Implement algorithmic concepts of DIP, Adaptive Signal Processing and Estimation & Detection. Focus is also put on implementation in C/C++ using standard image processing or scientific libraries. At the end of the lab course, a student should feel comfortable developing a project/executable that would manipulate images/real time signals using tunable filters. The students shall also develop an ability to perform in small groups.


  • Reading an image, use openCV. Access pixel values. Concept of pointer.

  • Perform basic image enhancement algorithm – histogram equalization. Display/save histogram as a binary image.

  • Perform basic image denoising algorithm - low pass filters, median filters.

  • Weiner filtering - denoising.

  • Estimation of noise - least squares approach. Use of GNU Scientific Library.

  • Image blurring, blurring system identification, image deblurring.

  • Adaptive prediction.


Design Project - 2 per group. Each group has the option of implementing in the DSP kit. Suggested topics:

  • Reading in speech signal, adding noise, LMS based noise removal filter.

  • Image compression - lossless compression using prediction.

  • JPEG implementation.

  • Classification using Bayes theory - skin color identification using bayes theory.