Research on Wheat Mold Degree Detection Based on Computer Vision Technology

Wheat, as one of the main cereals widely consumed globally, is the staple food for about one-third of the population in China. The production of wheat is closely related to people's lives and the development of society. Wheat kernels are known for their thin outer skin, loose tissue structure, and good storability. However, wheat kernels are susceptible to environmental factors, leading to heat-induced mold phenomena, which not only affect the quality of grain but also pose a threat to human health. Therefore, developing a rapid and accurate method for detecting the degree of mold in wheat kernels is of great significance for ensuring food safety. Wheat kernels with different degrees of mold were selected as the research subjects, and a dataset of their images was obtained using computer vision technology. Through preprocessing of the images, extraction of features, and training of the model, a classification model based on Support Vector Machine (SVM) was established. After the model underwent rigorous testing, it was found that it achieved a recognition rate of 97.3% for the degree of mold in wheat kernels. This finding indicated that the combination of computer vision technology with the SVM classification model could efficiently and accurately detect the degree of mold in wheat kernels, offering a new technical approach for food safety surveillance.