A5074531822- Smart Agriculture and AI
- Terahertz technology and applications
- Network Security and Intrusion Detection
- Spectroscopy and Chemometric Analyses
- Animal Behavior and Reproduction
- Advanced Chemical Sensor Technologies
- Silkworms and Sericulture Research
- Image and Signal Denoising Methods
- Image Processing Techniques and Applications
- Remote Sensing and Land Use
- Anomaly Detection Techniques and Applications
- Advanced Image Processing Techniques
- Wildlife Ecology and Conservation
- Time Series Analysis and Forecasting
East China Jiaotong University
2023-2025
Silkworm pupae (SP), the pupal stage of an edible insect, have strong potential in food, medicine, and cosmetic industries. Sex sorting is essential to enhance nutritional content genetic traits SP crossbreeding but it remains labor intensive time consuming. An intelligent method needed urgently improve efficiency productivity. To address problem, automatic sex-separation system was developed based on computer vision deep learning. Specifically, gonad features, a novel real-time sex...
During the process of automatic fruits picking by robot in orchard, movements invariably bring shakes camera, resulting spatially-varying motion blurs that may impair accuracy fruit detection and localization. Therefore, this research proposes a blurred image restoration approach based on an end-to-end multi-scale conditional generative adversarial network. Firstly, adopting residual module, it effectively improves network's ability to extract features reduce network parameters quantity....
Intending to solve the issues of similar color and background, uneven illumination, occlusion between fruit leaf during citrus detection under near-color this paper proposes a model by combing improved YOLOv5 data enhancement strategy. The image datasets are first preprocessed using algorithm expand datasets. Then, is trained on augmented dataset acquire weights. Finally, comparing different algorithms network, random erasure well-trained adopted. results show that presented method reaches...
Abstract A feasibility study on variety discrimination was organized for 12 wheat varieties before warehousing by applying the Terahertz time-domain spectroscopy (THz-TDS) technology in combination with convolutional neural network (CNN) considering time-consuming and complex traditional methods of that require great number pretreatment processes. In experiment, original spectra (TDS) range 0.1~2.0THz were acquired, frequency domain (FDS), refractive index absorption coefficient 0.1~1.4THz...
Leaf disease diagnosis is very critical for tea due to its importance increasing crop production. Tea leaf images pose unique challenges compared other types of their characteristics, including randomly distributed lesions, diverse symptoms, and complex backgrounds. As a result, capturing discriminative information from these becomes difficult. In this article, we propose effective model the by combing MobileViT deblurring preprocessing module. The composed with regular convolutions, MV2...