A5101992218- Smart Agriculture and AI
- Spectroscopy and Chemometric Analyses
- Minerals Flotation and Separation Techniques
- Advanced Control Systems Optimization
- Plant Disease Management Techniques
- Mineral Processing and Grinding
- Remote Sensing in Agriculture
- Water Quality Monitoring and Analysis
- Fault Detection and Control Systems
- Remote Sensing and Land Use
- Industrial Vision Systems and Defect Detection
- Adaptive Dynamic Programming Control
- Water Quality Monitoring Technologies
- Infrastructure Maintenance and Monitoring
- Leaf Properties and Growth Measurement
- Asphalt Pavement Performance Evaluation
- Plant and animal studies
- Intracerebral and Subarachnoid Hemorrhage Research
- Species Distribution and Climate Change
- Date Palm Research Studies
- Plant Virus Research Studies
- Remote Sensing and LiDAR Applications
- Structural Health Monitoring Techniques
- Advanced Algorithms and Applications
- Digital Imaging for Blood Diseases
Central South University of Forestry and Technology
2018-2025
Central South University
2015-2025
Taiyuan Central Hospital
2024
Shanghai Guanghua Hospital of Integrated Traditional Chinese and Western Medicine
2024
Sichuan University
2023
West China Hospital of Sichuan University
2023
Nanjing Tech University
2017-2022
New York University
2015
In this paper, a method for detecting rapid rice disease based on FCM-KM and Faster R-CNN fusion is proposed to address various problems with the images, such as noise, blurred image edge, large background interference low detection accuracy. Firstly, uses two-dimensional filtering mask combined weighted multilevel median filter (2DFM-AMMF) noise reduction, faster Otsu threshold segmentation algorithm (Faster 2D-Otsu) reduce of complex target blade in image. Then dynamic population firefly...
The identification of maize leaf diseases will meet great challenges because the difficulties in extracting lesion features from constant-changing environment, uneven illumination reflection incident light source and many other factors. In this paper, a novel disease recognition method is proposed. method, we first designed feature enhancement framework with capability enhancing under complex environment. Then neural network based on backbone Alexnet architecture, named DMS-Robust Alexnet....
Abstract Road cracks pose a serious threat to the stability of road structures and traffic safety. Therefore, this paper proposes an optimized accurate crack segmentation network called MBGBNet, which can solve problems complex background, tiny cracks, irregular edges in segmentation. First, multi‐scale domain feature aggregation is proposed address interference background. Second, bidirectional embedding fusion adaptive attention capture features finally, Gaussian weighted edge algorithm...
Maize, as one of the most important crops in world, faces severe challenges from various diseases and pests. The timely accurate identification maize leaf pests is great significance for ensuring agricultural production. Currently, two key challenges: (1) In actual process identifying pests, complex backgrounds can interfere with effect. (2) subtle features are difficult to accurately extract. To address these challenges, this study proposes a disease pest model called LFMNet. Firstly,...
Aiming at the problems of noise, background interference and low detection in peach disease image, this paper proposes a method based on asymptotic non-local means (ANLM) image algorithm fusion parallel convolution neural network (PCNN) extreme learning machine(ELM) optimized by linear particle swarm optimization(IPSO). Firstly, uses ANLM denoising to reduce complex then proposed identify characteristics disease, improved elu activation function instead conventional ReLu function, ELM...
Precise disease detection is crucial in modern precision agriculture, especially ensuring the health of tomato crops and enhancing agricultural productivity product quality. Although most existing methods have helped growers identify leaf diseases to some extent, these typically target fixed categories. When faced with new diseases, extensive costly manual annotation required retrain dataset. To overcome limitations, this study proposes a multimodal model PDC-VLD based on open-vocabulary...
In existing machine vision technology for fruit defects, the hue appears different, and defect area is small due to irregularity of illumination reflection from surface incident light source, this makes it difficult extract area. Thus, we proposed an apple detection method based on Fuzzy C-means Algorithm Nonlinear Programming Genetic (FCM-NPGA) in combination with a multivariate image analysis. First, was denoised enhanced through fractional differentiation. The noise points edge were...
In deep learning-based maize leaf disease detection, a identification method called Network based on wavelet threshold-guided bilateral filtering, multi-channel ResNet, and attenuation factor (WG-MARNet) is proposed. This can solve the problems of noise, background interference, low detection accuracy images. To begin, processing layer Wavelet threshold guided filtering (WT-GBF) WG-MARNet model employed to reduce image noise perform high low-frequency decomposition input using WT-GBF....
Pepper is a high-economic-value agricultural crop that faces diverse disease challenges such as blight and anthracnose. These diseases not only reduce the yield of pepper but, in severe cases, can also cause significant economic losses threaten food security. The timely accurate identification crucial. Image recognition technology plays key role this aspect by automating efficiently identifying diseases, helping workers to adopt implement effective control strategies, alleviating impact...
In this study, we propose a dynamics-learning multirate estimation approach to perceive the quality-related indices (QRIs) of feeding solution unit process. A index for is an intermediate technical indicator between process and proceeding process; hence, problem formulated as two-stage utilizing production data both processes. Dynamics-learning bidirectional long short-term memory (BiLSTM) with different inputs forward backward layers proposed manage input from BiLSTM, cycle control gate...