Abstract: In order to solve the problem of low efficiency and high subjectivity of traditional wheat identification, an improved CBAM-InceptionV3 wheat impurity identification method was proposed. Firstly, a machine vision online detection platform was built to collect dynamic image data, and the wheat impurity image was processed by data set enhancement, image preprocessing and KS classification. Then, GoogLeNet, ResNet34 and InceptionV3 models were used to classify and train the image data set. Secondly, based on InceptionV3 model, CBAM was introduced to enhance the sensitivity of the model to information and improve the recognition accuracy of the model. The improved convolutional neural network CBAM-InceptionV3 model is compared with CA-InceptionV3 and InceptionV3 models added in CA module. The results show that the accuracy of InceptionV3 model on test set is 83.5% and F₁-Score is 82.41%, and the accuracy of CA-InceptionV3 model on test set is 92.3% and F₁-Score is 92.29%. CBAM-InceptionV3 has 92.9% accuracy and 92.92% F₁-Score on the test set. The average prediction time of CBAM-InceptionV3 model for the test set is 0.045 pieces/s, which is significantly better than the other two models.