The impact of insect pests on the yield and quality of rice is extremely significant, and accurate detection of insect pests is of crucial significance to safeguard rice production. However, traditional manual inspection methods are inefficient and subjective, while existing machine learning-based approaches still suffer from limited generalization and suboptimal accuracy. To address these challenges, this study proposes an improved rice pest detection model, MTD-YOLO, based on the YOLOv8 framework. First, the original backbone is replaced with MobileNetV3, which leverages optimized depthwise separable convolutions and the Hard-Swish activation function through neural architecture search, effectively reducing parameters while maintaining multiscale feature extraction capabilities. Second, a Cross Stage Partial module with Triplet Attention (C2f-T) module incorporating Triplet Attention is introduced to enhance the model’s focus on infested regions via a channel-patial dual-attention mechanism. In addition, a Dynamic Head (DyHead) is introduced to adaptively focus on pest morphological features using the scale–space–task triple-attention mechanism. The experiments were conducted using two datasets, Rice Pest1 and Rice Pest2. On Rice Pest1, the model achieved a precision of 92.5%, recall of 90.1%, mAP@0.5 of 90.0%, and mAP@[0.5:0.95] of 67.8%. On Rice Pest2, these metrics improved to 95.6%, 92.8%, 96.6%, and 82.5%, respectively. The experimental results demonstrate the high accuracy and efficiency of the model in the rice pest detection task, providing strong support for practical applications.

To address false detections in small object detection within remote sensing imagery caused by complex backgrounds and minute target sizes, we propose an enhanced YOLOv8s detection algorithm, named AL-YOLOv8. The detection head is designed based on Adaptive Spatial Feature Fusion (ASFF) to resolve issues where shallow-level detail features of small remote sensing targets are easily disrupted by backgrounds, while deep-level semantic features lack sufficient representation. We embed Large-Kernel Separate Attention (LSKA) in the deep feature layer to expand the receptive field and enhance the response intensity of small target features. Additionally, an IFIoU loss function is introduced by combining the dynamic attention mechanism from FocalerIoU with InnerIoU, mitigating regression bias for small target bounding boxes and improving small target localization accuracy. On the DIOR, RSOD, and NWPU VHR-10 datasets, the AL-YOLOv8 model achieves precision rates of 91.5%, 94.2%, and 91.8%, respectively, with mAP@0.5 scores of 89.8%, 96.9%, and 92.2%. These results demonstrate consistent improvements over YOLOv8s and show that AL-YOLOv8 effectively reduces false detections and enhances detection accuracy for small object detection in remote sensing applications.