基于计算机视觉的胡萝卜外观品质分级系统与装备  

Quality Grading System and Equipment of Carrots Based on Computer Vision

作　　者：杜宏伟[1] 邓立苗[2,3] 熊凯[2] 徐艳[2] 韩仲志[2,3] 

机构地区：[1]青岛农业大学机电工程学院,山东青岛266109 [2]青岛农业大学理信学院,山东青岛266109 [3]中国石油大学(华东)地球科学学院,山东青岛266580

出　　处：《农机化研究》2015年第1期196-200,共5页Journal of Agricultural Mechanization Research

基　　金：国家自然科学基金项目(31201133);山东省自然科学基金项目(ZR2009DQ019);青岛市科技发展计划项目(11-2-3-20-nsh)

摘　　要：为了实现胡萝卜的自动化分级分选,在前期胡萝卜青头、须根、开裂检测算法研究的基础上,参照国家标准需求,进一步提出了胡萝卜外观等级规格,如弯曲、断折等缺陷,以及衡量异品种的锥形度参数的检测方法,并以此为基础设计了基于计算机视觉的胡萝卜外观品质实时分选生产线。该生产线由下位机控制系统、上位机软件系统和机械分级装置组成。通过对软硬件环境和机械系统性能的测试表明:该分选生产线每秒钟可检测20个胡萝卜的等级;对520个胡萝卜含有不同种类缺陷和异品种的胡萝卜进行分级检测,分级正确率达到了93.5%,能够满足胡萝卜外观品质分选实时检测的要求。该分选线能够同步实现胡萝卜等级规格、质量、各种缺陷的实时检测与在线分选。In order to achieve automatic classification of carrots by appearance quality, not long ago, we have been put forward kinds of algorithms which can detect the carrot green head, fibrous roots, crack on surface. Based on these work and the provisions of national standards, this paper raised some methods whichcan-measure the standard, detect the bend and break etc, also it studied a kind of method which can detect-othe^r kind carrot by measure the telescopicity. Based on these methods above and the computer vision theory, this paper established and developed a kind of production line which can separate different quality carrots. This separation production line composed of three parts. One is upper computer, which have integrated algorithms. The second is PLC which is the hardware control system. The last is the mechanical system （ILMS）. After these work, the environment of software ＆ hardware and the mechanical system has been tested. The results show that ： 1 ） the sorting line can detect 20 carrots per second, which can meet the real-time detection of car- rot by the appearance quality. 2 ） on 520 carrots contain a variety of defects and different varieties, and the correct rate of classification has reached 93.5%. The quality -time online detecting and sorting synchronously grading method and equipment constructed in this paper can realize real by grades and specifications, weight, various defects

关 键 词：胡萝卜 锥形度 图像检测 回归分析 在线检测 

分 类 号：S225.73[农业科学—农业机械化工程]