GI25403 Digital Image Processing – Semester IV – Geo – R-2025

Subject Code & Name: GI25403 – Digital Image Processing

Regulation: R-2025

Semester: IV (Fourth Semester)

Branch: B.E. Geoinformatics Engineering (Geo)

Credits / L-T-P: 5 Credits | L-T-P: 3-0-4

Course Objectives

• To provide comprehensive knowledge of remote sensing image formation, preprocessing, and digital image processing techniques.

Full Unit-wise Syllabus

Unit I – Image Formation and preprocessing

Image representation and properties – Visual perception and image formation – Sampling and quantization – Image acquisition and sensor characteristics – Histograms and scattergrams – Univariate and multivariate statistics – Geometry and radiometry – Noise models – Atmospheric, radiometric and geometric corrections – Interpolation and resampling techniques.

Practical: 1. Data download, visualization, histogram and scattergram generation, and image statistics computation. 2. Image resampling, reprojection, and geometric correction.

Unit II – Google Earth Engine (GEE) Programming

Overview of Google Earth Engine – JavaScript fundamentals for GEE – Data types, arrays, objects – Image and Image Collection handling – Feature Collection creation and filtering – Higher-order functions – Band operations – Image clipping, mosaicking, and visualization – Spectral indices (NDVI, NDBI, NDWI).

Practical: 1. Creating and manipulating ImageCollections; band combinations and index generation in GEE. 2. Image transformation, filtering, and export of processed outputs in GEE.

Unit III – Image Enhancement and Transformation

Point, local and regional operations – Contrast stretching and histogram equalization – Spatial filtering and edge detection – Frequency domain processing (DFT/FFT concepts) – Band ratioing – Principal Component Analysis (PCA) – Multi-image fusion – Image transformations and visualization techniques.

Practical: 1. Image enhancement using contrast stretching, histogram equalization, and filtering. 2. Band ratioing, PCA implementation, and feature extraction for vegetation/urban analysis.

Unit IV – Image Classification and Accuracy Assessment

Spectral signatures and training datasets – Supervised classification (Minimum Distance, Parallelepiped, Maximum Likelihood) – Unsupervised classification (K-means, ISODATA) – Neural network concepts – Change detection and time-series analysis – Accuracy assessment: Error matrix, Kappa statistics, RMSE, ROC – Evaluation of classification performance.

Practical: 1. Supervised and unsupervised classification using GEE. 2. Accuracy assessment using confusion matrix, Kappa coefficient, and change detection analysis.

Tasks

• Perform image preprocessing, enhancement, and spectral index generation using GEE to analyze land cover characteristics for application in environmental monitoring and resource assessment.
• Develop and apply supervised and unsupervised classification models on satellite imagery with feature extraction techniques for application in urban growth analysis and land use planning.
• Conduct accuracy assessment and change detection using multi-temporal satellite data for application in climate impact studies and sustainable development monitoring.

Course Outcomes (COs)

• CO1: Explain remote sensing image formation processes and preprocessing techniques.
• CO2: Apply Google Earth Engine programming concepts for image manipulation and spectral analysis.
• CO3: Implement image enhancement, transformation, and classification techniques for information extraction.
• CO4: Evaluate classification results and select appropriate image processing methods based on accuracy metrics and application needs.

Assessment Pattern (Quick Note)

• Weightage: Continuous Assessment 50% | End Semester Examinations 50% Methodology for Continuous Assessment : Quiz (5%), Project (15%), Assignment Programs (25%), Practical (25%), Internal Examinations (30%)

Source: Official Anna University – B.E. Geoinformatics Engineering R-2025 Syllabus
Last Updated: October 2026