摘要
目的开发客观、准确、无损的基于深度学习的牛肉大理石纹智能化分级技术。方法将深度学习的图像识别方法应用于牛肉大理石纹的特征提取和分类上,并进行相应的调试和学习。结果通过计算机调试和学习,评级正确率分别达到84.2%(一级)、89.4%(二级)、81.9%(三级)、84.1%(四级)、82.6%(五级)。各级牛肉的识别率均在80%以上,识别时间都在1 s以内,达到了预期目标。结论将深度学习的图像识别方法应用于牛肉大理石纹的特征提取和分类上,评级准确率非常高,且随着图片数据库样本数的不断增多,其识别的准确度将不断提高,可进行大量推广使用。
Objective To develop an objective, accurate and nondestructive intelligent grading method of beef marbling based on deep learning. Methods The image recognition method based on deep learning algorithm was used for feature extraction and classification of beef marbling, and the corresponding debugging and learning was carries out. Results The accuracy rates of beef marbling grading reached to 84.2% for the first level, 89.4% for the second level, 81.9% for the third level, 84.1% for the forth level and 82.6% for the fifth level, respectively, through the computer debugging and learning. The recognition rates of beef at all levels were above 80%, the recognition time was all within 1 s, and the desired goal was achieved. Conclusion The image recognition method based on deep learning algorithm can be used for feature extraction and classification of beef marbling, and the accuracy of this method is good. With the increasing number of samples in the picture database, the accuracy of its recognition will be improved continuously, and this method can be widely used.
作者
兰韬
初侨
刘文
戴岳
杨博
杨丽
张晓芳
席兴军
LAN Tao1, CHU Qiao1, LIU Wen1, DAI Yue1, YANG Bo2, YANG Li1, ZHANG Xiao-Fang1, XI Xing-Jun1.(1. Sub-Institutes of Food and Agriculture, China Institute of Standardization, Beijing 100191, China; 2. State Key Laboratory of Information Engineering in Surveying, Wuhan University, Wuhan 430072, Chin)
出处
《食品安全质量检测学报》
CAS
2018年第5期1059-1064,共6页
Journal of Food Safety and Quality
基金
中央级公益性科研院所基本科研业务费专项资金(562016Y-4489)~~
