Mutton adulteration qualitative judgment method based on temperature distribution difference

The invention discloses a mutton adulteration qualitative judgment method based on temperature distribution difference, and particularly relates to a method for optimally obtaining a thermal image with large temperature distribution difference by fully utilizing an infrared thermal image video, extracting a region of interest (ROI) of the thermal image and further carrying out mutton adulterationidentification by utilizing an optimized judgment model. The method comprises the steps of firstly establishing a thermogram judgment model for doping different meats into mutton, and then identifyingthe different meats doped into adulterated mutton by utilizing the judgment model. According to the method, a deep learning algorithm and a thermal imaging technology are combined to be applied to judgment of adulterated different meats in mutton, and high judgment accuracy is obtained; a convolutional neural network can automatically perform data normalization and feature extraction, so that thecomplexity of data preproc