In order to realize the fast and accurate optimization of corn seeds, a watershed algorithm combined with convolutional neural network was proposed to detect the quality of corn seeds with different quality as the research object. Firstly, the watershed algorithm was used to divide the single corn seed, and then the quality of each seed was classified by the convolutional neural network model. According to the position of the single seed obtained by the watershed algorithm, the results were labeled in the image to realize the quality detection of seeds. The improved InceptionV3 model was used to test the seeds. The test results showed that the average accuracy rate, the average recall rate and the F1 value (harmonic average evaluation) of the two kinds of seeds with good quality and defects were 94.18%, 94.61% and 94.39%. Meanwhile, in order to highlight the performance of the convolutional neural network model, the results were compared with the traditional machine learning method, and the F1 value of the convolutional neural network model was 20.39% higher than that of the LBP+SVM model.

This paper aims to solve the problems of low recognition rate of potato leaf diseases and difficulty in localization of late blight spots in the natural environment based on potato leaf images collected in the field environment. Firstly, the potato leaf disease was identified using five neural network models: ALEXNET, VGG16, InceptionV3, RESNET50, and MobileNet. The results show that the InceptionV3 model has the highest recognition accuracy, which can reach 98.00%. Secondly, the late blight spots of potato leaves were detected, and an improved CenterNet-SPP model was proposed. The model obtains the center point of the object through the feature extraction network and then obtains the image information such as the offset of the center point and the size of the target through the regression of the center point. The mAP of the trained model under the verification set was up to 90.03%.Using F ₁ as the evaluation value analysis, compared with other objective detection models, the CenterNet-SPP model has the best effect, whose accuracy is 94.93%, recall rate is 90.34%, F ₁ value is 92.58%, and the average detection time of an image is 0.10 s.This paper provides a relatively comprehensive deep learning algorithm and model research basis for potato disease recognition and detection in the natural environment.