An algorithm for image data compression based on vector quantization

Image data compression is concerned with minimization of the number of information carrying units used to represent an image. Fundamental goal of data compression is to store reduced number of bits or transmit fewer bits. New algorithms for data compression and, consequent image restoration are developed and implemented. Data compression is achieved by actual quantization, or conversion into a fixed or discrete quantities. Two dimensional discrete integral-valued sample of image pixels are continually quantized, until a desired set of codebook vectors is formed. This codebook is then used for image data compression. Thus, data can be either transmitted or stored with reduced bits. It is shown that using the same codebook, image data can be easily restored back to its original form, while maintaining the necessary fidelity of the data.