Aiming at difficult image acquisition and low recognition accuracy of two rice canopy pests, rice stem borer and rice leaf roller, we constructed a GA-Mask R-CNN (Generative Adversarial Based Mask Region Convolutional Neural Network) intelligent recognition model for rice stem borer and rice leaf roller, and we combined it with field monitoring equipment for them. Firstly, based on the biological habits of rice canopy pests, a variety of rice pest collection methods were used to obtain the images of rice stem borer and rice leaf roller pests. Based on different segmentation algorithms, the rice pest images were segmented to extract single pest samples. Secondly, the bug generator based on a generative adversarial network strategy improves the sensitivity of the classification network to the bug information, generates the pest information images in the real environment, and obtains the sample dataset for deep learning through multi-way augmentation. Then, through adding channel attention ECA module in Mask R-CNN and improving the connection of residual blocks in the backbone network ResNet101, the recognition accuracy of the model is improved. Finally, the GA-Mask R-CNN model was tested on a multi-source dataset with an average precision (AP) of 92.71%, recall (R) of 89.28% and a balanced score F1 of 90.96%. The average precision, recall, and balanced score F1 are improved by 7.07, 7.65, and 8.83%, respectively, compared to the original Mask R-CNN. The results show that the GA-Mask R-CNN model performance indexes are all better than the Mask R-CNN, the Faster R-CNN, the SSD, the YOLOv5, and other network models, which can provide technical support for remote intelligent monitoring of rice pests.

Rice membrane-covered cultivation offers notable agronomic advantages, including effective weed suppression and improved moisture retention. However, current mechanized approaches remain constrained by high labor requirements, low operational efficiency, and the inherent fragility of biodegradable membranes. To address these limitations, this study integrates a high-speed synchronous membrane-covering device, governed by a PSO-Fuzzy PID control algorithm, into a conventional rice transplanter. This integration enables precise coordination between membrane-laying and transplanting operations. The mechanical properties of the membranes were analyzed, and a tension evaluation model was developed considering structural parameters and roll diameter variation. Experimental tests on three biodegradable membranes revealed an average thickness of 0.012 mm, a longitudinal tensile force of 0.57 N, and a tensile strength of 2.85 N/mm. The PSO algorithm was employed to optimize fuzzy PID parameters (K = 5.3095, K_p = 10.6981, K_i = 0.0100, K_d = 8.2892), achieving adaptive synchronization between membrane output speed and transplanter travel speed. Simulation results demonstrated that the PSO-Fuzzy PID reduced rise time by 53.13%, stabilization time by 90.58%, and overshoot by 3.3% compared with the conventional PID. In addition, a dedicated test bench for the membrane-covering device was designed and fabricated. Orthogonal experiments determined the optimal parameters for the speed-measurement system: a membrane pressure of 5.000 N, a roller width of 28.506 mm, and a placement angle of 0.690°. Under these conditions, the minimum membrane-stretching tension was 0.55 N, and the rotational speed error was 0.359%. Field tests indicated a synchronization error below 1.00%, a membrane-width variation rate below 1.50%, and strong anti-interference capability. The proposed device provides an effective solution for intelligent and fully mechanized rice transplanting.