Wavelet based Image Compression using Self Organizing Feature Maps

Despite rapid progress in mass storage density, processor speeds and digital communication system performance demand for data storage capacity and data transmission bandwidth continues to outstrip the capabilities of available technologies. Various image compression techniques have been developed to reduce the transmission rates and increase the storage capacity for still images without sacrificing much of the image quality. Image compression has thus become a hub of contemporary research activity. In this paper, a new scheme for image compression combining Discrete Wavelet Transform with Vector Quantization has been proposed. This method is based on Kohonen's Self Organizing Feature Maps (SOFM) which takes into account the neighborhood property of an image and designs the codebook. Arithmetic Coding is then used to remove redundancies between the indexes of vectors corresponding to the neighboring blocks in the original image, which then leads to further compression. The simulation results demonstrate the improved coding efficiency of the proposed method, when compared with JPEG. The proposed scheme allows achieving a compression ratio upto approximately 40:1 with reasonable image quality.