Research on the Detected Technology of Cottonseeds Based on Image Recognition

摘要：

The paper is to achieve the detection to red cottonseeds and crashed cottonseeds of Xinjiang, built up inspection and grading system of red cottonseeds and crashed cottonseeds based on the image recognition technology. After took analysis and pretreatment to the image of cottonseeds, color model for red cottonseeds were obtained—HIS color model. Distilled the key character parameters of cottonseeds: proportion, perimeter, circularity, major and minor axis, Chose circularity as the detecting character of crashed cottonseeds based on the analysis of these character parameters. Selected 200 cottonseeds of Xinluzao-26＃at random for the experiment, the detecting precision of red cottonseeds had achieved 93.5％, and the detecting precision of crashed cottonseeds had achieved 85.7％.

关键词： image recognition color model marginal detection cottonseeds

作者: Li Jing-bin Zhu Jiang-li Shao Lu-hao Tmn Xu-shun Kan Za

作者单位: College of Mechanical and Electric Engineering Shihezi University Shihezi. China College of Mechanical and Electric Engineering Shihezi University Shihezi Engineering College Shihez College of Mechanical and Electric Engineering Shihezi University Shihezi, China College of Mechanical and Electric Engineering, Shihezi University, Shihezi. 83200, China Shihezi En

会议类型: 国际会议

会议名称: 2011 International Conference on Agricultural Engineering New Technology & Taishan Academic Forum(2011农业工程新技术国际学术会议暨泰山学术论坛 ICAE2011&TSAF)

会议地点: 山东淄博

会议语种:英文

页码: 176-179

在线出版日期: 2011-05-27（万方平台首次上网日期，不代表论文的发表时间）