ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Download
Publications Copernicus
Download
Citation
Articles | Volume V-2-2020
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-2-2020, 381–388, 2020
https://doi.org/10.5194/isprs-annals-V-2-2020-381-2020
ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci., V-2-2020, 381–388, 2020
https://doi.org/10.5194/isprs-annals-V-2-2020-381-2020

  03 Aug 2020

03 Aug 2020

LR-CNN: LOCAL-AWARE REGION CNN FOR VEHICLE DETECTION IN AERIAL IMAGERY

W. Liao1, X. Chen2, J. Yang3, S. Roth2, M. Goesele2, M. Y. Yang4, and B. Rosenhahn1 W. Liao et al.
  • 1Leibniz Universität Hannover, Germany
  • 2Technische Universität Darmstadt, Germany
  • 3Chinese Academy of Sciences, China
  • 4Faculty ITC, University of Twente, The Netherlands

Keywords: Deep Learning, Object Detection, Vehicle Detection, Twin Region Proposal, Feature Enhancement

Abstract. State-of-the-art object detection approaches such as Fast/Faster R-CNN, SSD, or YOLO have difficulties detecting dense, small targets with arbitrary orientation in large aerial images. The main reason is that using interpolation to align RoI features can result in a lack of accuracy or even loss of location information. We present the Local-aware Region Convolutional Neural Network (LR-CNN), a novel two-stage approach for vehicle detection in aerial imagery. We enhance translation invariance to detect dense vehicles and address the boundary quantization issue amongst dense vehicles by aggregating the high-precision RoIs’ features. Moreover, we resample high-level semantic pooled features, making them regain location information from the features of a shallower convolutional block. This strengthens the local feature invariance for the resampled features and enables detecting vehicles in an arbitrary orientation. The local feature invariance enhances the learning ability of the focal loss function, and the focal loss further helps to focus on the hard examples. Taken together, our method better addresses the challenges of aerial imagery. We evaluate our approach on several challenging datasets (VEDAI, DOTA), demonstrating a significant improvement over state-of-the-art methods. We demonstrate the good generalization ability of our approach on the DLR 3K dataset.