A5083093046- Image Enhancement Techniques
- Advanced Image Processing Techniques
- Advanced Image Fusion Techniques
- Video Surveillance and Tracking Methods
- Image and Signal Denoising Methods
- Advanced Vision and Imaging
- Advanced Neural Network Applications
- Visual Attention and Saliency Detection
- semigroups and automata theory
- Biomedical Text Mining and Ontologies
- Chaos control and synchronization
- Nonlinear Dynamics and Pattern Formation
- Fire Detection and Safety Systems
- Advanced Image and Video Retrieval Techniques
- Speech and Audio Processing
- DNA and Biological Computing
- Semantic Web and Ontologies
- Network Security and Intrusion Detection
- Medical Image Segmentation Techniques
- Advanced Computational Techniques and Applications
- Video Analysis and Summarization
- Advanced Data Compression Techniques
- Advanced Adaptive Filtering Techniques
- Cellular Automata and Applications
- Face and Expression Recognition
Southwest University
2009-2025
Nanjing Medical University
2018-2024
Nanyang Technological University
2024
Yuxi Normal University
2011-2024
Sichuan University
2007-2023
State Key Laboratory of Biotherapy
2023
East China University of Science and Technology
2023
Tencent (China)
2022
Beijing University of Posts and Telecommunications
2009-2022
Chongqing Technology and Business University
2022
Under the nighttime haze environment, quality of acquired images will be deteriorated significantly owing to influences multiple adverse degradation factors. In this paper, we develop a multi-purpose oriented removal framework focusing on hazy images. First, construct nonlinear model based classic Retinex theory formulate degradations image. Then, novel variational is presented simultaneously estimate smoothed illumination component and detail-revealed reflectance predict noise map from...
Exhaled breath analysis by nanosensors is a workable and rapid manner to diagnose lung cancer in the early stage. In this paper, we proposed Al-doped MoSe2 (Al–MoSe2) as promising biosensor for sensing three typically exhaled volatile organic compounds (VOCs) of cancer, namely, C3H4O, C3H6O, C5H8, using density functional theory (DFT) method. Single Al atom doped on Se-vacancy site surface, which behaves an electron-donor enhances electrical conductivity nanosystem. The adsorption desorption...
Nighttime image dehazing is a challenging task due to the presence of multiple types adverse degrading effects including glow, haze, blur, noise, color distortion, and so on. However, most previous studies mainly focus on daytime or partial degradations presented in nighttime hazy scenes, which may lead unsatisfactory restoration results. In this paper, we propose an end-to-end transformer-based framework for haze removal, called NightHazeFormer. Our proposed approach consists two stages:...
Snow removal causes challenges due to its characteristic of complex degradations. To this end, targeted treatment multi-scale snow degradations is critical for the network learn effective removal. In order handle diverse scenes, we propose a projection transformer (MSP-Former), which understands and covers variety degradation features in multi-path manner, integrates comprehensive scene context information clean reconstruction via self-attention operation. For local details various...
The aim of this paper is to propose a large scale dataset for image restoration (LSDIR). Recent work in has been focused on the design deep neural networks. datasets used train these networks 'only' contain some thousands images, which still incomparable with other vision tasks such as visual recognition and object detection. small training set limits performance To solve problem, we collect high-resolution (HR) images from Flickr restoration. ensure pixel-level quality collected dataset,...
Underwater Image Rendering aims to generate a true-to-life underwater image from given clean one, which could be applied various practical applications such as enhancement, camera filter, and virtual gaming. We explore two less-touched but challenging problems in rendering, namely, i) how render diverse scenes by single neural network? ii) adaptively learn the light fields natural exemplars, i,e., realistic images? To this end, we propose rendering method for imaging, dubbed UWNR (Underwater...
Most of existing dehazing algorithms are unable to deal with nighttime hazy scenarios well due complex degraded factors such as non-uniform illumination, low light and glows. To obtain high-quality image under haze imaging conditions, we present an effective single framework based on a variational decomposition model simultaneously address these undesirable issues. First, consisting three regularization terms is proposed decompose into structure layer, detail layer noise layer. Concretely,...
Outdoor images acquired under poor weather conditions are usually contaminated by suspended particles and aerosols in the atmosphere. These captured easily suffer from contrast reduction, low visibility, color distortion. In this paper, we develop a novel single image dehazing method based on large sky region segmentation multiscale opening dark channel model (MODCM). First, simple but effective for detection SVM classification is presented, which can be considered as first step of...
This paper describes a novel trajectory planning algorithm for an unmanned aerial vehicle (UAV) under the constraints of system positioning accuracy. Due to limitation structure, UAV cannot accurately locate itself. Once error accumulates certain degree, mission may fail. method focuses on correcting during flight process UAV. The improved genetic (GA) and A* are used in ensure has shortest length from starting point ending multiple least number corrections.
The effective potency and resistance of poly(ADP-ribose) polymerase (PARP) inhibitors limit their application. Here, we exploit a new paradigm that mimics the effects breast cancer susceptibility genes (BRCA) mutations to trigger possibility synthetic lethality, based on previous discovery potential lethality effect between bromodomain-containing protein 4 (BRD4) PARP1. Consequently, present study describes compound BP44 with high selectivity for BRD4 Fortunately, inhibits homologous...
Despite recent advancements in unified adverse weather removal methods, there remains a significant challenge of achieving realistic fine-grained texture and reliable background reconstruction to mitigate serious distortions.Inspired by codebook vector quantization (VQ) techniques, we present novel Adverse Weather Removal network with Codebook Priors (AWRCP) address the problem removal. AWRCP leverages high-quality priors derived from undistorted images recover vivid details faithful...
A lightweight underwater image enhancement network is of great significance for resource-constrained platforms, but balancing model size, computational efficiency, and performance has proven difficult previous approaches. In this work, we propose the Five A$^{+}$ Network (FA$^{+}$Net), a highly efficient real-time with only $\sim$ 9k parameters 0.01s processing time. The FA$^{+}$Net employs two-stage structure. strong prior stage aims to decompose challenging degradations into sub-problems,...
Single-image dehazing is a pivotal challenge in computer vision that seeks to remove haze from images and restore clean background details. Recognizing the limitations of traditional physical model-based methods inefficiencies current attention-based solutions, we propose new network combining an innovative Haze-Aware Attention Module (HAAM) with Multiscale Frequency Enhancement (MFEM). The HAAM inspired by atmospheric scattering model, thus skillfully integrating principles into...
Haze is a common weather phenomenon, which hinders many outdoor computer vision applications such as surveillance, navigation control, vehicle driving, and so on. In this paper, simple but effective unified variational model for single image dehazing presented based on the total variation regularization. From perspective of relationship between Retinex, problem can be formulated minimization Retinex model. The proposed incorporates two <inline-formula...