鲜香菇外部品质的视觉分选技术与图形缺陷检测  

Visual Sorting Technology and Graphic Defect Detection of External Quality of Fresh Lentinus edodes

作　　者：谢茜 XIE Qian(Nanyang Institute of Technology,Digital media and art Design College,Nanyang 473000,China)

机构地区：[1]南阳理工学院数字媒体与艺术设计学院,河南南阳473000

出　　处：《中国食用菌》2020年第12期153-156,共4页Edible Fungi of China

基　　金：2019年河南省教育科学规划领导小组办公室项目(2019-JKJXA-234)。

摘　　要：鲜香菇的外部品质和等级划分是影响我国香菇产业发展的重要因素,人工检测分级效率低、精度差。采用计算机视觉和图像处理技术,对鲜香菇的图像进行采集和预处理,使用图像灰度化、去噪和图像分割技术,对香菇图像进行菌盖直径、厚度和颜色的特征提取。依据鲜香菇分级标准完成香菇的缺陷检测和外观品质分级。该方法适用于多种香菇的缺陷检测,提高了香菇缺陷检测的速度和分级准确性。The external quality and grade classification of fresh Lentinus edodes are important factors affecting the development of L.edodes industry in China.The manual detection and grading efficiency is low and the accuracy is poor.Computer vision and image processing technology are used to collect and preprocess the image of fresh L.edodes.The image graying,denoising and image segmentation techniques are used to extract the characteristics of the pileus diameter,thickness and color.According to the grading standard of fresh L.edodes,the defect detection and appearance quality grading of L.edodes were completed.This method is suitable for a variety situations of Lentinus edodes defect detection,and improves the speed and accuracy of Lentinus edodes defect detection.

关 键 词：鲜香菇 图像处理 缺陷检测 

分 类 号：S646.9[农业科学—蔬菜学] TP391[农业科学—园艺学]