Published Paper Details:

Graphs are very powerful tools for describing many problems and structures in image processing. But graphs only describe some binary relations and are not always sufficient for modeling some complex problems or data. In order to overcome these difficulties and to provide a more viable tool for processing images by taking into account the topological and geometrical aspects of an image, combinatorial or hypergraph model is proposed. A relatively new class of algorithm termed combinatorial algorithm relies on approach to image processing. Combinatorics refers to arranging the elements into sub sets. This project aims three applications in particular to Denoising of Image, Image compression and Image Edge detection using combinatorial analysis. First the image can be converted into Hypergraph model by using Image Adaptive Neighborhood Hypergraph (IANH) model then this model can be used for different applications of image processing. Impulse noise is detected by analysing unit cardinality pixels in IANH model, that can be removed by different traditional methods and comparing PSNR and BER of these methods. In this project relates to a new method for lossless image compression, based on combinatorial (Hypergraph) analysis HGLC (Hyper Graph Lossless Compression). First the image can be represented into Hypergraph model by using IANH model. Then it can be modeled into number of rectangles to remove redundant data using algorithm to compress the image.