YOLO-FCE: A feature and clustering enhanced object detection model for species classification

• We proposed an accurate animal identification framework YOLO-FCE, highlighting its ability to handle high species count and intraspecific appearance variations. • We provide a detailed analysis of how different feature extraction techniques impact identification precision, offering insights into the factors that enhance model performance. • Despite the visual similarities among several species, YOLO-FCE obtains a mAP50 value of 90.8%, a mAP50:95 value of 87.5%, and a precision value of 98.2%, showcasing its robustness and potential for real-world applications. • By leveraging the advanced capabilities of YOLOv9, we aim to improve the efficiency and accuracy of wildlife monitoring in Australia, thereby contributing to better conservation and management of its natural heritage. Australia harbours a rich and unique diversity of wildlife, constituting a vital component of the nation’s ecological heritage. Accurate species identification in expansive and remote natural environments remains a significant challenge. In this study, we propose YOLO-Feature and Clustering Enhanced (YOLO-FCE), an improved model based on the YOLOv9 architecture. We conducted a series of cluster-distance-based analyses to evaluate and enhance the model’s feature extraction capabilities. The proposed model was trained and tested on a dataset containing 50 Australian animal species, with 700 images per species, resulting in a total of 35,000 images. YOLO-FCE achieved a mean Average Precision (mAP50:95) of 87.5% and a precision of 98.2%. On a separate validation set of previously unseen images, it attained a recognition accuracy of 91.29% with an average confidence score of 0.801. Compared with baseline models including YOLOv9, YOLOv11, and Faster R-CNN evaluated on the same dataset, YOLO-FCE demonstrated robust performance.