A5031332555- Spectroscopy and Chemometric Analyses
- Animal Behavior and Welfare Studies
- Effects of Environmental Stressors on Livestock
- Computer Graphics and Visualization Techniques
- Advanced Measurement and Detection Methods
- Medical Imaging Techniques and Applications
- Smart Agriculture and AI
- Radiomics and Machine Learning in Medical Imaging
- Fire Detection and Safety Systems
- Agriculture and Farm Safety
- Industrial Vision Systems and Defect Detection
- Leaf Properties and Growth Measurement
- Advanced X-ray and CT Imaging
- Advanced Image and Video Retrieval Techniques
- 3D Surveying and Cultural Heritage
- Advanced Neural Network Applications
- Advanced Chemical Sensor Technologies
- Human-Animal Interaction Studies
- Robotics and Sensor-Based Localization
- Water Quality Monitoring and Analysis
Inner Mongolia Agricultural University
2018-2023
Xijing Hospital
2023
Crop pests and diseases are one of the most critical disasters that limit agricultural production. In this paper, we trained a lightweight convolutional neural network model built Django framework-based potato disease leaf recognition system, which can recognize three types images including early blight, late healthy. A lightweight, network-based for identification significantly reduces number parameters, whereas accuracy Top-1 is over 93%. We imported into framework to build website thus...
Abstract Starch is an important quality index in potato, which contributes greatly to the taste and nutritional of potato. At present, determination starch depends on chemical analysis, time consuming laborious. Thus, rapid accurate detection content potatoes important. This study combined hyperspectral imaging with chemometrics predict potato content. Two varieties Kexin No.1 Holland No.15 were used as experimental samples. Hyperspectral data collected from three sampling sites (the top,...
In the video monitoring of piglets in pig farms, study precise segmentation foreground objects is work advanced research on target tracking and behavior recognition. view noninteractive real-time requirements such a system, this paper proposes method image based an improved GrabCut algorithm. The functions preserving edges noise reduction are realized through bilateral filtering. An adaptive threshold used to calculate local complete extraction target. simplified by morphological processing;...
In recent years, traditional farming methods have been increasingly replaced by more modern, intelligent techniques. This shift towards information and intelligence in is becoming a trend. When they are bitten dinoflagellates, cows display stress behaviors, including tail wagging, head tossing, leg kicking, ear flapping, skin fluttering. The study of cow protective behavior can indirectly reveal the health status their living patterns under different environmental conditions, allowing for...
The study of the self-protective behaviors dairy cows suffering dipteral insect infestation is important for evaluating breeding environment and cows’ selective breeding. current practices measuring diary are mostly by human observation, which not only tedious but also inefficient inaccurate. In this paper, we develop an automatic monitoring system based on video analysis. First, improved optical flow tracking algorithm Shi-Tomasi corner detection presented. By combining morphological...
In the process of upgrading electric power industry from new infrastructure cultivation to development dynamic energy in enterprises, safety and security work staff will also face challenges. The demand capacity increased intensity site operators, some were paralyzed without wearing helmets, which posed serious hazards. use video detection technology based on target can effectively monitor whether personnel at wear further reduce risks accidents protect personal operators. this paper,...
Various approaches for CT-less PET acquisition have been proposed and evaluated in the standalone scanners or low-dose clinical applications. parametric imaging has yet investigated poses challenges terms of both attenuation correction estimation image-derived input function (IDIF). In this work, we a deep learning-based solution with IDIF. A synthesized map from non-attenuation-corrected (NAC) was used dynamic image reconstruction, as well segmentation aorta mask Clinical static <sup...
We describe and demonstrate an algorithm for 3D Reconstruction that output a beautiful model with textures. Taking set of images as input, we detect features from by SIFT, estimate camera parameters points positions Bundle Adjustment, use PMVS to acquire point cloud. can accurate object scene models fine surface detail. Then, propose novel cheap triangulation method suited the cloud generated PMVS. Our make full advantage BFS(Breadth-First Search), choosing constraints in optical condition....