Design of Vector Quantizer for Image Compression Using Self-Organizing Feature Map and Surface Fitting

Abstract

We propose a new scheme of designing a vector quantizer for image compression. First, a set of codevectors is generated using the self-organizing feature map algorithm. Then, the set of blocks associated with each code vector is modeled by a cubic surface for better perceptual fidelity of the reconstructed images. Mean-removed vectors from a set of training images is used for the construction of a generic codebook. Further, Huffman coding of the indices generated by the encoder and the difference-coded mean values of the blocks are used to achieve better compression ratio. We proposed two indices for quantitative assessment of the psychovisual quality (blocking effect) of the reconstructed image. Our experiments on several training and test images demonstrate that the proposed scheme can produce reconstructed images of good quality while achieving compression at low bit rates. Index Terms-Cubic surface fitting, generic codebook, image compression, self-organizing feature map, vector quantization.

New Approaches for Image Compression Using Neural Network

“…Ideally, the mean square error should be zero for ideal decompression. The compression-decompression error is evaluated by comparing the input image and decompressed image using normalized mean square error formula [16].…”

“…True image compression occurs when the hidden layer outputs are digitized before transmission. The Figure 2 shows a typical digitization scheme using 3 bits as reported in [16]. In this scheme, there are 8 possible binary codes: 000, 001, 010, 011, 100, 101, 110, and 111.…”

A Research on Image CompressionTechniques

“…Data sent by a camera using video standards contain still image mixed with data containing changes, so that unchanged data (for instance the background) are not sent in every image. Consequently the frame rate measured in frames per second (fps) is much greater [11].…”

Fast codebook searching in a SOM-based vector quantizer for image compression

“…SOM is easy to implement and has found numerous applications (for a list of references see [13,26]). A discussion of general features of SOM algorithm and their suitability with respect to design of a codebook can be found in [19]. In [6], Chang and Gray introduced an on-line technique for VQ design using stochastic gradient algorithm which can be considered a special case of the SOM algorithm and it is shown to perform slightly better than GLA.…”

“…Amerijckx et al [2] used Kohonen's SOM algorithm to design a VQ for the coefficients of discrete cosine transform of the image blocks and further compressed the output of VQ encoder using entropy coding. In [19], the present authors used SOMs for designing VQ and proposed a novel surface-fitting method for refining the codebook for achieving better psychovisual quality of the reproduced image. However, all the above works use SOM for generating good codebooks.…”