Ensemble Deep Network for Dense Vehicle Detection in Large Image (고해상도 영상의 군집형 차량 검출을 위한 앙상블 딥러닝 네트워크 구성 방법)

Jae-Hyoung Yu (유재형); Youngjun Han (한영준); KIM Jong Kuk (김종국); Hernsoo Hahn (한헌수)

doi:10.9708/jksci.2021.26.01.045

Ensemble Deep Network for Dense Vehicle Detection in Large Image

Journal of The Korea Society of Computer and Information
Abbr : JKSCI
2021, 26(1), pp.45-55
DOI : 10.9708/jksci.2021.26.01.045
Publisher : The Korean Society Of Computer And Information
Research Area : Engineering > Computer Science
Received : September 17, 2020
Accepted : November 30, 2020
Published : January 29, 2021

Jae-Hyoung Yu ¹, Youngjun Han ¹, KIM Jong Kuk ¹, Hernsoo Hahn ¹

¹숭실대학교

Accredited

ABSTRACT

This paper has proposed an algorithm that detecting for dense small vehicle in large image efficiently. It is consisted of two Ensemble Deep-Learning Network algorithms based on Coarse to Fine method. The system can detect vehicle exactly on selected sub image. In the Coarse step, it can make Voting Space using the result of various Deep-Learning Network individually. To select sub-region, it makes Voting Map by to combine each Voting Space. In the Fine step, the sub-region selected in the Coarse step is transferred to final Deep-Learning Network. The sub-region can be defined by using dynamic windows. In this paper, pre-defined mapping table has used to define dynamic windows for perspective road image. Identity judgment of vehicle moving on each sub-region is determined by closest center point of bottom of the detected vehicle's box information. And it is tracked by vehicle's box information on the continuous images. The proposed algorithm has evaluated for performance of detection and cost in real time using day and night images captured by CCTV on the road.

KEYWORDS

Ensemble Deep-Learning Network, Voting Map, Dense Small Object Detection, High Resolution Image, Dynamic Windows

Citation status

* References for papers published after 2023 are currently being built.

[journal] Q. QIN / 2000 / A New Approach to Object Recognition on High Resolution Satellite Image / International Archives of Photogrammetry and Remote Sensing XXXIII

[confproc] M. Müller / 1999 / Object Recognition Based on High Spatial Resolution Panchromatic Satellite Imagery / Proc. ISPRS Joint Workshop - Sensors and Mappingfrom Space 1999

[confproc] M. Yang / 2018 / Avoiding Pitfalls when Using NVIDIA GPUs for Real-Time Tasks in Autonomous Systems / ECRTS, Euromicro Conference on Real-Time Systems (20) : 20:1~20:21

[confproc] M. Cordts / 2016 / The cityscapes dataset for semantic urban scene understanding / Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition : 3213~3223

[journal] Y. Lv / 2015 / Traffic Flow Prediction with Big Data : A Deep Learning Approach / IEEE Trans. Intelligent Transportation Systems 16(2) : 865~873

[journal] L. B. Guaman / 2020 / Deep Learning Framework for Vehicle and Pedestrian Detection in Rural Roads on an Embedded GPU / Electronics 9 : 589~

[other] A. Marcu / Dual local-global contextual pathways for recognition in aerial imagery / Computing Research Repository (CRR) abs/1605.05462

[journal] O. Russakovsky / 2015 / ImageNet large scale visual recognition challenge / Int. J. Comput. Vis 115(3) : 211~252

[journal] S. Razakarivony / 2016 / Vehicle detection in aerial imagery:A small target detection benchmark / Journal of Visual Communication and Image Representation 34 : 187~203

[other] H. I. Fawaz / 2019 / Deep Neural Network Ensembles for Time Series Classification / International Joint Conference on Neural Networks (IJCNN), arXiv:1605.05462

[other] H. Noh / 2015 / Learning Deconvolution Network for Semantic Segmentation / Computer Vision and Pattern Recognition, arXiv:1505.04366

[other] S. Ren / Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks / Computer Vision and Pattern Recognition Cite as:arXiv:1506.01497

[confproc] W. Liu / 2016 / SSD: Single shot multibox detector / Proc. ECCV 21(37) : 21~37

[confproc] J. Redmon / 2016 / YOLO9000 : Better, Faster, Stronger / Proceedings of the IEEE conference on computer vision and pattern recognition : 779~788

[other] M. Gao / 2018 / Dynamic Zoom-in Network for Fast Object Detection in Large Images / Computer Vision and Pattern Recognition (CVPR), arXiv:1711.05187

[other] V. Růžička / 2018 / Fast and accurate object detection in high resolution 4K and 8K video using GPUs / IEEE High Performance extreme Computing Conference (HPEC), arXiv:1810.10551

[journal] H. Tayara / 2018 / Object Detection in Very High-Resolution Aerial Images Using One-Stage Densely Connected Feature Pyramid Network / Sensors 18(10) : 3341~

[journal] Z. Wu / 2018 / Rapid Target Detection in High Resolution Remote Sensing Images Using YOLO Model / The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences XLII-3

[journal] Y. Ren / 2018 / Small Object Detection in Optical Remote Sensing Images via Modified Faster R-CNN / SCI 8(5) : 813~

[journal] D. M. W. POWERS / 2011 / EVALUATION: FROM PRECISION, RECALL AND F-MEASURE TO ROC, INFORMEDNESS, MARKEDNESS & CORRELATION / Journal of Machine Learning Technologies 2(1) : 37~63

KJCKorea
Journal Central

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

Ensemble Deep Network for Dense Vehicle Detection in Large Image

ABSTRACT

KEYWORDS

Citation status

* References for papers published after 2023 are currently being built.

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

Ensemble Deep Network for Dense Vehicle Detection in Large Image

ABSTRACT

KEYWORDS

Statistics

Tools

Issue List

Citation status

KCI Citation Counts (1)

REFERENCES (20) * References for papers published after 2023 are currently being built.

Search PDF

Citation

* References for papers published after 2023 are currently being built.