A colorization-based anisotropic variational model for vector-valued image compression

Abstract

<p style='text-indent:20px;'>Image compression is an important technology in digital image processing. In this paper, a novel colorization-based codec for vector-valued images is proposed. In compression, we first define the concept of "structure image", which contains rich geometric structure information of the vector-valued image. Then, to extract representative pixels from the original vector-valued image, a "one-iteration method" is proposed. It can tremendously improve compression efficiency. In decompression, starting from colorizing the structure image, an anisotropic variational model is proposed. The existence and uniqueness of minimizers for the proposed variational model are established. Besides, we develop a fast and efficient algorithm for solving the model numerically by employing the scaled form of the alternating direction method of multipliers (ADMM). Numerical experiments on natural color images demonstrate that the proposed method outperforms the state-of-art colorization-based image compression method. Compared with the transform-based approaches, experiments on satellite multispectral images illustrate that the proposed method is superior to the JPEG and JPEG2000 standards.</p>

An anisotropic variational pansharpening model with adaptive coefficients

“…Isotropic regularization applies the same diffusion rate function in all directions, which may result in a loss of significant directional information and potential blurring of edges and fine details. In contrast, by adopting the diffusion process in different directions based on the local structure and orientation of image features, anisotropic regularization can provide a more suitable approach for vector-valued image processing tasks [39,64]. In the following section, we will show how to construct an anisotropic regularization term based on the available PAN image.…”

“…That is, the spatial information of a high-resolution MS image can be expressed by measuring the gradient field of the corresponding PAN image. Inspired by the impressive performance achieved by the anisotropic variational models for image compression [39,64], we propose the following anisotropic regularization term for pansharpening,…”

Computer vision-based image data processing of digital media video images with ultra-clear picture quality

“…However, due to subjective and objective constraints, much information cannot be obtained directly or accurately by the human visual system, so humans naturally want to use external devices to help them process or understand information, which brings a new research topic for human science and technology development -computer vision [2]. Computer vision is the science of how to make machines "see", which can simulate, extend or expand human intelligence to help humans solve large-scale complex problems [3]. Computer vision tasks have a wide range of applications, such as human recognition, vehicle or pedestrian detection, target tracking, image generation, etc.…”