Affine-Function Transformation-Based Object Matching for Vehicle Detection from Unmanned Aerial Vehicle Imagery

• Published in: Remote sensing (Basel, Switzerland) Vol. 11; no. 14; p. 1708
• Main Authors: Cao, Shuang, Yu, Yongtao, Guan, Haiyan, Peng, Daifeng, Yan, Wanqian
• Format: Journal Article
• Language: English
• Published: Basel MDPI AG 19.07.2019
• Subjects: Classification; Comparative studies; comparative study; Construction; data collection; Deep learning; Genetic transformation; Image detection; Image processing; Image segmentation; Information science; International conferences; lighting; Linear programming; Matching; Neural networks; object matching; quantitative analysis; Remote sensing; remote sensing imagery; Response elimination; Roads & highways; Semantics; Sensors; superpixel segmentation; Traffic police; unmanned aerial vehicle; Unmanned aerial vehicles; Variation; vehicle detection; Vehicles; China
• ISSN: 2072-4292; 2072-4292
• DOI: 10.3390/rs11141708

Vehicle detection from remote sensing images plays a significant role in transportation related applications. However, the scale variations, orientation variations, illumination variations, and partial occlusions of vehicles, as well as the image qualities, bring great challenges for accurate vehicle detection. In this paper, we present an affine-function transformation-based object matching framework for vehicle detection from unmanned aerial vehicle (UAV) images. First, meaningful and non-redundant patches are generated through a superpixel segmentation strategy. Then, the affine-function transformation-based object matching framework is applied to a vehicle template and each of the patches for vehicle existence estimation. Finally, vehicles are detected and located after matching cost thresholding, vehicle location estimation, and multiple response elimination. Quantitative evaluations on two UAV image datasets show that the proposed method achieves an average completeness, correctness, quality, and F1-measure of 0.909, 0.969, 0.883, and 0.938, respectively. Comparative studies also demonstrate that the proposed method achieves compatible performance with the Faster R-CNN and outperforms the other eight existing methods in accurately detecting vehicles of various conditions.