摘要
根据苹果光学反射特性建立了一套适于苹果坏损自动检测的计算机图像系统。鉴于坏损出现位置、大小等因素的不可预测性，提出了一种基于知识的坏损点单调检测方法以及坏损区域判别方法。试验表明：系统有较高的坏损检出率，且能有效地消除果梗区和花萼区对坏损区域判别的影响；检测方法具有较强的鲁棒性。
On the optical reflectance properties of apples a computer image system for automatic detection of defects on apples is built. As the position and size of the defects can not be forecasted, a method based on the knowledge to detect defects pixel monotonically and a method to judge the defect region are proposed. Tests prove that the system has high occurrence of detecting defects and can effectively eliminate the effect of the stalk and calyx. The detecting way has strong robustness.
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
1998年第4期81-86,共6页
Transactions of the Chinese Society for Agricultural Machinery
关键词
计算机
图像分析
检测
苹果
Computer, Image analysis, Detection, Apple
