The course aims on introducing digital image processing (IP) and computer vision (CV) concepts. It explores the methods and algorithms enabling the students to important IP/CV systems or use IP/CV software with emphasis on remote-sensing and photogrammetry applications. The course components include: digital image acquisition and sampling image enhancement in the spatial and frequency domain, Color image processing, image restoration, image segmentation, image compression and multi-source image/ data fusion.

Course outline:

  1. Introduction (1 hour)
    1. digital image processing, origins of digital image processing
    2. Fundamentals in digital image processing, components in image processing system
    3. discipline of surveying significance to civil engineering

  2. Fundamentals of digital image (2 hours)
    1. Elements of visual perception, light and electromagnetic spectrum,
    2. image sensing and acquisition, image sampling and quantization

  3. Image enhancement in spectral domain (6 hours)
    1. Feature space, multispectral ratios, and vegetation indexes
    2. Principal components, standard principal components
    3. Maximum noise fraction, tasselled cap components
    4. Contrast enhancement global transfer forms, linear stretch, non linear stretch, stretch, reference stretch, local transforms
    5. Color images, mean max stretch, normalization stretch, decorrelation stretch, color space transforms, spatial domain blending

  4. Spatial processing (6 hours)
    1. Background, histogram processing (histogram equalization, matching)
    2. Spatial filtering, convolution filters, mathematical properties of filters, linear filters, convolution, low pass and high pass filters, high boost filters, directional filters
    3. Characteristics of filtered images, the box filter algorithm, cascade filters, statistical filters, morphological filters, gradient filters,

  5. Image enhancement in frequency domain (9 hours)
    1. Introduction to fourier transforms, fourier analysis and synthesis, discrete fourier transforms in two- D
    2. The fourier components, filtering with fourier transforms, transfer functions system modelling using the fourier transform
    3. The power spectrum, scale space transforms, image resolution pyramids, zero crossing filters, laplacian of gaussian filters, difference of gaussain filters wavelet transforms

  6. Image segmentation (6 hours)
    1. Background, bottom of approach, talk down approach
    2. Point line and edge detection, thresholding, region based segmentation
    3. Multi-resolution segmentation, watershed based segmentation

  7. Image restoration and reconstruction (6 hours)
    1. Model of image degradation and restoration process, noise models
    2. Restoration in the presence of noise only, periodic noise reduction, linear
    3. Position -invariant degradations, estimating degradation function, inverse filtering wiener filtering, geometric mean filtering, image reconstruction from projection

  8. Color image processing (3 hours)
    1. Color fundamentals, color models, pseudo color processing
    2. Color transformation, colors segmentation, smoothing and sharpening

  9. Image compression (6 hours)
    1. Fundamentals (coding redundancy, spatial and temporal redundancy, irrelevant information, measuring image information, fidelity criteria)
    2. Compression methods (Huffman coding, The Discrete Cosine Transform (DCT), golomb coding, arithmetic coding, run-length compression, symbol based coding, black transform coding, wavelet coding), Quantization, JPEG_LS and MPEG.

Tutorials and Practical:
The tutorial should be detail guideline and should be explain at least once before going though the practical. The handout should contain questions pertaining to the basic concepts, analysis and application of each practical such that student can understand and visualize the concept behind the tutorial and practical component of the course. The tutorial and practical can be carried out using MATLAB or ENVI/IDL tools

  1. Image Enhancement techniques at the spectral, spatial and frequency domain:
    1. Spectral domain
    2. Spatial domain
    3. Frequency domain
  2. Image Segmentation
  3. Colour Image Processing
  4. Image compression

Three Assessments: The average of Three assessment will be the final internal mark

  1. Text Book: Digital Image Processing, R.C. Gonzalez, R.E. Woods
  2. Digital Image Processing: A.K Jain
  3. Digital Picture Processing, Azriel Rosenfeld and Avinash Kak, Academic Press
  4. Digital Image Processing by Rafael C. Gonzalez and Richard E. Woods , Pearson Prentice Hall
  5. Remote Sensing and Image Interpretation by Thomas M. Lillesand, Ralph W. Kiefer and Jonathan W. Chipman, Wiley student edition

Evaluation scheme:
The question will cover the entire chapter in the syllabus. The evaluation scheme will be as mentioned in the table below




Marks Allocation*










6, 9.1














*There will be minor deviation in the marks distribution

Sponsered Ads