- Image Enhancement Techniques
- Visual Attention and Saliency Detection
- Visual perception and processing mechanisms
- Color Science and Applications
- Advanced Image Fusion Techniques
- Advanced Clustering Algorithms Research
- Advanced Image and Video Retrieval Techniques
- Advanced Vision and Imaging
- Neural dynamics and brain function
- Advanced Radiotherapy Techniques
- Advanced Image Processing Techniques
- Complex Network Analysis Techniques
- Video Surveillance and Tracking Methods
- Retinal Imaging and Analysis
- Data Management and Algorithms
- Medical Image Segmentation Techniques
- Blind Source Separation Techniques
- Face and Expression Recognition
- Angiogenesis and VEGF in Cancer
- Data Mining Algorithms and Applications
- Generative Adversarial Networks and Image Synthesis
- Radiomics and Machine Learning in Medical Imaging
- Human Pose and Action Recognition
- Face Recognition and Perception
- Advanced Algorithms and Applications
University of Electronic Science and Technology of China
2016-2025
Naval University of Engineering
2013-2025
Jiangsu Second Normal University
2025
Sun Yat-sen University
2017-2025
Hefei University of Technology
2025
Sun Yat-sen Memorial Hospital
2022-2025
Southwest Medical University
2016-2025
Chinese Academy of Medical Sciences & Peking Union Medical College
2022-2025
Suzhou Guangji Hospital
2025
Soochow University
2025
An ultrasensitive photoelectrochemical immunoassay of cancer biomarker α-fetoprotein (AFP) is proposed that uses titanium dioxide (TiO(2)) coupled with AFP-CdTe-GOx bioconjugate, which featured AFP antigen and glucose oxidase (GOx) labels linked to CdTe quantum dots (QDs) for signal amplification. The synthesized QDs yielded a homogeneous narrow size distribution, allowed the binding GOx on QDs. Greatly enhanced sensitivity came from dual amplification strategy. First, an effective matching...
Motor imagery-based brain-computer interface (MI-BCI) systems hold promise in motor function rehabilitation and assistance for impaired people. But the ability to operate an MI-BCI varies across subjects, which becomes a substantial problem practical BCI applications beyond laboratory.Several previous studies have demonstrated that individual performance is related resting state of brain. In this study, we further investigate offline variations through perspective resting-state...
We propose an underwater image enhancement model inspired by the morphology and function of teleost fish retina. aim to solve problems degradation raised blurring nonuniform color biasing. In particular, feedback from color-sensitive horizontal cells cones a red channel compensation are used correct bias. The center-surround opponent mechanism bipolar amacrine interplexiform then serve enhance edges contrasts output image. ganglion with color-opponent for correction. Finally, we adopt...
Illuminant estimation is a key step for computational color constancy. Instead of using the grey world or edge assumptions, we propose in this paper novel method illuminant by information pixels detected given color-biased image. The underlying hypothesis that most natural images include some detectable are at least approximately grey, which can be reliably utilized estimation. We first validate our assumption through comprehensive statistical evaluation on diverse collection datasets and...
The double-opponent (DO) color-sensitive cells in the primary visual cortex (V1) of human system (HVS) have long been recognized as physiological basis color constancy. In this work we propose a new constancy model by imitating functional properties HVS from single-opponent (SO) retina to DO V1 and possible neurons higher cortexes. idea behind proposed double-opponency based (DOCC) originates substantial observation that distribution responses color-biased images coincides well with vector...
A traffic driving environment is a complex and dynamically changing scene. When driving, drivers always allocate their attention to the most important salient areas or targets. Traffic saliency detection, which computes prior targets in specific environment, an indispensable part of intelligent transportation systems could be useful supporting autonomous sign training, car collision warning, other tasks. Recently, advances visual models have provided substantial progress describing eye...
The morphology of retinal vessels is closely associated with many kinds ophthalmic diseases. Although huge progress in vessel segmentation has been achieved the advancement deep learning, some challenging issues remain. For example, can be disturbed or covered by other components presented retina (such as optic disc lesions). Moreover, thin are also easily missed current methods. In addition, existing fundus image datasets generally tiny, due to difficulty labeling. this work, a new network...
Benefitting from insensitivity to light and high penetration of foggy environments, infrared cameras are widely used for sensing in nighttime traffic scenes. However, the low contrast lack chromaticity thermal (TIR) images hinder human interpretation portability high-level computer vision algorithms. Colorization translate a TIR image into daytime color (NTIR2DC) may be promising way facilitate scene perception. Despite recent impressive advances translation, semantic encoding entanglement...
Object detection is an important task for self-driving vehicles or advanced driver assistant systems (ADASs). Additionally, visual selective attention a crucial neural mechanism in driver's vision system that can rapidly filter out unnecessary information driving scene. Some existing models detect all objects scenes from the aspect of computer vision. However, changing environment, detecting salient critical appearing drivers' interested safety-relevant areas more useful ADASs. In this...
Recently, a new density-peak-based clustering method, called with local density peaks-based minimum spanning tree (LDP-MST), was proposed, which has several attractive merits, e.g., being able to detect arbitrarily shaped clusters and not very sensitive noise parameters. Nevertheless, we also found the limitation of LDP-MST in efficiency. Specifically, <inline-formula><tex-math notation="LaTeX">$O(N\log N+M^{2})$</tex-math></inline-formula> time, where...
Thyroid nodules are one of the most common nodular lesions. The incidence thyroid cancer has increased rapidly in past three decades and is cancers with highest incidence. As a non-invasive imaging modality, ultrasonography can identify benign malignant nodules, it be used for large-scale screening. In this study, inspired by domain knowledge sonographers when diagnosing ultrasound images, local global feature disentangled network (LoGo-Net) proposed to classify nodules. This model imitates...
The selection of suitable beam angles in external radiotherapy is at present generally based upon the experience human planner. requirement to automatically select particularly highlighted intensity-modulated radiation therapy (IMRT), which a smaller number modulated beams hoped be used, comparison with conformal radiotherapy. It has been proved by many researchers that most valuable for plan small (< or = 5). In this paper an efficient method presented investigate how improve dose...
Brightness and color are two basic visual features integrated by the human system (HVS) to gain a better understanding of natural scenes. Aiming combine these cues maximize reliability boundary detection in scenes, we propose new framework based on color-opponent mechanisms certain type color-sensitive double-opponent (DO) cells primary cortex (V1) HVS. This DO has oriented receptive field with both chromatically spatially opponent structure. The proposed is feedforward hierarchical model,...
This paper presents an effective nighttime vehicle detection system that combines a novel bioinspired image enhancement approach with weighted feature fusion technique. Inspired by the retinal mechanism in natural visual processing, we develop method modeling adaptive feedback from horizontal cells and center-surround antagonistic receptive fields of bipolar cells. Furthermore, extract features based on convolutional neural network, histogram oriented gradient, local binary pattern to train...
To effectively perform visual tasks like detecting contours, the system normally needs to integrate multiple features. Sufficient physiological studies have revealed that for a large number of neurons in primary cortex (V1) monkeys and cats, neuronal responses elicited by stimuli placed within classical receptive field (CRF) are substantially modulated, inhibited, when difference exists between CRF its surround, namely, non-CRF, various local The exquisite sensitivity V1 center-surround...
Nicotiflorin is a flavonoid extracted from Carthamus tinctorius. Previous studies have shown its cerebral protective effect, but the mechanism undefined. In this study, we aimed to determine whether nicotiflorin protects against ischemia/reperfusion injury-induced apoptosis through JAK2/STAT3 pathway. The injury model was established by middle artery occlusion/reperfusion. (10 mg/kg) administered tail vein injection. Cell in ischemic cortex examined hematoxylin-eosin staining and terminal...
Image enhancement is an important pre-processing step for many computer vision applications especially regarding the scenes in poor visibility conditions. In this work, we develop a unified two-pathway model inspired by biological vision, early visual mechanisms, which contributes to image tasks including low dynamic range (LDR) and high (HDR) tone mapping. Firstly, input separated sent into two pathways: structure-pathway detail-pathway, corresponding M- P-pathway system, code low-...
Abstract Metformin, an anti-diabetic drug commonly used for type 2 diabetes therapy, is associated with anti-angiogenic effects in conditions beyond diabetes. miR-21 has been reported to be involved the process of angiogenesis. However, precise regulatory mechanisms by which metformin-induced endothelial suppression and its on miR-21-dependent pathways are still unclear. Bioinformatic analysis identification targets their antiangiogenic activity were assessed using luciferase assays,...
The geometry of retinal layers is an important imaging feature for the diagnosis some ophthalmic diseases. In recent years, layer segmentation methods optical coherence tomography (OCT) images have emerged one after another, and huge progress has been achieved. However, challenges due to interference factors such as noise, blurring, fundus effusion, tissue artifacts remain in existing methods, primarily manifesting intra-layer false positives inter-layer boundary deviation. To solve these...