Progressive Transmission of Medical Images via a Bank of Generative Adversarial Networks

The healthcare sector is currently undergoing a major transformation due to the recent advances in deep learning and artificial intelligence. Despite a significant breakthrough in medical imaging and diagnosis, there are still many open issues and undeveloped applications in the healthcare domain. In particular, transmission of a large volume of medical images proves to be a challenging and time-consuming problem, and yet no prior studies have investigated the use of deep neural networks towards this task. The purpose of this paper is to introduce and develop a deep-learning approach for the efficient transmission of medical images, with a particular interest in the progressive coding of bit-planes. We establish a connection between bit-plane synthesis and image-to-image translation and propose a two-step pipeline for progressive image transmission. First, a bank of generative adversarial networks is trained for predicting bit-planes in a top-down manner, and then prediction residuals are encoded with a tailored adaptive lossless compression algorithm. Experimental results validate the effectiveness of the network bank for generating an accurate low-order bit-plane from high-order bit-planes and demonstrate an advantage of the tailored compression algorithm over conventional arithmetic coding for this special type of prediction residuals in terms of compression ratio.

An Effective Progressive Image Transmission using Superpixel based Saliency Detection and Modified SPIHT Compression Algorithm

A new progressive image transmission system was proposed in this research paper for effective usage of communication bandwidth. At first, the superpixel based saliency detection method was used for segmenting the foreground region from the background region, because it gives more saliency information of an image with the benefit of color contrast. Then, Integer Wavelet Transform (IWT) was applied in the foreground image, which delivers A good quality of the image and also the compression ratio of the image was decent. Additionally, optimized neural network and modified Set Partitioned in Hierarchical Tree (SPIHT) algorithm were applied in the background image that delivers good rate distortion properties in the noise free environment and also enhances the image visual experience. In modified SPIHT, the sub-tree roots were not excluded that helps to encode and quantize the wavelet coefficients effectively. Also, it delivers more information to the image edges that effectively improves the subjective visual experience. Experiment report showed that the proposed work enhanced the Peak Signal to Noise Ratio (PSNR) upto 5dB compared to the existing work.

Study on Nearly Lossless Compression with Progressive Decoding

A progressive transmission and decoding nearly lossless compression algorithm is proposed. The image data are grouped according to different frequencies based on DCT transform, then it uses the JPEG-LS core algorithmtexture prediction and Golomb coding on each group of data, in order to achieve progressive image transmission and decoding. Experimentation on the standard test images with this algorithm and comparing with JPEG-LS shows that the compression ratio of this algorithm is very similar to the compression ratio of JPEG-LS, and this algorithm loses a little image information but it has the ability of the progressive transmission and decoding.

Research on Image Technology with Progressive Image Compression Based on JPEG and Laplacian Pyramid

Progressive image transmission is a kind of image technology has been widely used in various fields, it can not only save bandwidth but also improve the user experience to meet user demand for different image quality. According to user's demand for image quality, realizing the progress of image compression coding flow can meet the demand of users. This article mainly introduce by means of JPEG and Laplacian pyramid coding principle implement progressive image compression.