Xinyi Ma

Aiming at the challenges in practical production lines, including the difficulty in accurately capturing external defects on continuously rolling walnuts, distinguishing subtle defects, and differentiating narrow fissures from natural walnut textures, this paper proposes an improved walnut external defect detection model named YOLO11-GME, based on YOLO11n. Firstly, the original backbone network is replaced with the lightweight GhostNetV1 network, enhancing model precision while meeting real-time detection speed requirements. Secondly, a Mixed Local Channel Attention (MLCA) mechanism is incorporated into the neck to strengthen the network’s ability to capture features of subtle defects, thereby improving defect recognition accuracy. Finally, the EIoU loss function is adopted to enhance the model’s localization capability for irregularly shaped defects and reduce false detection rates by improving the scale sensitivity of bounding box regression. Experimental results demonstrate that the improved YOLO11-GME model achieves a mean Average Precision (mAP) of 96.2%, representing improvements of 8.6%, 7%, and 5.8% compared to YOLOv5n, YOLOv8n, and YOLOv10n, respectively, and a 5.9% improvement over the original YOLOv11. Precision rates for the normal, fissure, and inferior categories increased by 8.7%, 5.3%, and 3.7%, respectively. The frame rate remains at 43.92 FPS, approaching the original model’s 51.02 FPS. These results validate that the YOLO11-GME model enhances walnut external defect detection accuracy while maintaining real-time detection speed, providing robust technical support for defect detection and classification in industrial walnut production.

Since it takes too long time for computer to train to recognize individual under SVM and other traditional algorithms. in this study, this paper used a better model called Convolutional Neural Network (CNN) as the main technology to reduce the training time for computer. This paper showed every step of my proposed model openly and transparently. Eleven of my own pictures are mixed with CASIA-FaceV5 dataset since Asian people are similar to each other, so it would help with exercising the accuracy of the CNN model. This paper used the file called haarcascade_frontalface_default.xml to extract people’s face for further training. Sequential mode in Keras was used for better operating. More specifically, this paper also used ReLu function as the activation function for better result. Stochastic gradient descent (SGD) was employed with momentum to minimize the loss function. Finally, computer camera was used for taking the picture and visualize the final result. Experimental results indicate that my CNN model achieves a satisfactory performance that it can provide accurate feedback for recognition.