Vehicle Detection in Aerial Images Based on Hyper Feature Map in Deep Convolutional Network = Vehicle Detection in Aerial Images Based on Hyper Feature Map in Deep Convolutional Network

다국어 초록 (Multilingual Abstract)

Vehicle detection based on aerial images is an interesting and challenging research topic. Most of the traditional vehicle detection methods are based on the sliding window search algorithm, but these methods are not sufficient for the extraction of object features, and accompanied with heavy computational costs. Recent studies have shown that convolutional neural network algorithm has made a significant progress in computer vision, especially Faster R-CNN. However, this algorithm mainly detects objects in natural scenes, it is not suitable for detecting small object in aerial view. In this paper, an accurate and effective vehicle detection algorithm based on Faster R-CNN is proposed. Our method fuse a hyperactive feature map network with Eltwise model and Concat model, which is more conducive to the extraction of small object features. Moreover, setting suitable anchor boxes based on the size of the object is used in our model, which also effectively improves the performance of the detection. We evaluate the detection performance of our method on the Munich dataset and our collected dataset, with improvements in accuracy and effectivity compared with other methods. Our model achieves 82.2% in recall rate and 90.2% accuracy rate on Munich dataset, which has increased by 2.5 and 1.3 percentage points respectively over the state-of-the-art methods.

참고문헌 (Reference)

1 "https://pan.baidu.com/s/1mz-phfgwG3VdF0OASAI14g"

2 Joseph Redmon, "You Only Look Once: Unified, Real-Time Object Detection" 779-788, 2016

3 Joseph Redmon, "YOLO9000: Better , Faster , Stronger" 6517-6525, 2017

4 Xiangbo Shu, "Weakly-Shared Deep Transfer Networks for Heterogeneous-Domain Knowledge Propagation" 35-44, 2015

5 Matthew D Zeiler, "Visualizing and understanding convolutional networks" 818-833, 2014

6 Karen Simonyan, "Very Deep Convolutional Networks for Large-Scale Image Recognition" 2015

7 Aniruddha Kembhavi, "Vehicle detection using partial least squares" 33 (33): 1250-1265, 2011

8 Ziyi Chen, "Vehicle detection in high-resolution aerial images via sparse representation and superpixels" 54 (54): 103-116, 2015

9 Yanjun Liu, "Vehicle detection in high resolution satellite images with joint-layer deep convolutional neural networks" 1-6, 2016

10 Tianyu Tang, "Vehicle detection in aerial images based on region convolutional neural networks and hard negative example mining" 17 (17): 336-352, 2017

11 Hsu-Yung Cheng, "Vehicle Detection in Aerial Surveillance Using Dynamic Bayesian Networks" 21 (21): 2152-2159, 2012

12 Zhipeng Deng, "Toward Fast and Accurate Vehicle Detection in Aerial Images Using Coupled Region-Based Convolutional Neural Networks" 10 (10): 3652-3664, 2017

13 Juan-juan Zhu, "Surf points based Moving Target Detection and Long-term Tracking in Aerial Videos" 한국인터넷정보학회 10 (10): 5624-5638, 2016

14 Wei Liu, "SSD: Single shot multibox detector" 21-37, 2016

15 Zhong Jiandan, "Robust Vehicle Detection in Aerial Images Based on Cascaded Convolutional Neural Networks" 17 (17): 2720-2737, 2017

16 Ross B Girshick, "Region-Based Convolutional Networks for Accurate Object Detection and Segmentation" 38 (38): 142-158, 2016

17 PDA Kraaijenbrink, "Object-based analysis of unmanned aerial vehicle imagery to map and characterise surface features on a debris-covered glacier" 186 (186): 581-595, 2016

18 Pedro F Felzenszwalb, "Object Detection with Discriminative Trained Part Based Models" 32 (32): 1627-1645, 2010

19 Paul Viola, "Multiple instance boosting for object detection" 1417-1424, 2005

20 Gong Cheng, "Learning Rotation-Invariant Convolutional Neural Networks for Object Detection in VHR Optical Remote Sensing Images" 54 (54): 7405-7415, 2016

21 Meng Yi, "Improved image alignment algorithm based on projective invariant for aerial video stabilization" 한국인터넷정보학회 8 (8): 3177-3195, 2014

22 Nouman Ali, "Image retrieval by addition of spatial information based on histograms of triangular regions" 54 : 539-550, 2016

23 Bushra Zafar, "Image classification by addition of spatial information based on histograms of orthogonal vectors" 13 (13): 2018

24 Navneet Dalal, "Histograms of oriented gradients for human detection" 886-893, 2005

25 Timo Ojala, "Gray-scale and rotation invariant texture classification with local binary patterns" 404-420, 2000

26 Chia-Feng Juang, "Fuzzy Classifiers Learned Through SVMs with Application to Specific Object Detection and Shape Extraction Using an RGB-D Camera" 777 : 253-274, 2018

27 Guimei Cao, "Feature-Fused SSD: Fast Detection for Small Objects" 2018

28 Shaoqing Ren, "Faster R-CNN : Towards Real-Time Object Detection with Region Proposal Networks" 39 (39): 1137-1149, 2017

29 Ross Girshick, "Fast R-CNN" 1440-1448, 2015

30 Liu Kang, "Fast Multiclass Vehicle Detection on Aerial Images" 12 (12): 1938-1942, 2015

31 D. N. Chandrappa, "Face Detection Using a Boosted Cascade of Features Using OpenCV" 399-404, 2012

32 Hailing Zhou, "Efficient Road Detection and Tracking for Unmanned Aerial Vehicle" 16 (16): 297-309, 2015

33 David G. Lowe, "Distinctive image features from scale-invariant keypoints" 60 (60): 91-110, 2004

34 Thomas Moranduzzo, "Detecting cars in UAV images with a catalog-based approach" 52 (52): 6356-6367, 2014

35 Amir Ghodrati, "Deepproposal: Hunting objects by cascading deep convolutional layers" 2578-2586, 2015

36 Lars Wilko Sommer, "Deep learning based multi-category object detection in aerial images" 10202 : 2017

37 Nassim Ammour, "Deep learning approach for car detection in UAV imagery" 9 (9): 1-15, 2017

38 K. Liu, "DLR 3k Munich Vehicle Aerial Image Dataset"

39 Yakoub Bazi, "Convolutional SVM Networks for Object Detection in UAV Imagery" 56 (56): 3107-3118, 2018

40 Igor Sevo, "Convolutional Neural Network Based Automatic Object Detection on Aerial Images" 13 (13): 740-744, 2016

41 Xiangbo Shu, "Concurrence-Aware Long Short-Term Sub-Memories for Person-Person Action Recognition" 2176-2183, 2017

42 Yongzheng Xu, "Car detection from low-altitude UAV imagery with the faster R-CNN" 2017 : 1-10, 2017

43 Wen Shao, "Car detection from high-resolution aerial imagery using multiple features" 4379-4382, 2012

44 Thomas Moranduzzo, "Automatic car counting method for unmanned aerial vehicle images" 52 (52): 1635-1647, 2014

45 Yongzheng Xu, "An Enhanced Viola-Jones Vehicle Detection Method from Unmanned Aerial Vehicles Imagery" 18 (18): 1-12, 2017

46 Yanxiang Chen, "Accurate seat belt detection in road surveillance images based on CNN and SVM" 274 : 80-87, 2018

47 Gong Cheng, "A survey on object detection in optical remote sensing images" 117 : 11-28, 2016

48 Mesay Belete Bejiga, "A convolutional neural network approach for assisting avalanche search and rescue operations with UAV imagery" 9 (9): 100-121, 2017

49 Faisal Riaz, "A collision avoidance scheme for autonomous vehicles inspired by human social norms" 69 : 690-704, 2018

50 Nouman Ali, "A Novel Image Retrieval Based on Visual Words Integration of SIFT and SURF" 11 (11): 2016