Tomato Recognition and Identification of Growth States based on Shape and Texture Feature Analysis
In order to guide robot for harvesting mature tomato, an automatic tomato recognition and identification of growth states based on shape and texture features analysis was developed. On the basis of morphological processes of binary image generated from image segmentation, child images of tomato and disturbed objects are extracted from 80 tomato images captured under outdoor scene Three shape parameters and five texture parameters are extracted respectively from child images and statistically analyzed according to the instances that growth states of tomato are separated, connected, overlapped and partially covered by branches and leaves of plant Finally flow of tomato recognition and identification of growth states are determined By processing of other 674 images, the tests show that right recognition ratio of harvesting mature tomato is more than 90 percent except that tomatoes are submerged in flaky withered leaves and image segmentation method, cant separate them .and single tomato is divided into several parts by branches and leaves of plant
Agricultural robot Image recognition Feature extraction
YIN Jianjun XIE Yongliang
Key Laboratory of Modern Agricultural Equipment and Technology, Ministry of Education & Jiangsu Prov ZheJiang Technology Institute of Economy, Hangzhou, Zhejiang Province, China, 310018
国际会议
2008 International Conference on IAEA(2008农业信息化、自动化与电气化国际会议)
镇江
英文
554-558
2008-11-01（万方平台首次上网日期，不代表论文的发表时间）