The precise execution of semantic segmentation on remote sensing data is a pivotal factor. It determines the achievements and impact of geoscience endeavors and their applications. However, challenges caused by target edge blurring and scale variability in high-resolution remote sensing imagery hinder the improvement of segmentation accuracy. In this work, to address these issues, a Boundary-Enhanced Multi-Scale Semantic Segmentation Network (BEMS-UNetFormer) based on UNetFormer is proposed for remote sensing data. Firstly, an improved Boundary Awareness Module (BAM) is used to extract the edge information of the target from the low-level features to enhance the recognition of the target edges. Secondly, the improved Boundary-Guided Fusion Module (BFM) incorporates the edge information from BAM into subsequent decoding, further refining the precise representation of boundary regions. Finally, at the pivotal junction between the encoder and decoder, the Multi-Scale Cascaded Atrous Spatial Pyramid Pooling (MSC-ASPP) is designed, capable of deeply mining and integrating multi-scale deep features. The method was tested on two mainstream datasets, Potsdam and Vaihingen, achieving 86.12% and 83.10% MIoU, respectively, improving by 1.38% and 1.79% over the baseline model. Notably, the IoU and F1 Score for the small-scale target “Car” in the Potsdam dataset reached 91.20% and 95.57%, respectively, while the “Building” and “LowVeg” categories in the Vaihingen dataset achieved the highest IoU and F1 Score. The experimental results indicate that the proposed method demonstrates higher precision in segmenting small-scale targets and target boundaries, surpassing mainstream methods overall.

Due to the existence of cotton weeds in a complex cotton field environment with many different species, dense distribution, partial occlusion, and small target phenomena, the use of the YOLO algorithm is prone to problems such as low detection accuracy, serious misdetection, etc. In this study, we propose a YOLOv8-DMAS model for the detection of cotton weeds in complex environments based on the YOLOv8 detection algorithm. To enhance the ability of the model to capture multi-scale features of different weeds, all the BottleNeck are replaced by the Dilation-wise Residual Module (DWR) in the C2f network, and the Multi-Scale module (MSBlock) is added in the last layer of the backbone. Additionally, a small-target detection layer is added to the head structure to avoid the omission of small-target weed detection, and the Adaptively Spatial Feature Fusion mechanism (ASFF) is used to improve the detection head to solve the spatial inconsistency problem of feature fusion. Finally, the original Non-maximum suppression (NMS) method is replaced with SoftNMS to improve the accuracy under dense weed detection. In comparison to YOLO v8s, the experimental results show that the improved YOLOv8-DMAS improves accuracy, recall, mAP0.5, and mAP0.5:0.95 by 1.7%, 3.8%, 2.1%, and 3.7%, respectively. Furthermore, compared to the mature target detection algorithms YOLOv5s, YOLOv7, and SSD, it improves 4.8%, 4.5%, and 5.9% on mAP0.5:0.95, respectively. The results show that the improved model could accurately detect cotton weeds in complex field environments in real time and provide technical support for intelligent weeding research.

Cotton is a crucial crop in the global textile industry, with major production regions including China, India, and the United States. While smart agricultural mechanization technologies, such as automated irrigation and precision pesticide systems, have improved crop management, weeds remain a significant challenge. These weeds not only compete with cotton for nutrients but can also serve as hosts for diseases, affecting both cotton yield and quality. Existing weed detection models perform poorly in the complex environment of cotton fields, where the visual features of weeds and crops are similar and often overlap, resulting in low detection accuracy. Furthermore, real-time deployment on edge devices is difficult. To address these issues, this study proposes an improved lightweight weed detection model, YOLO-WL, based on the YOLOv8 architecture. The model leverages EfficientNet to reconstruct the backbone, reducing model complexity and enhancing detection speed. To compensate for any performance loss due to backbone simplification, CA (cross-attention) is introduced into the backbone, improving feature sensitivity. Finally, AFPN (Adaptive Feature Pyramid Network) and EMA (efficient multi-scale attention) mechanisms are integrated into the neck to further strengthen feature extraction and improve weed detection accuracy. At the same time, the model maintains a lightweight design suitable for deployment on edge devices. Experiments on the CottonWeedDet12 dataset show that the YOLO-WL model achieved an mAP of 92.30%, reduced the detection time per image by 75% to 1.9 ms, and decreased the number of parameters by 30.3%. After TensorRT optimization, the video inference time was reduced from 23.134 ms to 2.443 ms per frame, enabling real-time detection in practical agricultural environments.

When aiming at the problems such as missed detection or misdetection of recognizing green walnuts in the natural environment directly by using target detection algorithms, a method is proposed based on improved UNet3+ for green walnut image segmentation, which incorporates the channel and spatial attention mechanism CBAM (convolutional block attention module) and cross-entropy loss function (cross-entropy loss) into the UNet3+ network structure, and introduces the five-layer CBAM in the encoder module to construct the improved UNet3+ network model. The model consists of an encoder module (down-sampling), a decoder module (up-sampling) and a full-scale skip connection module, a full-scale feature supervision module, and a classification guidance module. After utilizing data-enhanced approaches to expand the green walnut dataset, the improved UNet3+ model was trained. The experimental findings demonstrate that the improved UNet3+ network model achieves 91.82% average precision, 96.00% recall rate, and 93.70% F1 score in the green walnut segmentation task; the addition of five-layer CBAM boosts the model segmentation precision rate by 3.11 percentage points. The method can precisely and successfully segment green walnuts, which can serve as a guide and research foundation for precisely identifying and localizing green walnuts and finishing the autonomous sorting for intelligent robots.

The identification and classification of traditional Chinese herbal medicines demand significant time and expertise. We propose the dual-teacher supervised decay (DTSD) approach, an enhancement for Chinese herbal medicine recognition utilizing a refined knowledge distillation model. The DTSD method refines output soft labels, adapts attenuation parameters, and employs a dynamic combination loss in the teacher model. Implemented on the lightweight MobileNet_v3 network, the methodology is deployed successfully in a mobile application. Experimental results reveal that incorporating the exponential warmup learning rate reduction strategy during training optimizes the knowledge distillation model, achieving an average classification accuracy of 98.60% for 10 types of Chinese herbal medicine images. The model boasts an average detection time of 0.0172 s per image, with a compressed size of 10 MB. Comparative experiments demonstrate the superior performance of our refined model over DenseNet121, ResNet50_vd, Xception65, and EfficientNetB1. This refined model not only introduces an approach to Chinese herbal medicine image recognition but also provides a practical solution for lightweight models in mobile applications.

Yak daily behaviors, including feeding, standing, lying down, and walking, are closely related to their health status, making accurate behavior recognition essential for intelligent monitoring and management in yak husbandry. However, real-world grazing environments present significant challenges due to complex backgrounds, occlusions, small or distant targets, and high visual similarity between behavior categories. To address these issues, we propose a problem-driven, multi-scale behavior recognition framework based on an enhanced YOLOv11n architecture specifically designed for outdoor yak monitoring. A dedicated real-world dataset is constructed to capture four fundamental behaviors under diverse natural conditions. Based on this dataset, we develop the DPAP-YOLOv11n model, which incorporates Dynamic Convolution for adaptive feature modulation and Pinwheel-shaped Convolution (PConv) for fine-grained spatial representation. Additionally, a YOLOv7-Aux auxiliary training head is introduced to strengthen intermediate feature learning, and a Focal-PIoU loss function is adopted to improve robustness against hard or ambiguous samples. Experimental results show that DPAP-YOLOv11n outperforms the baseline YOLOv11n, achieving gains of 2.4% in mAP@50 and 2.8% in mAP@50–95. These findings demonstrate the practical potential of the proposed approach for high-precision, real-time yak behavior recognition in complex field environments.