作者: Ying Yang , Bo Peng , Jianqin Wang
DOI: 10.1007/978-3-642-18333-1_16
关键词:
摘要: This paper presents a system for detection and recognition of pests in stored-grain based on video analysis. Unlike current systems which conduct analysis static images, the proposed uses data captured by camera performs to detect recognize grain. By using instead techniques such as motion estimation multiple-frame verification are used locate, count pests. Compared image processing, is more robust moving avoids missing re-counting Furthermore, analyzing video, can only living ignore dead ones, recommended national standard grain quality cannot be achieved processing.