自然场景下果实目标的识别和定位精选.pdfVIP

第35 卷第3 期 Vol. 35 No. 3 2007 年6 月 JOURNAL OF ZHEJIANG UNIVERSITY OF TECH NOLOGY Jun. 2007 2 , , , , ( . , 3 0032; 2. , 3 0032) : 通过分析目前果蔬图像中识别果蔬对象存在的问题, 研究了果蔬采摘机 人视觉中的两大难 题: 识别果蔬和定位果蔬. 在此大背景下, 研究提出了一种结合多种颜色特征和纹理特征进行分割 的方法, 有效解决果蔬目标和背景颜色差异较小时的果蔬对象识别问题. 同时, 提出一种全新的理 念解决果实被部分遮挡时中心点和采摘点确定问题. 另外, 为了克服已有的类球形果蔬的定位方法 不能适用于偏斜下垂生长的果蔬定位不能有效定位叠加的多 个果蔬的不足, 利用几何校正方法, 研究提供一种能够适应于偏斜下垂生长的果蔬定位的方法, 能有效定位叠加的多个果蔬. 经实验证 明, 效果良好. : 自然场景; 颜色; 纹理; 果实目标; 识别; 定位 : T P39 . 4 : A : 006-4303( 2007) 03-0267-07 The recognizing and locating method for fruit objects under nature scenes 2 GU Hui , LU Ya-ya , DING We-i long , WA NG Jie , ZHANG We-i tong ( . Colleg e of Inform ation Engineering, Zhejiang University of Technology, H angzhou 3 0032, China; 2. College of Mechanical Engineering, Zhejiang University of T echnology, H angzhou 3 0032, China) Abstract: T his paper discussed about the most difficult problems in the vision of harvesting ro- bot, including recognizing and locating problem existed in the recognition of the fruit object in fruit images. A new segmentation method combined w ith color features and texture features is presented, in w hich the recognition problem w ith little difference betw een w ith the target and the background is solved. It also put forw ard a novel conception to resolve the locating center and ab- scission point under the condition that the fruit or vegetables are partially sheltered. Meanw hile, in order to overcome the demerits of the going locating methods w hich unfits for the declining f