A5017598677- Advanced Neural Network Applications
- Image Processing and 3D Reconstruction
- Explainable Artificial Intelligence (XAI)
- Handwritten Text Recognition Techniques
- Image and Object Detection Techniques
- Brain Tumor Detection and Classification
- IoT and Edge/Fog Computing
- Medical Image Segmentation Techniques
- Higher Education and Employability
- Reservoir Engineering and Simulation Methods
- Hydraulic Fracturing and Reservoir Analysis
- Seismic Imaging and Inversion Techniques
- Digital Transformation in Industry
- Adversarial Robustness in Machine Learning
- Smart Agriculture and AI
China Agricultural University
2024
Chongqing University of Posts and Telecommunications
2023
China University of Petroleum, Beijing
2014
University of Hong Kong
1999
In this study, a deep-learning-based intelligent detection model was designed and implemented to rapidly detect cotton pests diseases. The integrates cutting-edge Transformer technology knowledge graphs, effectively enhancing pest disease feature recognition precision. With the application of edge computing technology, efficient data processing inference analysis on mobile platforms are facilitated. Experimental results indicate that proposed method achieved an accuracy rate 0.94, mean...
Proposes a method to tackle the problem of detecting skilled forgeries in off-line signature verification. Inspired by approach adopted expert examiners, it is based on smoothness criterion. From collection genuine and forged signatures, observed that, although forgery signatures are very similar ones global scale, they generally less smooth natural detailed scale than ones, especially for those which consist cursive graphic patterns. A index derived from such signatures. This combined with...
Due to the rapid societal development and impact of novel coronavirus, employment rate graduates from agricultural forestry colleges has been significantly affected. Considering specialized nature talent training in this field limited prospects, issue garnered substantial attention society. This paper aims address concern by utilizing Python crawl data pertaining "employment situation universities" major internet sources. The gathered text information is subsequently categorized using LDA...
Summary In this paper, a percentile-half-thresholding approach is proposed in the transformed domain thresholding process for iterative shrinkage (IST). The percentile-thresholding strategy more convenient implementing than constant-value, linear-decreasing, or exponential-decreasing because it’s data-driven. novel half-thresholding inspired from recent advancement researches on optimization using non-convex regularization. We summarize general framework IST and show that only difference...