鲜香菇外部品质计算机视觉检测与分级研究

根据鲜香菇图像特点和分级标准,运用计算机视觉技术和神经网络算法对香菇进行自动检测与分级。采用掩模去背景、中值滤波、边缘亮度补偿等技术对图像进行处理。选取香菇菇盖最大直径、圆形度、色调均值及缺陷区域总面积与香菇图像总面积的比值作为鲜香菇分级的特征参数。通过BP神经网络建立了特征参数与鲜香菇等级之间的关系模型,试验结果表明,其预测识别结果达到94.2%。

Detection and Gradeing on Exterior Quality of Fresh Entinus Edodes Based on Computer

According to image characteristics and grading standards of fresh entinus edodes, computer vision technology and neural network is used to realize automatic detection and grading. The mask background segmentation, median filtering and edge light compensation method are applied to process images. The maximal diameter of entinus edodes lid, degree of circularity, average of hue and ratio value between total areas of defective regions and total areas of entinus edodes image are selected as character parameters to grate all samples. The model referring the relation between character parameters and entinus edodes grading is set up by BP neural network. The results show that the prediction identification can reach 94.2%.