A5035352898- Remote Sensing and LiDAR Applications
- Animal Behavior and Welfare Studies
- Smart Agriculture and AI
- Food Supply Chain Traceability
- Advanced Nanomaterials in Catalysis
- Automated Road and Building Extraction
- Remote Sensing and Land Use
- Wood and Agarwood Research
- Remote Sensing in Agriculture
- Advanced Photocatalysis Techniques
- Effects of Environmental Stressors on Livestock
- Meat and Animal Product Quality
- Wood Treatment and Properties
- Leaf Properties and Growth Measurement
- Quantum Dots Synthesis And Properties
- Generative Adversarial Networks and Image Synthesis
- Industrial Vision Systems and Defect Detection
- Advances in Cucurbitaceae Research
- Forensic Fingerprint Detection Methods
- 3D Surveying and Cultural Heritage
- Rough Sets and Fuzzy Logic
- Advanced biosensing and bioanalysis techniques
- Botanical Research and Chemistry
- Advanced Vision and Imaging
- Date Palm Research Studies
Zhejiang A & F University
2016-2024
State Forestry and Grassland Administration
2019-2024
East China University of Science and Technology
2022-2023
Accurate plant leaf image segmentation provides an effective basis for automatic area estimation, species identification, and disease pest monitoring. In this paper, based on our previous publicly available dataset, approach that fuses YOLOv8 improved DeepLabv3+ is proposed precise of individual leaves. First, the object detection algorithm-based was introduced to reduce interference backgrounds second stage task. Then, method more efficiently capture bar leaves slender petioles. Densely...
Pig tracking contributes to the assessment of pig behaviour and health. However, on real farms is very difficult. Owing incomplete camera field view (FOV), pigs frequently entering exiting FOV affect accuracy. To improve pig-tracking efficiency, we propose a method that based skeleton feature similarity, which named GcnTrack. We used YOLOv7-Pose extract key points design dual-tracking strategy. This strategy combines IOU matching keypoint-based graph convolutional reidentification (Re-ID)...
For industrial enzymatic catalysis, efficient regeneration of nicotinamide adenine dinucleotide (NADH) is necessary. Herein, hollow resorcinol-formaldehyde resin nanospheres loaded with Au nanoparticles (HRF-Au) shape-controllable nanostructures for photocatalysis were successfully prepared via the hard template method and photodeposition. As a novel photocatalyst, HRF-Au efficiently achieved high NADH yield 80.39% total reaction rate 0.6 mmol·gcatalyst–1·h–1 under visible light...
In urban forest management, individual street tree segmentation is a fundamental method to obtain phenotypes, which especially critical. Most existing image models have been evaluated on smaller datasets and lack experimental verification larger, publicly available datasets. Therefore, this paper, based large, dataset, proposes YOLO-SegNet for segmentation. the first stage of object detection task, BiFormer attention mechanism was introduced into YOLOv8 network increase contextual...
Lumber pieces usually contain defects such as knots, which strongly affect the strength and stiffness. To develop a model for rapid, accurate grading of lumbers based on Douglas fir, spruce-pine-fir (SPF), Chinese hemlock, Dragon spruce were used. The experiments explored effects modelling methods spectral preprocess knot detection, investigated feasibility using built within one species to discriminate samples from other species, novel variable selection method-random frog (RF)- select...
Continuous movement monitoring helps quickly identify pig abnormalities, enabling immediate action to enhance welfare. However, continuous and precise of daily on farms remains challenging. We present an approach automatically precisely monitor the group-housed pigs. The instance segmentation model YOLOv8m-seg was applied detect presence then a spatial moment algorithm quantitatively summarize each detected pig's contour as corresponding center point. agglomerative clustering (AC)...
In automated, intelligent harvesting tasks on grape farms, reliable methods are needed to accurately locate bunches of stems in complex contexts. On-farm environments with stacked clusters and interlocking branches make it more difficult unstructured environments. this study, we propose a top-down stem localization method based the characteristics growth. The is human pose estimation uses target detection algorithm identify apply cluster keypoint model localize clusters. For purpose, use...
Tree species identification and tree organ segmentation using images are challenging problems that useful in many forestry-related tasks. Despite researchers’ efforts data collection, there still few datasets covering a wide range of with detailed annotations. In this paper, the urban street dataset is proposed as comprehensive, publicly available 50 contains 41467 high-resolution classification (22872 annotated images) from 10 city scenes. Our includes leaf, tree, trunk, branch trunk...
To solve problems of lacking adequate evidences and real-time monitoring in the process sewage disposal, improve efficiency disposal emergency response capability pig farms, we presented an exploratory smart strategy, model Ecological analysis method on hog under ecological treatment mode. Based principle mode, research proposed to use multiple sensors monitor collect data conduct qualitative quantitative with mathematical method. Using method, this paper analyzed built a for procedures...