Wide-Area and Real-Time Object Search System of UAV

Abstract

The method of collecting aerial images or videos by unmanned aerial vehicles (UAVs) for object search has the advantages of high flexibility and low cost, and has been widely used in various fields, such as pipeline inspection, disaster rescue, and forest fire prevention. However, in the case of object search in a wide area, the scanning efficiency and real-time performance of UAV are often difficult to satisfy at the same time, which may lead to missing the best time to perform the task. In this paper, we design a wide-area and real-time object search system of UAV based on deep learning for this problem. The system first solves the problem of area scanning efficiency by controlling the high-resolution camera in order to collect aerial images with a large field of view. For real-time requirements, we adopted three strategies to accelerate the system, as follows: design a parallel system, simplify the object detection algorithm, and use TensorRT on the edge device to optimize the object detection model. We selected the NVIDIA Jetson AGX Xavier edge device as the central processor and verified the feasibility and practicability of the system through the actual application of suspicious vehicle search in the grazing area of the prairie. Experiments have proved that the parallel design of the system can effectively meet the real-time requirements. For the most time-consuming image object detection link, with a slight loss of precision, most algorithms can reach the 400% inference speed of the benchmark in total, after algorithm simplification, and corresponding model’s deployment by TensorRT.

A Survey on UAV Applications in Smart City Management: Challenges, Advances, and Opportunities

“…Users can deploy corresponding DL algorithms on such platforms to realize the real-time processing of data acquired by airborne sensors and make corresponding decisions in real time based on the processing results to make UAVs perform tasks more intelligently and automatically. This mode of work has been widely used, such as in the automatic inspection of urban power lines [34], wide-area real-time target search [84], traffic management [85], etc. However, this hardware device is not necessary.…”

A Computational Model for Simulating the Performance of UAS-Based Construction Safety Inspection through a System Approach

“…The term UAS refers to a platform that includes unmanned aerial vehicles (UAVs), commonly called drones, as well as communication sensors and a ground control station including a human flight crew [33,34]. Using the UAS, a wide area can be filmed in real time, and communication sensors can be used to provide real-time data [24,35,36]. The safety manager of the construction site detects safety risks at the site and takes corrective actions but cannot perform real-time monitoring simultaneously on sites with extensive areas [37,38].…”

Editorial for Special Issue: “Recent Progress in UAV-AI Remote Sensing”

“…Ref. [9] discusses the design of a wide-area and real-time object search system for UAVs. The system utilizes high-resolution cameras to collect aerial images with a large field of view, and parallel processing techniques to achieve real-time object search in a wide area.…”