Poisson Wavelet Quantized Piecewise Regressive Distributed Coding for Image Compression and Transmission

Abstract

Image compression is the process of minimizing the size of a digital image file while attempt to preserve its visual quality as much as possible. This file size reduction process is achieved by eliminating redundant or unnecessary information in an image. Therefore image compression is essential for various applications, including reducing storage space requirements, improving data transmission efficiency, and enhancing the user experience in multimedia applications. The existing compression method faces most important challenges to enhance the quality of the reconstructed image with minimum time and higher compression ratio.  A novel technique called Poisson Wavelet Quantized Piecewise Regressive Distributed Coding (PWQPRDC) technique for achieving higher compression ratio.  First, the numbers of natural images are collected from the dataset. The PWQPRDC first performs the Camargo's indexive gamma map filtering technique to preprocess the input image and to remove the noisy pixels from input image. Then the PWQPRDC technique performs the image compression to obtain the storage efficiency and transmission. The image compression process includes three different process namely Poisson Wavelet Transformation, Dead-Zone Quantization and Absolute piecewise regressive Geometric Distributed Coding. In PWQPRDC techniques, Poisson wavelet transformation is used for decomposing the input image into low and high frequency coefficients. After that, Dead-Zone Quantization is applied to quantize the low frequency coefficients and effectively rounding to the nearest discrete value. Finally, the Absolute piecewise regressive Geometric Distributed Coding is applied for mining the size of the quantized image with higher compression ratio. This compressed image is transmitted through the network channel. The decompression is performed as a reverse process of the compression technique such as Decoding, Dead-Zone De-Quantization and inverse Poission wavelet transformation. Experimental evaluation is carried out on factors such as peak signal-to-noise ratio (PSNR), compression ratio, compression time and Storage saved, with respect to image size. The analyzed results indicates that the superior performance of our proposed PWQPRDC techniques model when compared with existing methods.