Classification of Digital Image Processing Techniques (Image Rectification, Image Enhancement, Image Classification) (Especially for GATE-Geospatial 2022)

Get unlimited access to the best preparation resource for CTET/Paper-1 : get questions, notes, tests, video lectures and more- for all subjects of CTET/Paper-1.

Examrace Books on Mapping, GIS, and Remote Sensing prepares you throughly for a wide range of practical applications.

Digital Image Processing Services|Shukka Methods - GIS and …
  • ERDAS Imagine 8.4 version was used for the digital image processing of this study.
  • Image processing technique used can be broadly grouped into three categories namely –
    • Image Rectification
    • Image Enhancement
    • Image Classification

Image Rectification

  • This operation also termed as image restoration. This is because satellite data is prepared in this step for further processing and analysis and hence it is generally called data preparation or pre-processing.
  • These operations are intended to eliminate or correct the distortions or errors caused due to geometric distortions, radiometric distortions, and presence of noise in the data. Etc. The standard products available from National Remote Sensing Agency are pre-processed to the extent of radiometric and geometric corrections.
  • For these images obtained, geo-referencing and a suitable projection has to be done using the corresponding rectified toposheet.
  • A scanned toposheet (on 1: 250,000 scale) is taken and its graticule intervals are used as ground control points (GCPs) for rectification of soft copy toposheet. These rectified softcopy toposheets were used for rectification or geo-referencing of the satellite image.
  • GCPs are identified on the toposheet and same point is located on the satellite image and marked as a registration mark. Likewise many GCPs or registration marks were marked by the satellite image, and then the image was resampled.
  • Following this operation a suitable projection i.e.. in this case polyconic projection parameters are added and the image is resampled. Thus, during this operation the satellite data is geo-referenced and projected.

Image Enhancement

  • Image enhancements are done to improve the interpretability of the satellite image. Different techniques of image enhancements were used, of which linear stretch, standard deviation stretch, histogram equalization and stretching by using break points are the prominent methods used for image stretching.
  • Different techniques give different types of result and one single technique is good for all types of scenes or interpretations. The image characteristics and the land features at a particular moment of time i.e.. season or a different location have a predominant role in giving different results to different types of stretches.
  • Hence, a trial and error method has to be used for getting the best results of the seen under investigation. Normalized Differential Vegetation Index (NDVI) , Ratio Vegetation Index (RVI) , Fourier Transformations, Edge Enhancements, Principle Component Analysis (PCA) are some of the other advanced techniques which enhances the output of the image and provide better feature exhibitance i.e.. increases the visual distinction between features contained in an image.
  • These enhanced images are also used for further processing like forgiving training sets for supervised classification etc.

Image Classification

  • Image classification is the process of assigning land cover classes to pixels. Image classification operation is essentially met to substitute the visual analysis of the remotely sensed satellite data and quantitative assessment.
  • The classification of the remotely sensed data can be carried out either without a priori knowledge about the features present in the scene (unsupervised classification) or with a priori knowledge about the terrain features (supervised classification) .
  • In our study both of these techniques were used. The digital image classification was found to be inferior to the visual image interpretation in a sense that only the tone of the image i.e.. DN values is a sole criteria for classification of the digital image.
  • Whereas the visual image interpretation considers the nine elements of image interpretation, which had been discussed earlier.
  • Supervised classification. In supervised classification the user or image analyst “supervises” the pixel classification process. The user specifies the various pixels values or spectral signatures that should be associated with each class
  • Unsupervised classification is where the outcomes (groupings of pixels with common characteristics) are based on the software analysis of an image without the user providing sample classes. The computer uses techniques to determine which pixels are related and groups them into classes

Developed by: