Noise reduction of VQ encoded Medical image

In the past few years, many techniques have been explored to investigate the transmission of vector quantization (VQ) indices over noisy channels. In this paper, we propose an algorithm to correct or conceal the channel errors found in a VQ encoded Medical image. In the proposed method, some verification information (bloc mean and complexity variation) is embedded into the indices of the VQ encoded data. This information can help with the detection of errors in the indices transmitted over a noisy channel. Our method does not need to rearrange the codebook. In addition, a new error concealment technique is proposed to correct the erroneous image. According to our experiments, the proposed scheme provides a high detection result in the BER 0.01 noisy channel. Comparing to the related method, Anti-Gray Coding (AGC) and checksums, the proposing method performs better under BER 10^-3 -10^-2. Moreover, our method does not have the codebook rearrangement that it is obligatory in AGC and other methods and need to more time. Therefore, our method is suitable to noisy and less noisy channels.