基于近红外图像的温室小型西瓜采摘信息获取技术.pdfVIP

20 12 年7 月 农 业 机 械 学 报 第43 卷 第7 期 DOI:10.6041/ j.issn.1000鄄1298.2012.07.032 基于近红外图像的温室小型西瓜采摘信息获取技术* 袁摇 挺摇 纪摇 超摇 张震华摇 张俊雄摇 谭豫之摇 李摇 伟 (中国农业大学工学院,北京 100083) 摇 摇 揖摘要铱摇 为实现温室立体栽培模式下小型西瓜的识别与空间定位,研究了基于近红外图像的西瓜采摘信息获 取方法。 测定、比较西瓜果实与茎、叶的光谱反射率,确定波长850nm 附近波段为区分西瓜与背景的最佳波段,在 光强差异较大的两时段内采集了最佳波段下的西瓜近红外图像;通过Otsu算法滤除背景信息,利用“米冶字型模板 检测得到“浓缩西瓜冶区域,实现西瓜果实识别;使用形心坐标计算公式获得采摘点坐标;根据西瓜果梗生长特性, 利用分块定位算法获得切割点坐标信息。 在温室环境下随机选择拍摄50 幅有西瓜图像和20 幅无西瓜图像进行 识别算法验证,并对识别成功的有西瓜图像进行采摘点与切割点提取算法验证。 结果表明,有西瓜图像识别成功 率为86%,无西瓜图像为95%;采摘点、切割点定位准确度分别为93郾0%、88郾4%。 关键词:西瓜摇 采摘摇 近红外图像摇 信息获取 中图分类号:TP391郾4 文献标识码:A 文章编号:1000鄄1298(2012)07鄄0174鄄05 Information Acquisition Technique of Mini鄄watermelon for Harvesting Based on Near鄄infrared Image in Greenhouse Yuan Ting摇 Ji Chao摇 Zhang Zhenhua摇 Zhang Junxiong摇 Tan Yuzhi摇 Li Wei (College of Engineering,ChinaAgricultural University,Beijing 100083,China) Abstract In order to realize the recognition and localization of the mini鄄watermelon with stereoscopic cultivation in greenhouse,a machine vision method for acquiring harvesting information of watermelon based on the near鄄infrared spectral image was presented. By comparing the spectral reflectance of fruit, leaf and stem,a wavelength of about 850 nm was chosen as the best wavelength,of which the images taken at different illumination conditions were tested for fruit recognition. At first,the Otsu threshold algorithm was adopted to eliminate most background information. Then,a template liked circlewasused to detect fruitregion andreducethenoises. Thirdly,accordingtothemorphologicalfeature,thecentroid of fruit was considered as the optimum point for picking and the cutting point was judged by “block鄄