项目名称: 基于机器视觉的果蝇行为分析技术研究
项目编号: No.61273281
项目类型: 面上项目
立项/批准年度: 2013
项目学科: 自动化技术、计算机技术
项目作者: 郑玻
作者单位: 安阳师范学院
项目金额: 81万元
中文摘要: 以果蝇为动物模型来系统地对动物行为的研究是生物学研究领域的重要方法,然而需要耗费大量的人力物力来观测果蝇的行为,自动果蝇行为数据获取与行为分析对生物学研究意义重大。本课题针对以果蝇为模型的,基于计算机视觉的动物行为学研究的难点和瓶颈,探索利用人工智能与计算机视觉技术进行客观快速准确鲁棒的以果蝇为模型的行为分析方法,主要包括:研究果蝇群体密度分析、高密集情况下果蝇运动多目标视觉跟踪、基于实例的果蝇运动行为识别研究、果蝇运动频繁行为模式挖掘等问题,并将以上研究形成的一些方法和技术进行应用验证,评价果蝇运动行为分析方法获取的信息的质量和系统的价值,提出具有独立知识产权的如何进行自动分析果蝇行为和行为模式挖掘的关键技术的算法,完成一个包含这些关键技术的原型系统,为自动果蝇行为分析提供技术准备和工具。
中文关键词: 行为分析;运动跟踪;半监督学习;;
英文摘要: Using Drosophila as an animal model to systematically study animal behavior is a very important approach to biological research, however, it requires a lot of manpower and resources to observe the behavior of the drosophila, How to automatically obtain Drosophila behavior data and to analyze Drosophila behavior is an open problem to be solved. This proposal focus on the study of Drosophila's behavior based on artificial intelligence and computer vision technology to provide fast, accurate and robust behavior analysis methods and tools for animal behavior research based on drosophila model, which include density estimation of the drosophila group, Multi-target visual tracking, Drosophila's behavior recognition, behavior patterns mining etc. We propose to design a system to allow high-throughput, quantitative analysis of behavior in freely moving flies based on the previous mentioned techniques. Our primary goal in this project is to provide quantitative behavior analysis tools to the biological research community. Lastly, we will work out a demo of drosophila behavior analysis.
英文关键词: behaviour analysis;tracking;semi-supervised learning;;