| 基于图像纹理特征的牛肉嫩度预测方法研究 |
| |
| 引用本文： | 王卫,沈明霞,彭增起,陈士进,吴海娟,刘超超,梁林,谌启亮.基于图像纹理特征的牛肉嫩度预测方法研究[J].食品科学,2012,33(15):61-65. |
| |
| 作者姓名： | 王卫 沈明霞 彭增起 陈士进 吴海娟 刘超超 梁林 谌启亮 |
| |
| 作者单位： | 1. 南京农业大学工学院
2.南京农业大学 农业部农畜产品加工与质量控制重点开放实验室 |
| |
| 基金项目： | 国家现代农业(肉牛)产业技术体系项目(080600231;080600232);农业科技成果转化资金项目(SQ2011ECC100043);江苏高校优势学科建设工程资助项目(PAPD) |
| |
| 摘 要： | 在经过图像预处理，背最长肌与大理石花纹的分割，并实现大理石花纹特征值的提取后，利用灰度共生矩阵提取4个对嫩度剪切力贡献较大的纹理特征参数，并统计这些参数应用多元线性回归建立牛肉嫩度剪切力预测模型。结果表明：可见光下利用纹理特征预测牛肉嫩度的方法能够以96%的准确率实现嫩度剪切力等级的预测，具有较高的商用开发价值。
|
| 关 键 词： | 牛肉 嫩度 纹理 灰度共生矩阵 多元线性回归 |
| 收稿时间： | 2011-07-29 |
Prediction of Beef Tenderness Based on Image Texture Features |
| |
| WANG Wei,SHEN Ming-xia,PENG Zeng-qi,CHEN Shi-jin,WU Hai-juan,LIU Chao-chao,LIANG Lin,CHEN Qi-liang.Prediction of Beef Tenderness Based on Image Texture Features[J].Food Science,2012,33(15):61-65. |
| |
| Authors: | WANG Wei SHEN Ming-xia PENG Zeng-qi CHEN Shi-jin WU Hai-juan LIU Chao-chao LIANG Lin CHEN Qi-liang |
| |
| Affiliation: | 1.College of Engineering,Nanjing Agricultural University,Nanjing 210031,China;2.Key Laboratory of Agricultural and Animal Products Processing and Quality Control,Ministry of Agriculture,Nanjing Agricultural University,Nanjing 210095,China) |
| |
| Abstract: | A mathematical modeling method for predicting beef tenderness utilizing image texture features under visible light was proposed.After image preprocessing,beef longissimus dorsi muscle and marbling were segmented,and then four marbling features that greatly influence beef shear force as a measure of meat tenderness were extracted by grey-level co-occurence matrix(GLCM) technique and statistically analyzed to establish a multiple linear regression model for predicting beef shear force.The proposed predictive method for beef shear force allowed 96% accurate prediction of beef tenderness,indicating its high value for commercial application. |
| |
| Keywords: | beef tenderness texture gray level co-occurrence matrix multiple linear regression |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《食品科学》浏览原始摘要信息 |
|
点击此处可从《食品科学》下载全文 |
|
