基于机器视觉的竹片颜色分选系统设计

Design of Bamboo Color Sorting System Based on Machine Vision

受消费者欢迎的竹制品加工中,竹片颜色分拣是竹制品生产中重要的工序。本文设计了一种基于机器视觉的竹片颜色自动分选系统,把工业相机采集的竹片颜色RGB颜色空间图像,通过低通滤波器降噪和直方图规定化增强处理,利用MATLAB软件转换到L~*a~*b~*空间后提取L~*、a~*、b~*、相关的均值、均差和标准差等19个特征值构成的特征向量,应用距离测度模板匹配法进行分类。为国内研究适合竹片颜色的分选系统提供有力参考。

Bamboo color sorting is an important process in the production of bamboo products,which is popular with consumers. In this paper,an automatic bamboo color sorting system based on machine vision is designed. The RGB color space image of bamboo color collected by industrial camera is reduced by low-pass filter and enhanced by histogram specification. The feature vectors composed of 19 feature values,such as L^＊a^＊b^＊,and related mean,mean and standard deviation,are extracted by converting to L^＊a^＊b^＊ space by MATLAB software,and then classified by distance measure template matching method. It provides a powerful reference for domestic research on sorting system suitable for bamboo color.