滩羊胴体眼肌切面图像处理及分级特征值的提取  

Processing and Feature Values Extraction of Tan Sheep Carcass Rib-Eye Cross-Section Images for Carcass Classification

作　　者：张丽文[1] 田银[1] 牛佳[1] 柏霜[1] 罗瑞明[1] 

机构地区：[1]宁夏大学农学院,宁夏银川750021

出　　处：《肉类研究》2016年第11期6-10,共5页Meat Research

基　　金："十二五"国家科技支撑计划项目(2015BAD29B05)

摘　　要：为获得高质量滩羊胴体眼肌切面图像,采集图像用工业摄像机,并对样本图像分割算法进行研究。首先预处理样本图像,3×3模板中值滤波器去除干扰,采用加权平均值灰度化,用自动选择最优分割值OTSU算法去除复杂背景;再对其进行二值化,通过区域逐步分割对目标区域进行粗提取,通过区域标记和形态学处理进行有效眼肌区域的精确提取;最后对有效眼肌区域面积、背膘厚度进行计算。结果表明,图像处理算法具有较好的分割效果和计算效果。High-quality images of Tan sheep carcass rib-eye cross-sections were taken with an industrial camera and used to develop an algorithm to segment them.Firstly,the images were pre-processed and grayed by the weighted average method after removing the interference by 3× 3 template median filter.The automatic selection of the optimal segmentation threshold by the Otsu algorithm was carried out to remove the complex background.Then,after the binarization of the images,and the target region was roughly extracted through gradual region separation for accurate extraction of the effective rib-eye area by region labeling and morphological processing.At last,the effective rib-eye area and the back fat thickness were calculated.The image processing algorithm has a good segmentation and calculation performance.

关 键 词：滩羊胴体 眼机切面 图像处理 分级特征值 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]