Analysis of X-ray images using MRA based biorthogonal wavelets for detection of minute fractures

Edge feature extraction of X-ray images plays an important role in detecting the fractures in bones. It deals with the detection of discontinuities in the gray level values which helps the medical practisioners to extract the edges of an images to detect the minute fractures of an X-ray bone image. Edge feature extraction of X-ray images using Multiresolution Analyis (MRA) based on biorthogonal wavelet is more preferable when compared with orthogonal wavelets because of more flexibility and less computation time. The orthogonal wavelet transforms like Haar, daubechies, coiflet, symlets can be used as edge detector, but a lot of false edge information will be extracted. Therefore biorthogonal wavelet transforms like bior1.3, bior2.4, bior3.9, bior5.5, bior6.8 are applied to detect the edges and are compared. Among all the methods, bior1.3 performs well in detecting the edges with better quality. The various performance metrics like Ratio of Edge pixels to size of image (REPS), peak signal to noise ratio (PSNR) and computation time are compared for various wavelets.