非结构环境下成熟草莓分割方法  

Method of ripe strawberry segmentation in unstructured environment

作　　者：邢婧瑜 解迎刚[1] 高博斌 XING Jingyu;XIE Yinggang;GAO Bobin(School of Information&Communication Engineering,Beijing Information Science&Technology University,Beijing 100101,China;School of Metallurgy and Energy Engineering,Kunming University of Science and Technology,Kunming 650093,China)

机构地区：[1]北京信息科技大学信息与通信工程,北京100101 [2]昆明理工大学冶金与能源工程,云南昆明650093

出　　处：《现代电子技术》2022年第5期36-41,共6页Modern Electronics Technique

基　　金：北京市自然科学基金面上项目(4192023);北京市自然科学基金面上项目(4202024);北京信息科技大学“勤信人才”培育计划资助项目(QXTCPC201704)。

摘　　要：针对目前在非结构环境下草莓识别准确率较低的问题,提出一种改进的Otsu阈值分割与分水岭算法相结合的成熟草莓图像分割算法。首先,将采集的图像进行双边滤波去除噪声,其次利用果实色差的关系,将RGB颜色空间转换成YCrCb颜色空间并提取Cr分量。通过改进的Otsu阈值分割算法对草莓图像进行初次分割,针对连通区域的草莓目标利用分水岭分割算法进行二次分割。实验结果表明,改进的Otsu分割算法在复杂场景下分割成熟草莓果实的正确率为90%,较传统Otsu分割算法提高了15%。分水岭算法正确分割重叠、遮挡草莓目标的分割率为85%,为成熟草莓在非结构环境下精准识别提供了理论基础。In allusion to the low recognition accuracy rate of strawberry in unstructured environment,a ripe strawberry image segmentation algorithm combining improved Otsu threshold segmentation and watershed algorithm is proposed. The collected images are bilaterally filtered to remove noise. According to the relationship of fruit color difference,the RGB color space is converted into YCrCb color space,and the Cr component is extracted. The improved Otsu threshold segmentation algorithm is used for the elementary segmentation of strawberry images,and the watershed segmentation algorithm is used for the secondary segmentation of the objective strawberries in the connected region. The experimental results show that the improved Otsu segmentation algorithm has an accuracy rate of 90% in segmenting ripe strawberry fruits in complex scenes,which is 15% higher than that of the traditional Otsu segmentation algorithm. The watershed algorithm has an accuracy rate of 85% in segmenting the overlapping and occluded objective strawberries,which provides a theoretical basis for accurate recognition of ripe strawberries in unstructured environments.

关 键 词：草莓分割 图像分割 图像去噪 双边滤波 颜色空间转换 Cr分量提取 

分 类 号：TN911.73-34[电子电信—通信与信息系统] TP391.4[电子电信—信息与通信工程]