In this paper, we explore the secrecy performance of a visible light communication (VLC) system consisting of distributed light-emitting diodes (LEDs) and multiple users (UEs) randomly positioned within an indoor environment while considering the presence of an eavesdropper. To enhance the confidentiality of the system, we formulate a problem of maximizing the sum secrecy rate for UEs by searching for an optimal LED for each UE. Due to the non-convex and non-continuous nature of this security maximization problem, we propose an LED selection algorithm based on tabu search to avoid getting trapped in local optima and expedite the search process by managing trial vectors from previous iterations. Moreover, we introduce three LED selection strategies with a low computational complexity. The simulation results demonstrate that the proposed algorithm achieves a secrecy performance very close to the global optimal value, with a gap of less than 1%. Additionally, the proposed strategies exhibit a performance gap of 28% compared to the global optimal.

Deep learning-based algorithms for enhancing underwater images have demonstrated outstanding performance in recent years. However, numerous limitations exist in applying existing methods to various underwater environments, viewed as water types. The lack of generalization ability and visual consistency remain significant challenges, as they neglect the domain gaps of underwater scenes and the differences in the multi-frequency nature of information. A critical desideratum in underwater image enhancement is to establish a strong correspondence between the learned features and water types while also exploring multi-frequency analysis. In response to these issues, we propose a novel domain adaptive multi-frequency underwater image enhancement network. This network leverages the capability of adapting feature learning to varying water types while also supporting a simultaneous feature decode for high and low frequencies. Specifically, we first construct a module for domain adaptation with a water type classifier to distinguish the impacts of different water types, which naturally reveals domain adaptation between water types. This enhances the model’s sensitivity to water types by generating the domain-sensitive feature. In addition, we design a dual-branch decoder with self-attention to decompose the domain-sensitive features into sub-bands of high and low frequencies. Each branch contains distinct semantic content aimed at promoting multi-frequency feature complementarity. This helps the model to accurately capture local detail and global context, thereby further improving image quality. Finally, we design a multi-stage training strategy equipped with multiple loss functions to improve the stability of the model. The proposed method can adaptively deal with degraded images of different water types and enhance underwater images with high quality by processing high-frequency and low-frequency components synchronously. Extensive experiments demonstrate that the proposed method outperforms state-of-the-art methods in terms of both visual quality and evaluation metrics.

Underwater object detection is an important computer vision task that has been widely used in marine life identification and tracking. However, there are a series of challenges such as low-contrast condition, occlusion condition, unbalanced light condition, and small dense objects in underwater object detection. Attention mechanism has been proven powerful in feature extraction. However, attention mechanisms usually divide image into fixed patches. This methods lead to the splitting up of continuous structures, which hinders the use of similar information in other areas to enhance image details. In fact, using graph structure is more flexible and effective for visual perception, because it can capture complex target relation and context information effectively. Thus, we apply graph attention mechanisms to irregular patches and propose an irregular-patch graph attention network (IGA-Net). First, the superpixel segmentation method is used to segment the image to reduce noise. Second, the global graph and local graph are constructed to obtain internal structures. Finally, to handle occlusion and small objects, a distinctive feature three-way handshake (F3H) module is proposed to fuse information from global and local graph. To demonstrate the effectiveness of the proposed method, we conduct comprehensive evaluations on five challenging underwater datasets UTDAC2020, RUOD, Brackish, TrashCan, and WPBB. Experimental results demonstrate that the proposed IGA-Net achieves superior performance on five challenging underwater datasets.

This paper proposes a novel approach for structure-sensitive image processing based on the rigorous mathematical derivation of data-specific anisotropic Mexican hat wavelets (DAM). Our DAM is derived from the negative first-order derivative of the fundamental solution of heat diffusion equation with respect to time, which not only shares similar properties with Mexican hat wavelet but also intrinsically embeds the image-specific properties. Through the scale-aware DAM transform and its inverse transform, we are capable of conducting structure-sensitive image processing. Our key idea is to represent the images as undirected graphs, whose edge weights are governed by the normalized intensity/color differences within the local neighboring pixel window. Based on the rigorous theory of global graph Laplacian and heat diffusion, our original DAM can also encode the local/global structure of images. We employ the Krylov subspace technique to reduce the computational cost of our DAM transform. Furthermore, aiming at various structure-preserving image processing applications such as filtering, detail enhancement, tone manipulation, and stylization, we conduct comprehensive experiments and make quantitative comparisons with other state-of-the-art methods, which demonstrate the versatility and superiority of our method.