A Method for Identifying Meat Quality Based on CNN-SVM

To identify the quality variation in meat during storage and transportation, and improve the efficiency of rapid detection of meat quality changes, a meat quality change recognition method based on an improved CNN-SVM model is proposed. Firstly, a CNN model is constructed using meat images as input samples. Replacing the fully connected layer of traditional CNN models with a global average pooling layer to reduce the number of model parameters while extracting image features. The data enhancement and batch normalization techniques are used to improve the model generalization ability and reduce the model overfitting. Secondly, nonlinear support vector machine SVM is used to replace the Softmax classifier in traditional CNN models to improve the accuracy of image change recognition. Finally, an example verification was conducted on the proposed model. The results show that compared with traditional CNN models, the improved CNN-SVM model has stronger recognition ability for changes in meat quality.