基于图像分割的筷子瑕疵检测研究  

Detection of chopsticks defects research based on image segmentation

作　　者：胡小慧[1] 江虹[1] 郭秋梅[1] 

机构地区：[1]西南科技大学信息工程学院,四川绵阳621010

出　　处：《微型机与应用》2016年第2期39-42,46,共5页Microcomputer & Its Applications

摘　　要：针对筷子瑕疵人工检测存在效率低、成本高、工作环境恶劣等缺点,提出了一种基于图像分割的筷子瑕疵检测研究分析方法。该方法采用图像采集设备获取筷子图像,利用图像处理技术对筷子图像进行增强、边缘检测及图像分割,获取筷子图像的RGB数据;对RGB三通道所占百分比及百分比方差进行统计分析。实验结果表明,该方法能有效地将目标筷子的彩色图像与筷子的背景图像分离,得到准确的筷子RGB数据,数据分析结果对筷子瑕疵检测有很大的参考价值。Artificial chopstick defects detection have some problems such as low efficiency,high cost,poor working conditions,etc.The method of detecting chopstick defects based on image segmentation is put forward in this paper.Firstly,using the image acquisition device to accumulate chopsticks image.Then,this paper applies the image processing technology of image enhancement,edge detection and image segmentation to deal with the chopsticks images.With those technologies,we get the RGB data of chopsticks image and realize the statistical analysis of RGB percentage and the percentage of variance values.The experimental results show that the method of this paper can effectively separate target chopsticks and background image,and can get the exact chopsticksRGB data,and the data analysis results for chopsticks defects detection have great reference value.

关 键 词：筷子瑕疵 边缘检测 图像增强 图像分割 

分 类 号：TN06[电子电信—物理电子学]