Accurate detection of maize tassels plays a crucial role in yield estimation of maize in precision agriculture. Recently, UAV and deep learning technologies have been widely introduced in various applications of field monitoring. However, complex field backgrounds pose multiple challenges against the precision detection of maize tassels, including maize tassel multi-scale variations caused by varietal differences and growth stage variations, intra-class occlusion, and background interference. To achieve accurate maize tassel detection in UAV images under complex field backgrounds, this study proposes an MSMT-RTDETR detection model. The Faster-RPE Block is first designed to enhance multi-scale feature extraction while reducing model Params and FLOPs. To improve detection performance for multi-scale targets in complex field backgrounds, a Dynamic Cross-Scale Feature Fusion Module (Dy-CCFM) is constructed by upgrading the CCFM through dynamic sampling strategies and multi-branch architecture. Furthermore, the MPCC3 module is built via re-parameterization methods, and further strengthens cross-channel information extraction capability and model stability to deal with intra-class occlusion. Experimental results on the MTDC-UAV dataset demonstrate that the MSMT-RTDETR significantly outperforms the baseline in detecting maize tassels under complex field backgrounds, where a precision of 84.2% was achieved. Compared with Deformable DETR and YOLOv10m, improvements of 2.8% and 2.0% were achieved, respectively, in the mAP50 for UAV images. This study proposes an innovative solution for accurate maize tassel detection, establishing a reliable technical foundation for maize yield estimation.
Instance segmentation of moso bamboo cells is a critical step in quantitative structural analysis of bamboo materials and plant phenomics research. Moso bamboo tissues are mainly composed of vascular bundles and parenchyma cells. Within vascular bundles, fiber cells exhibit thick cell walls and extremely dense arrangements, whereas vessel cells are characterized by large diameters and complex internal structures. These features frequently lead to blurred boundaries, structural complexity, and local overexposure in microscopic images, making it difficult for traditional segmentation algorithms to achieve stable and accurate results. Although the U-Net has demonstrated outstanding performance in biological microscopic image analysis, its feature extraction capability and boundary recognition stability remain insufficient when dealing with the composite structure of moso bamboo. To address these challenges, this study proposes an improved model based on a multi-scale attention mechanism, termed MBMSA-UNet (Moso Bamboo Multi-Scale Attention U-Net). Building upon the encoder–decoder architecture of U-Net, the proposed model introduces a multi-scale channel-spatial attention block, aiming to handle the pronounced morphological and scale differences among vessels, fibers, and parenchyma cells. By adaptively reweighting features at different scales, the model enhances cross-layer feature fusion and strengthens responses to key regions, thereby effectively suppressing local overexposure interference and emphasizing boundary features between different cell types. Experimental results demonstrate that, compared with the U-Net and several of its improved variants, MBMSA-UNet achieves higher segmentation accuracy and greater robustness on microscopic images of moso bamboo, providing a solid foundation for fine-grained quantitative analysis of complex bamboo tissues.