- Industrial Vision Systems and Defect Detection
- Remote Sensing and LiDAR Applications
- Wood and Agarwood Research
- Retinal Diseases and Treatments
- Retinal Imaging and Analysis
- Educational Games and Gamification
- Hydrocarbon exploration and reservoir analysis
- Pancreatic and Hepatic Oncology Research
- Photorefractive and Nonlinear Optics
- Gastrointestinal Tumor Research and Treatment
- Gastrointestinal disorders and treatments
- Minimally Invasive Surgical Techniques
- Retinal and Macular Surgery
- Infrastructure Maintenance and Monitoring
- Enhanced Oil Recovery Techniques
- Glaucoma and retinal disorders
- Solid State Laser Technologies
- Advanced Fiber Laser Technologies
- Teaching and Learning Programming
- MicroRNA in disease regulation
- Atmospheric aerosols and clouds
- Alcoholism and Thiamine Deficiency
- Extracellular vesicles in disease
- Advanced Graph Neural Networks
- Date Palm Research Studies
Guangxi Normal University
2024-2025
First People's Hospital of Yunnan Province
2018-2024
Kunming University of Science and Technology
2018-2024
Hechi University
2024
Guangdong Academy of Medical Sciences
2022-2024
China National Petroleum Corporation (China)
2023-2024
Guangdong Provincial People's Hospital
2022-2024
Beijing Agricultural Machinery Research Institute
2024
Southern Medical University
2022-2023
Anhui Polytechnic University
2023
The detection of wood defect is a crucial step in processing and manufacturing, determining the quality reliability products. To achieve accurate detection, novel method named BPN-YOLO proposed. ordinary convolution ELAN module YOLOv7 backbone network replaced with Pconv partial convolution, resulting P-ELAN module. Wood performance improved by this modification while unnecessary redundant computations memory accesses are reduced. Additionally, Biformer attention mechanism introduced to more...
Wood surface defect detection is a critical step in wood processing and manufacturing. To address the performance degradation caused by small targets multi-scale features detection, novel deep learning model proposed this study, FDD-YOLO, specifically designed for task. In feature extraction stage, C2f module funnel attention (FA) mechanisms are integrated into design of C2f-FA to enhance model’s ability extract defects various sizes. Additionally, Dual Spatial Pyramid Pooling-Fast (DSPPF)...
The underrepresentation within computer science of women, domestic students color, and with lower socioeconomic-status remains a national issue. Recent studies demonstrate two critical factors: Persistent stereotypes about "who does science" can preclude interest in the field for members these groups; many also perceive computing as "irrelevant" "asocial". While issues must be addressed at multiple ages levels, suggest that we should start early, before have developed stereotypes. As step...
Colorectal cancer (CRC) is one of the most common diagnoses. Histone deacetylase (HDAC) overactivity in CRC could promote progression. HDAC1, a member HDAC family, found aberrantly expressed CRC, but it remains unclear whether expression HDAC1 can be regulated by microRNA. In present study, we confirmed overexpression status tissues and cell lines, its proliferation invasion vitro. We saw that was direct target gene miR-761 bioinformatic luciferase reporter analyses. negatively correlated...
<title>Abstract</title> Wood surface defect detection technology offers the advantages of being non-destructive, rapid, accurate, and economical. It plays a crucial role in wood grade sorting, detection, improving quality sawn timber, accelerating automation processing. Currently, there are challenges accurately identifying multi-scale defects insufficient overall accuracy field detection. To address these issues, new model named DRR-YOLO is proposed this study. This combines DWR module DRB...
Quasi‐phase matching (QPM) is a technique in nonlinear optics for achieving efficient energy exchange among optical waves at different frequencies, by spatially modulating the quadratic nonlinearity ( χ (2) ) of medium. To realize full potential QPM, 3D spatial modulation required. This has become experimentally feasible recently thanks to invention femtosecond laser‐based engineering ferroelectric crystals. Herein, first experimental demonstration QPM second harmonic generation (SHG) cubic...
<title>Abstract</title> Using deep learning methods is a promising approach to improving bark removal efficiency and enhancing the quality of wood products. However, lack publicly available datasets for plate segmentation in processing poses challenges researchers this field. To address issue, benchmark named WPS-dataset proposed study, which consists 4863 images. We designed an image acquisition device assembled it on equipment capture images real industrial settings. evaluated using six...
Accurate detection of wood surface defects plays a pivotal role in enhancing grade sorting precision, maintaining high standards processing quality, and safeguarding forest resources. This paper introduces an efficient precise approach to detecting defects, building upon enhancements the YOLOv8 model, which demonstrates significant performance handling multi-scale small-target commonly found wood. The proposed method incorporates dilation-wise residual (DWR) module trunk deformable large...
Wood surface defect detection is a challenging task due to the complexity and variability of types. To address these challenges, this paper introduces novel deep learning approach named SiM-YOLO, which built upon YOLOv8 object framework. A fine-grained convolutional structure, SPD-Conv, introduced with aim preserving detailed information during feature extraction process, thus enabling model capture subtle variations complex details wood defects. In fusion stage, SiAFF-PANet-based module...
Thirty-six, fifty-two, and seventy-nine MAs showed no, mild, severe leakage on FA, respectively. Most (61.7%) were centered in the inner nuclear layer. Cystoid spaces observed adjacent to 60 (35.9%) MAs. with had a statistically higher flow proportion compared no or mild (both P < 0.001). Only 112 (67.1%) visualized OCTA en face images, while 165 (98.8%) could be OCT images. The location of did not associate significantly FA status. presence nearby cystoid by B-scan overlay correlated...
Over the past decade, politicians, leaders, and pundits have called for computing computer science education opportunities to be made available earlier earlier. Such calls led creation of a wide variety offerings students at middle-school even elementary levels, including summer "code camps" targeted students. camps often emphasize fun aspects computing, such as games robots. In contrast, research collegiate level suggests that meaningful applications social good, are more successful...
To investigate body fluid status in diabetic macular edema (DME) patients and the extent to which it is affected by renal function. One hundred thirty-two eyes from 132 with diabetes mellitus (DM) were prospectively collected this cross-sectional, observational study. Thirty-five DM without retinopathy (DR), 31 DR DME, 66 DME patients. The of each participant was quantified extracellular water-to-total water ratio (ECW/TBW) using a composition monitor. Central subfield thickness (CST) volume...
Objectives To explore the possible role of peripheral lesions (PLs) detected by ultrawide field (UWF) imaging system on central neurovascular structure and retinal function. Methods Ninety-seven diabetic patients were included in this cross-sectional study using UWF pseudocolour colour with Optos Daytona (Optos, PLC). images graded as predominantly (PPLs) without PPL. Macular alterations function measured optical coherence tomography angiography (OCTA) RETeval device, respectively. Central...
This article has been retracted. A retraction notice can be found at https://doi.org/10.3233/JIFS-219433.
Using deep learning methods is a promising approach to improving bark removal efficiency and enhancing the quality of wood products. However, lack publicly available datasets for plate segmentation in processing poses challenges researchers this field. To address issue, benchmark named WPS-dataset proposed study, which consists 4863 images. We designed an image acquisition device assembled it on equipment capture images real industrial settings. evaluated using six typical models. The models...
Attempting to apply deep learning methods wood panels bark removal equipment enhance the quality and efficiency of is a significant challenging endeavor. This study develops tests learning-based equipment. In accordance with practical requirements sawmills, equipped vision inspection system designed. Based on substantial collection panel images obtained using visual system, first general semantic segmentation dataset constructed for training BiSeNetV1 model employed in this study....
Wood plate bark removal processing is critical for ensuring the quality of wood and its products. To address issue lack datasets available application deep learning methods to this field, fill research gap in field equipment, a benchmark segmentation proposed study. Firstly, costumed image acquisition device designed assembled on equipment capture images real industrial settings. After data filtering, enhancement, annotation, recording, partitioning, dataset named WPS-dataset containing 4863...
Fault diagnosis is of great significance for industrial equipment maintenance, and feature extraction a key step the entire scheme. The symbolic aggregate approximation (SAX) popular approach with potential recently. In spite achievements SAX has made, adverse information aliasing still exists in its calculation procedure, it may make fail to guarantee correctness. This work focuses on analyzing phenomenon SAX, followed by developing novel alternative method, i.e. parallel (PSAX). proposed...