Discusses the image compression technology where the aim is to narrow the transmitted band-width as much as possible. Such compression reduces the cost of transmission and for military use it reduces the susceptibilit...
详细信息
Discusses the image compression technology where the aim is to narrow the transmitted band-width as much as possible. Such compression reduces the cost of transmission and for military use it reduces the susceptibility to interference, but the problem is that the higher the compression, the greater the loss of image quality.
Despite deep learning's progress in semantic communication, traditional fixed-length encoding does not adequately address the variable complexity of semantic content, often leading to loss of critical nuances and ...
详细信息
Despite deep learning's progress in semantic communication, traditional fixed-length encoding does not adequately address the variable complexity of semantic content, often leading to loss of critical nuances and reduced communication accuracy. Current methods also introduce unnecessary redundancy, compromising transmission efficiency. Addressing these challenges, our work introduces an innovative adaptive rate encoding mechanism that captures the intrinsic semantics of images and fine-tunes the coding rate based on semantic interconnection probability. We employ a cross-attention model to construct a layered semantic probability graph parsed into a hierarchical semantic tree, which represents the probabilistic relationships of image semantics and unravels the latent semantic structure. This not only delineates the image's semantic architecture but also enables our adaptive encoding to dynamically allocate resources, minimizing redundancy and enhancing efficiency. Our experiments confirm that our approach provides a more judicious bit allocation to complex image features and allocates more bits to semantically rich features while achieving superior compression of simpler content. The proposed method not only improves upon existing semantic fidelity metrics but also reduces the bit demand for transmitting complex images. Our adaptive encoding strategy represents a significant stride in leveraging the endogenous semantic information of images for more accurate and efficient communication.
This paper introduces a novel coding scheme based on Tree-Structured Vector Quantisation (TSVQ) scheme for image compression. The genealogical relationship among the indices of the neighbouring blocks generated from v...
详细信息
This paper introduces a novel coding scheme based on Tree-Structured Vector Quantisation (TSVQ) scheme for image compression. The genealogical relationship among the indices of the neighbouring blocks generated from vector quantisation is exploited to improve the coding performance of TSVQ. The proposed coding scheme provides about 3.5 dB improvement over the basic TSVQ scheme and outperforms VQ schemes with memory and JPEG coding standard at low bit-rates. In addition its performance is comparable with address VQ but with much less complexity. (C) 1999 Elsevier Science B.V. All rights reserved.
Various variants and hybrid approaches evolved from the origin of Delp and Mitchell's block truncation coding (BTC) or moment preserving quantizer (MPQ) have formed a niche as an effective and simple image compres...
详细信息
Various variants and hybrid approaches evolved from the origin of Delp and Mitchell's block truncation coding (BTC) or moment preserving quantizer (MPQ) have formed a niche as an effective and simple image compression methodology with attractive coding performance achieved at moderate bitrates. As ETC is still lacking a fundamental error analysis, in this paper we present some fundamental insights regarding one-bit (or two-level) BTC's truncation error by providing mathematical analysis as well as novel geometric interpretation. We further show that the mean-square error (MSE) of Lema and Mitchell's absolute moment block truncation coding (AMBTC) is always bounded below (i.e., less than or equal to) that of ETC. Therefore, with additional advantages in computation and implementation, AMBTC is always superior. Furthermore, we developed a new adaptive equal sign position optimization (ESPO) algorithm for optimum pixel classification. Our quantization error analysis shows that incorporating the ESPO algorithm into conventional AMBTC or ETC achieves minimum MSE in either case. (C) 2000 Elsevier Science B.V. All rights reserved.
Compressive learning (CL) is an emerging framework that integrates signal acquisition via compressed sensing (CS) and machine learning for inference tasks directly on a small number of measurements. It can be a promis...
详细信息
Compressive learning (CL) is an emerging framework that integrates signal acquisition via compressed sensing (CS) and machine learning for inference tasks directly on a small number of measurements. It can be a promising alternative to classical image-domain methods and enjoys great advantages in memory saving and computational efficiency. However, previous attempts on CL are not only limited to a fixed CS ratio, which lacks flexibility, but also limited to MNIST/CIFAR-like datasets and do not scale to complex real-world high-resolution (HR) data or vision tasks. In this article, a novel transformer-based compressive learning framework on large-scale images with arbitrary CS ratios, dubbed TransCL, is proposed. Specifically, TransCL first utilizes the strategy of learnable block-based compressed sensing and proposes a flexible linear projection strategy to enable CL to be performed on large-scale images in an efficient block-by-block manner with arbitrary CS ratios. Then, regarding CS measurements from all blocks as a sequence, a pure transformer-based backbone is deployed to perform vision tasks with various task-oriented heads. Our sufficient analysis presents that TransCL exhibits strong resistance to interference and robust adaptability to arbitrary CS ratios. Extensive experiments for complex HR data demonstrate that the proposed TransCL can achieve state-of-the-art performance in image classification and semantic segmentation tasks. In particular, TransCL with a CS ratio of 10% can obtain almost the same performance as when operating directly on the original data and can still obtain satisfying performance even with an extremely low CS ratio of 1%. The source codes of our proposed TransCL is available at https://***/MC-E/TransCL/.
We introduce new transforms for efficient compression of image blocks with directional preferences. Each transform, which is an orthogonal basis for a specific direction, is constructed from an eigen-decomposition of ...
详细信息
We introduce new transforms for efficient compression of image blocks with directional preferences. Each transform, which is an orthogonal basis for a specific direction, is constructed from an eigen-decomposition of a discrete directional Laplacian system matrix. The method is a natural extension of the DCT, expressing the Laplacian in Cartesian coordinates rotated to some predetermined angles. Symmetry properties of the transforms over square domains lead to efficient computation and compact storage of the directional transforms. A version of the directional transforms was implemented within the beyond HEVC software and demonstrated significant improvement for intra block coding.
Hyperspectral imaging is a powerful technology for remotely inferring the material properties of the objects in a scene of interest. Hyperspectral images consist of spatial maps of light intensity variation across a l...
详细信息
Hyperspectral imaging is a powerful technology for remotely inferring the material properties of the objects in a scene of interest. Hyperspectral images consist of spatial maps of light intensity variation across a large number of spectral bands or wavelengths; alternatively, they can be thought of as a measurement of the spectrum of light transmitted or reflected from each spatial location in a scene. Because chemical elements have unique spectral signatures, observing the spectra at a high spatial and spectral resolution provides information about the material properties of the scene with much more accuracy than is possible with conventional three-color images. As a result, hyperspectral imaging is used in a variety of important applications, including remote sensing, astronomical imaging, and fluorescence microscopy.
Single-chip cameras usually incorporate a CFA (color filter array) on the sensor to obtain color information. Color interpolation is then needed to recover the color images. Color coding was conventionally implemented...
详细信息
Single-chip cameras usually incorporate a CFA (color filter array) on the sensor to obtain color information. Color interpolation is then needed to recover the color images. Color coding was conventionally implemented after the interpolation process. Two drawbacks inherited are a long processing time, and a requirement for a large memory buffer. Direct coding of the sensor data before color interpolation results in enormous artifacts and poor compression efficiency. This paper will point out the problems theoretically and introduce a signal processing method incorporating a DCT (Discrete Cosine Transform) compression scheme to avoid these problems. Simulation results show good compression performance with good image quality and relatively low artifacts. This approach is also appropriate for real-time implementation in single-chip cameras.
We present a new lossy technique of runlength coding applied to gray-level images. In lossless runlength coding applied to gray-level images, one finds runs of gray levels that have the same value along a horizontal r...
详细信息
We present a new lossy technique of runlength coding applied to gray-level images. In lossless runlength coding applied to gray-level images, one finds runs of gray levels that have the same value along a horizontal row and then encodes them in the form (C, L), where C is gray-level value and L is the runlength. The drawback to this approach is that it does not give much compression, as L is usually small. In our approach we relax the criterion of requiring runs of exactly the same gray level;instead we find runs of pixel gray-level values that lie within a dynamic range, the dynamic window range, and encode them in the form (C, L). Here, C represents approximations of the pixel gray levels and L the runlength. This approach, although lossy, promises to give better compression. Additionally, if lossless Huffman coding is used in the final stage a slightly higher compression ratio is achieved. This technique was used to process a set of images and the results based on windowlength, compression ratio, root-mean-square error, and peak signal-to-noise ratio are tabulated and plotted. The compressed images and the corresponding error images are provided to illustrate the results. Additionally, performance of this algorithm was compared with the runlength coding of comparable coarsely quantized images. (C) 1995 Academic Press, Inc.
In this paper, it is shown that cubic-spline interpolation (CSI) [1] can be performed by a fast and efficient computation for the encoding and decoding processes of image coding. It requires substantially fewer additi...
详细信息
In this paper, it is shown that cubic-spline interpolation (CSI) [1] can be performed by a fast and efficient computation for the encoding and decoding processes of image coding. It requires substantially fewer additions and multiplications than the original CSI algorithm. Furthermore, a new type of overlap-save scheme is utilized to solve the boundary-condition problems that occur between two neighboring subimages in the actual image. It is also shown that a very efficient nine-point Winograd discrete Fourier transform (WDFT) can be used to replace the fast Fourier transform (FFT) needed to implement the CSI scheme in the modified JPEG encoder. Finally, the proposed fast new CSI algorithm with a compression ratio of 9:1 is used along with the JPEG standard to speed up the modified JPEG encoder-decoder and still obtain a better quality of reconstructed image for higher compression ratios.
暂无评论