Comparison of lossless image compression algorithms

The article considers the relevance of the development of lossless image compression and transmission algorithms and their application for creating transport video surveillance systems. A brief overview of lossless transport image compression methods is provided. We propose a method for compressing transport plots based on the pyramid-recursive method of splitting the source image into polygons of various shapes and sizes. We consider two new algorithms for implementing the proposed method that are fundamentally different from each other: with a transition to the spectral region and without a transition to the spectral region of the original signal to ensure lossless compression. The results of testing various well-known lossless compression algorithms are analyzed: series length, Huffman, and arithmetic encoding, and compared with the proposed algorithms. It is shown that the proposed algorithms are more efficient in terms of compression ratio (2—3 times) compared to the known ones, while the computational complexity increases approximately by more than 3-4 times.

Lossless text compression methods involve some form of moderately high-order exact string matching. However, this work cannot easily be carried over to lossless image compression, because images are two-dimensional and (more important) essentially quantized analog data. A better plan is to find and encode as much of the image structure of the data as possible, and then to encode efficiently the unstructured, noisy residual. In three steps the authors predict the value of each pixel, model the error of the prediction, and encode the error of the prediction. Having a probabilistic model for the errors, they can use arithmetic coding to encode the errors efficiently with respect to the model.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

This paper presents a new method of lossless image compression. An image is characterized by homogeneous parts. The bit planes, which are of high weight are characterized by sequences of 0 and 1 are successive encoded with RLE, whereas the other bit planes are encoded by the arithmetic coding (AC) (static or adaptive model). By combining an AC (adaptive or static) with the RLE, a high degree of adaptation and compression efficiency is achieved. The proposed method is compared to both static and adaptive model. Experimental results, based on a set of 12 gray-level images, demonstrate that the proposed scheme gives mean compression ratio that are higher those compared to the conventional arithmetic encoders.

A color-mapped image is composed of a palette and an image of indices. The color of each pixel is the palette color referenced by the pixel's index. The indices image is usually far from being smooth. This affects its compressibility by lossless image compression methods, due to their reliance on the natural smoothness of images. The indices images can be smoothed by re-ordering the palette, ie, assigning the indices to the colors in a better way. We propose a new method for palette ordering based on an approximate solution of the traveling salesman problem. The proposed method has low complexity, dependent only on the size of the palette. It is found to significantly improve the compression ratios of color-mapped images coded with the JPEG-LS lossless compression standard. The resulting compression ratios are better than those achieved by the popular GIF image file format.

In this paper, a method for lossless and near lossless compression of large digital mammograms is proposed. This method is based on a predictive coder that uses integer-to-integer operations, and generates a two-layer embedded bit stream (i.e. near-lossless and refinement layer). The maximum tolerated pixel distortion is guaranteed in the near lossless component. Comparisons with other proposed approaches are based on a database of high-resolution 12 bits/pixel digital mammograms. The simulation results indicate that our lossless compression method can offer average bit rates 39, 5%, 24%, 7% and 2.5% better than PNG, JPEG 2000, JPEG-lossless, and LOCO, respectively. Besides, for the same images, the proposed method offers near lossless compression at relatively low computational cost, providing an average bit rate of 2.57 bits/pixel and PSNR of 47.77 dB.

This paper presents a new lossless image compression method based on the learning of pixel values and contexts through multilayer perceptrons (MLPs). The prediction errors and contexts obtained by MLPs are forwarded to adaptive arithmetic encoders, like the conventional lossless compression schemes. The MLP-based prediction has long been attempted for lossless compression, and recently convolutional neural networks (CNNs) are also adopted for the lossy/lossless coding. While the existing MLP-based lossless compression schemes focused only on accurate pixel prediction, we jointly predict the pixel values and contexts. We also adopt and design channel-wise progressive learning, residual learning, and duplex network in this MLP-based framework, which leads to improved coding gain compared to the conventional methods. Experiments show that the proposed method performs better than the conventional non-learning algorithms and also recent learning-based compression methods with practical computation time.

We consider and evaluate the context clustering method for lossless image compression based on the existing LOCO-I algorithm used in JPEG-LS - the latest lossless image compression standard. We employ the LOCO-I Medpredictor to enroll the error pixels. The contexts are defined by calculating gradient of current pixels. The three directional gradients are quantized with different codebook size (7, 9, 19) respectively. The error pixels are then corrected and encoded by the clustered-contexts. A main advantage of using the context clustering method is that it can eliminate the storage of probability vector. An adaptive arithmetic encoder is also introduced to yield a higher compression rate.

Space research organizations, hospitals and military air surveillance activities, among others, produce a huge amount of data in the form of images hence a large storage space is required to record this information. In hospitals, data produced during medical examination is in the form of a sequence of images and are very much correlated; because these images have great importance, some kind of lossless image compression technique is needed. Moreover, these images are often required to be transmitted over the network. Since the availability of storage and bandwidth is limited, a compression technique is required to reduce the number of bits to store these images and take less time to transmit them over the network. For this purpose, there are many state-of the-art lossless image compression algorithms like CALIC, LOCO-I, JPEG-LS, JPEG20000; Nevertheless, these compression algorithms take only a single file to compress and cannot exploit the correlation among the sequence frames of MRI or CE images. To exploit the correlation, a new algorithm is proposed in this paper. The primary goals of the proposed compression method are to minimize the memory resource during storage of compressed data as well as minimize the bandwidth requirement during transmission of compressed data. For achieving these goals, the proposed compression method combines the single image compression technique called super spatial structure prediction with inter-frame coding to acquire grater compression ratio. An efficient compression method requires elimination of redundancy of data during compression; therefore, for elimination of redundancy of data, initially, the super spatial structure prediction algorithm is applied with the fast block matching approach and later Huffman coding is applied for reducing the number of bits required for transmitting and storing single pixel value. Also, to speed up the block-matching process during motion estimation, the proposed method compares those blocks that have identical sum and leave the others; therefore, the time taken by the block-matching process is reduced by minimizing the unnecessary overhead during the blockmatching process. Thus, in the proposed fast lossless compression method for medical image sequences, the twostage redundant data elimination process ultimately reduces the memory resource required for storing and transmission. The method is tested on the sequences of MRI and CE images and produces an improved compression rate.

Lossless image compression plays a vital role in improving data storage and transmission efficiency without compromising data integrity. However, the throughput of current lossless compression and decompression systems remains limited and is unable to meet the growing demands of high-speed data transfer. To address this challenge, a previously proposed hybrid lossless compression and decompression algorithm has been implemented on an FPGA platform. This implementation significantly improves processing speed and efficiency. A multi-core system architecture is introduced, utilizing the processing system (PS) and programmable logic (PL) of a Xilinx Zynq-706 evaluation board. The PS handles coordination. The PL performs compression and decompression using multiple cores. Each core can process up to eight image tiles at the same time. The compression process is designed with a four-stage pipeline, and decompression is managed by a dynamic state machine to ensure optimized control. The parallel architecture and innovative algorithm design enable high-throughput operation, achieving compression and decompression rates of 480 Msubpixels/s and 372 Msubpixels/s, respectively. Through this work, a practical and high-performance solution for real-time lossless image compression is demonstrated.

Pixelwise linear prediction using backward-adaptive least-squares or weighted least-squares estimation of prediction coefficients is currently among the state-of-the-art methods for lossless image compression. While current research is focused on mean intensity prediction of the pixel to be transmitted, best compression requires occurrence probability estimates for all possible intensity values. Apart from common heuristic approaches, we show how prediction error variance estimates can be derived from the (weighted) least-squares training region and how a complete probability distribution can be built based on an autoregressive image model. The analysis of image stationarity properties further allows deriving a novel formula for weight computation in weighted least-squares proofing and generalizing ad hoc equations from the literature. For sparse intensity distributions in non-natural images, a modified image model is presented. Evaluations were done in the newly developed C++ framework volumetric, artificial, and natural image lossless coder (Vanilc), which can compress a wide range of images, including 16-bit medical 3D volumes or multichannel data. A comparison with several of the best available lossless image codecs proofs that the method can achieve very competitive compression ratios. In terms of reproducible research, the source code of Vanilc has been made public.

Problem statement: Identifying the new lossless image compression algorithm for high performance applications like medical and satellite imaging; a high quality lossless image is most important when reproduction which leads to classify the data for decision making. Approach: A new lossless hybrid algorithm based on simple selective scan order with Bit Plane Slicing method is presented for lossless Image compression of limited bits/pixel images, such as medical images, satellite images and other still images common in the world. Efficient coding is achieved by run length and modified Huffman coding. This approach is combined with efficient selective scan order for entire image in one pass through. Results: The new hybrid algorithm achieves good compression rate, compared to the existing schemes of coding with different test images. Conclusion: Compared to the existing standards JPEG-LS and CALIC, the compression rate is reduced with our proposed algorithm for different standard test images.

A novel near lossless image compression method is proposed in this paper. In the proposed method, a redundant representation of the image is firstly obtained by pixel level shifting and then partitioned into image blocks with the same size. Next, the image blocks are classified into smooth image blocks and non-smooth image blocks. For the smooth image blocks, improved block coding which includes mean residual error is utilized for pixel rectangles with the same pixel value in each to remove the redundancy. Lastly, the whole image is scanned by efficient image scanning that only one pixel is scanned for each pixel rectangle and then compressed by Huffman coding. Simulation results indicate that the proposed method has good performance compared with traditional lossless/near lossless image compression methods.

In this paper, a new method lossless for real time compression of still images in the field of the artificial vision, is presented. The proposed method is based on a new focus, which incorporates the simultaneous execution of the acquisition and compression process of an image. The method consists of three processes. In the first one, a linear camera captures a line, which is stored in memory. In the next step, the compression process of this line is executed at the same time as the acquisition process reads the next line. The process is repeated downwards starting from the upper line of the image. The novelty and main contribution of this method is in the compression process. A new coding algorithm is formulated and developed. The proposed method guarantees a ratio of compression of 3:1 with high probability and a real time execution.

With the success of deep learning in computer vision applications, deep learning based algorithms have also been proposed for image processing applications. One such application is lossless image compression. Most traditional lossless image compression algorithms use pixel-by-pixel processing algorithms. In the pixel-by-pixel algorithms, each pixel is predicted from the previously coded neighbor pixels and the prediction error is compressed without loss. Deep learning based lossless image compression algorithms can be categorized into two categories, namely prior based algorithms and pixel-by-pixel (or masked convolution based) algorithms. In the pixel-by-pixel algorithms, each pixel's probability distribution is obtained by processing the previously coded neighboring pixels with a neural network, which is then used by an arithmetic coder for lossless compression. This paper explores a deep learning based architecture, which utilizes masked convolutions, to model probability distributions of pixels and also presents a method to improve the parallelization of the algorithms. The obtained compression performance is competitive and is compared to both state-of-the art traditional and deep learning based methods.

In this paper a new lossless image compression method named DPCM-IWPT is proposed. First, wavelet packets transform (WPT) and lifting scheme (LS) are described. Then an application of the LS to WPT is presented which leads to the generation of integer wavelet packets transform (IWPT). After analyzing and implementing the effect of differential pulse code modulation (DPCM) on lossless image compression, a new method combining DPCM and IWPT (DPCM-IWPT) is proposed. Comparing with IWT and IWPT, the method DPCM-IWPT shows better performance in lossless image compression.