To improve the long-distance trajectory planning efficiency of unmanned aerial vehicles (UAV) in 3 D complex scene, the study proposed a real-time UAV trajectory planning method based on local soft-constrained optimization. Firstly, the safety distance constraint was added to Theta^* algorithm, and the heuristic function was improved by using the turning cost to reduce the time consumption caused by UAV turning, and finally the initial path composed of a small number of key points was generated. Then, the segments with potential safety hazards in the initial path were optimized by the local optimization strategy based on soft constraints. The Hodograph property of Bézier curve was used for time allocation to ensure the continuity, smoothness and dynamic feasibility of the trajectory and to improve the UAV flight efficiency. Experimental results show that the proposed method has the advantages of shorter flight distance and less flight time, and higher planning efficiency while ensuring the safety of quadrotor. This method was successfully tested through the actual quadrotor flight.

Early dental caries detection by endoscope can prevent complications, such as pulpitis and apical infection. However, automatically identifying dental caries remains challenging due to the uncertainty in size, contrast, low saliency, and high interclass similarity of dental caries. To address these problems, we propose the Global Feature Detector (GFDet) that integrates the proposed Feature Selection Pyramid Network (FSPN) and Adaptive Assignment-Balanced Mechanism (AABM). Specifically, FSPN performs upsampling with the semantic information of adjacent feature layers to mitigate the semantic information loss due to sharp channel reduction and enhance discriminative features by aggregating fine-grained details and high-level semantics. In addition, a new label assignment mechanism is proposed that enables the model to select more high-quality samples as positive samples, which can address the problem of easily ignored small objects. Meanwhile, we have built an endoscopic dataset for caries detection, consisting of 1318 images labeled by five dentists. For experiments on the collected dataset, the F1-score of our model is 75.6%, which out-performances the state-of-the-art models by 7.1%.