of haze, fog, or atmospheric scattering, there always exists some kind of unwanted signal value called bias. The bias is the amount of offset for each spectral band. Bias can be determined by regressing the visible band vs. infrared bands.

50. First of all , it is essential to identify some areas like an airport, deep homogenous non-turbid water bodies and some shadow area in the scene. The brightness values of all these features from each band are then extracted. The grey value of the visible band (for example, IRS- band 1) is plotted against corresponding values at the same pixel location in the infrared band (IRS-band 4). The plot will result in a scatter diagram. (iv) Image enhancement The main aim of digital enhancement is to amplify these slight differences for better clarity of the image scene. This means digital enhancement increases the separability (contrast) between the interested classes or features. The digital image enhancement may be defined as some mathematical operations that are to be applied to digital remote sensing input data to improve the visual appearance of an image for better interpretability or subsequent digital analysis. Many digital enhancement algorithms are available. They are contrast stretching enhancement, ratioing , linear combinations , principal component analysis, and spatial filtering.

51. t he enhancement techniques are categorised as point operations and local operations. Point operations modify the values of each pixel in an image data set independently, whereas local operations modify the values of each pixel in the context of the pixel values surrounding it. Point operations include contrast enhancement and band combinations, but spatial filtering is an example of local operations. Contrast Enhancement The sensors mounted on board the aircraft and satellites have to be capable of detecting upwelling radiance levels ranging from low (oceans) to very high (snow or ice). The sensitivity of remote sensing detectors was designed to record a wide range of terrain brightness from black asphalt and basaltic rocks to white sea ice under a wide range of lighting conditions. In general, few of the scenes have the full brightness range. To produce an image with optimum contrast ratio, it is inevitable to utilize the entire dynamic range. Digital contrast enhancement is thus of prime importance.

52. Image Classification Image classification is a procedure to automatically categorize all pixels in an image of a terrain into land cover classes. Normally, multispectral data are used to perform the classification of the spectral pattern present within the data for each pixel is used as the numerical basis for categorization. This concept is dealt under the broad subject, namely, Pattern Recognition. Spectral pattern recognition refers to the family of classification procedures that utilises this pixel-by-pixel spectral information as the basis for automated land cover classification. Spatial pattern recognition involves the categorization of image pixels on the basis of the spatial relationship with pixels surrounding them. Image classification techniques are grouped into two types, namely supervised and unsupervised.

53. Supervised Classification A supervised classification algorithm requires a training sample for each class, that is, a collection of data points known to have come from the class of interest. The classification is thus based on how "close" a point to be classified is to each training sample. The training samples are representative of the known classes of interest to the analyst. Classification methods that relay on use of training patterns are called supervised classification methods. i ) Training stage: The analyst identifies representative training areas and develops numerical descriptions of the spectral signatures of each land cover type of interest in the scene. (ii) The classification stage: Each pixel in the image data set is categorised into the land cover class it most closely resembles. If the pixel is insufficiently similar to any training data set it is usually labeled 'Unknown

54. (iii) The output stage: The results may be used in a number of different ways. Three typical forms of output products are thematic maps, tables and digital data files which become input data for GIS

55. There are a number of powerful supervised classifiers based on the statistics, which are commonly, used for various applications. A few of them are a minimum distance to means method, average distance method, parallelepiped method, maximum likelihood method, modified maximum likelihood method, Baysian's method, decision tree classification, and discriminant functions.

56. Training Dataset: A training dataset is a set of measurements (points from an image) whose category membership is known by the analyst. This set must be selected based on additional information derived from maps, field surveys, aerial photographs, and analyst's knowledge of usual spectral signatures of different cover classes. Unsupervised Classification: Unsupervised classification algorithms do not compare .points to be classified with training data. Rather, unsupervised algorithms examine a large number of unknown data vectors and divide them into classes based on properties inherent to the data themselves. Vegetation Indices: Vegetation and land use can be considered one of the most essential key elements as the variations of vegetation and land use from one area to the other act as an indicator of terrain condition.