Projects

Image Compression using Discrete Cosine Transform
Used Discrete Cosine Transform to compress images
Image Segmentation using Watershed Transform
Implemented a modified watershed algorithm using adaptive thresholding and adaptive masking techniques
Model to predict flight performance
Implemented models to predict flight on-time performance, whether it was delayed or not using flight arrival and departure data· Used machine learning techniques like SVM, random forests and neural networks.
Shape-Preserving Half-Projective Warps for Image Stitching
Implemented a novel parametric warp, a spatial combination of a projective transformation and a similarity transformation. By this, the field of view could be extended by stitching images with less projective distortion.