SDE-DET: a precision network for Shatian pomelo detection in complex orchard environments

• We present STP-AgriData, a new dataset for Shatian pomelo detection, establishing a robust foundation for future research. To our knowledge, this is the first dataset that combines field-collected and publicly sourced data, thereby enhancing the diversity and comprehensiveness of the Shatian pomelo dataset. • We propose a new network SDE-DET, designed to achieve automatic, contactless, and accurate detection of Shatian pomelo in complex orchard environments. In SDE-DET, the Star Block, Deformable Attention, and EMA mechanisms are adopted to reduce feature loss, preserve key information from the original image, and enhance the detection of small and occluded Shatian pomelos. • Experimental results show that SDE-DET outperforms the state-of-the-art on the STP-AgriData dataset, achieving scores of 0.883, 0.771, 0.838, 0.497 and 0.823 in Precision, Recall, [email protected], [email protected]:0.95 and F1-score, respectively. Pomelo detection is an essential process for their localization, automated robotic harvesting, and maturity analysis. However, detecting Shatian pomelo in complex orchard environments poses significant challenges, including multi-scale issues, obstructions from trunks and leaves, small object detection, etc. To address these issues, this study constructs a custom dataset STP-AgriData and proposes the SDE-DET model for Shatian pomelo detection. SDE-DET first utilizes the Star Block to effectively acquire high-dimensional information without increasing the computational overhead. Furthermore, the presented model adopts Deformable Attention in its backbone, to enhance its ability to detect pomelos under occluded conditions. Finally, multiple Efficient Multi-Scale Attention (EMA) mechanisms are integrated into our model to reduce the computational overhead and extract deep visual representations, thereby improving the capacity for small object detection. In the experiment, we compared SDE-DET with the Yolo series and other mainstream detection models in Shatian pomelo detection. The presented SDE-DET model achieved scores of 0.883, 0.771, 0.838, 0.497, and 0.823 in Precision, Recall, [email protected], [email protected]:0.95 and F1-score, respectively. SDE-DET has achieved state-of-the-art performance on the STP-AgriData dataset. Experiments indicate that the SDE-DET provides a reliable method for Shatian pomelo detection, laying the foundation for the further development of automatic harvest robots.