A5019289884- Software Engineering Research
- Smart Agriculture and AI
- Industrial Vision Systems and Defect Detection
- Educational and Technological Research
- Imbalanced Data Classification Techniques
- Food Supply Chain Traceability
- Plant Virus Research Studies
- Advanced Image and Video Retrieval Techniques
- Image Processing Techniques and Applications
- E-commerce and Technology Innovations
- Technology and Security Systems
- CCD and CMOS Imaging Sensors
- Online Learning and Analytics
- Machine Learning and Data Classification
- Educational Technology and Pedagogy
- Software Reliability and Analysis Research
- Plant Pathogenic Bacteria Studies
- Infrastructure Maintenance and Monitoring
- Medical Image Segmentation Techniques
- Integrated Circuits and Semiconductor Failure Analysis
Yunnan Agricultural University
2024
Harbin University of Science and Technology
2022-2024
Background: Given the severe economic burden that citrus greening disease imposes on fruit farmers and related industries, rapid accurate detection is particularly crucial. This not only effectively curbs spread of disease, but also significantly reduces reliance manual within extensive planting areas. Objective: In response to this challenge, address issues posed by resource-constrained platforms complex backgrounds, paper designs proposes a novel method for recognition localization named...
Background: Corn is the main grain crop grown in China, and ear shape index of corn an important parameter for breeding new varieties, including length, diameter, row number ears, grains per ear, so on. Objective: In order to solve problem limited field view associated with computer detection against a complex background, this paper proposes panoramic splicing method ears which can splice 10 images at same time, improve information collection efficiency, display comprehensive information,...
Software defect prediction is an important part of software development, which aims to use existing historical data predict future defects. Focusing on the model performance and communication efficiency cross-project prediction, this paper proposes efficient communication-based federated meta-learning (ECFML) algorithm. The lightweight MobileViT network used as meta-learner Model Agnostic Meta-Learning (MAML) By learning common knowledge local multiple clients, then fine-tuning model, number...
Heterogeneous defect prediction (HDP) constructs a model through the source project to realize tendency of target project. HDP based on federated learning can combine multi-party data improve effect under condition protecting privacy. However, with increase task complexity and performance requirements, number neural network layers gradually deepens, parameters also increase. accordingly. In scenario multiple parameters, limited communication bandwidth, clients, amount received by server is...
Software defect prediction is used to identify modules in software projects that may have defects. Heterogeneous Defect Prediction (HDP) establishes a cross project model based on different datasets. However, due the heterogeneity of multi-source data, performance usually not ideal. In addition, data holder unwilling disclose privacy regulations and other reasons, resulting islands. This paper presents federal prototype learning averaging (FPLPA), which combines federated (FL) with for...
With the rise of food safety traceability, unmanned supermarkets and autonomous shopping, automatic identification technology agricultural products such as vegetables in circulation sales has become an urgent problem. This paper designs intelligent vegetable system based on MobileNet to solve problem supermarkets.
 The includes main control core, visual processing module, pressure sensor, voice broadcasting module display module. When detects that there are be weighed, completes...