[1]高 群,朱 均,王芊芊,等.基于鱼眼图像的目标检测算法研究[J].控制与信息技术(原大功率变流技术),2019,(03):43-47.[doi:10.13889/j.issn.2096-5427.2019.03.100]
 GAO Qun,ZHU Jun,WANG Qianqian,et al.Research on the Object Detection Algorithm Based on Fisheye Image[J].High Power Converter Technology,2019,(03):43-47.[doi:10.13889/j.issn.2096-5427.2019.03.100]
点击复制

基于鱼眼图像的目标检测算法研究()
分享到:

《控制与信息技术》(原《大功率变流技术》)[ISSN:2095-3631/CN:43-1486/U]

卷:
期数:
2019年03期
页码:
43-47
栏目:
人工智能技术与应用
出版日期:
2019-06-05

文章信息/Info

Title:
Research on the Object Detection Algorithm Based on Fisheye Image
文章编号:
2096-5427(2019)03-0043-05
作者:
高 群1朱 均2王芊芊1曹 杰3许 超2
(1.国网浙江温岭市供电有限公司,浙江温岭 317500;2.浙江大学控制科学与工程学院,浙江 杭州 310000; 3.温岭市非普电气有限公司,浙江温岭 317500)
Author(s):
GAO Qun1 ZHU Jun2 WANG Qianqian1 CAO Jie3 XU Chao2
( 1. State Grid Zhejiang Wenling Power Supply Co., Ltd., Wenling, Zhejiang 317500, China; 2. College of Control Science and Engineering, Zhejiang University, Hangzhou, Zhejiang 310000, China; 3. Wenling Feipu Electric Co., Ltd., Wenling, Zhejiang 317500, China )
关键词:
目标检测鱼眼图像深度学习畸变矫正嵌入式特征提取
Keywords:
object detection fisheye image deep learning distortion correction embedded system feature extraction
分类号:
TN911.73
DOI:
10.13889/j.issn.2096-5427.2019.03.100
文献标志码:
A
摘要:
鱼眼图像在使用前一般会先进行畸变矫正,但畸变严重图像的矫正会降低图像质量。为了提高目标检测精度与速度,文章提出了一种利用单个下视鱼眼摄像头代替多个普通摄像头的目标检测方案。其采用特征金字塔结构检测多尺度物体,并结合下视鱼眼的旋转与畸变特性进行算法优化,直接在原始鱼眼图像上进行目标检测。通过构建下视鱼眼数据集并进行实验,结果显示,所提出的鱼眼目标检测模型不仅精度较高,而且还能在嵌入式设备上快速运行。
Abstract:
Fisheye image is usually corrected before it is used, but image correction with serious distortion will reduce image quality. In order to improve the accuracy and speed of object detection, this paper presented a method which adopts a single bottom-view fisheye camera instead of multiple common cameras. It uses the feature pyramid structure to detect multi-scale objects, and combines the rotation and distortion characteristics of the fisheye to optimize the algorithm and directly detects the objects from the original fisheye image. In this paper, a downside fisheye dataset was constructed and experimented. The experimental results show that the fisheye detection model has high accuracy and can run fast on embedded devices.

参考文献/References:

[1] REN S , HE K , GIRSHICK R , et al. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2015, 39(6):1137-1149.
 [2] DAI J , LI Y , HE K , et al. R-FCN: Object Detection via Region-based Fully Convolutional Networks[C]// Proceedings of the 30th International Conference on Neural Information Processing Systems. Barcelona, Spain,2016:379-387 .
[3] HE K, ZHANG X, REN S, et al. Deep Residual Learning for Image Recognition[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Las Vegas, USA,2016:770-778.
[4] LIU W, ANGUELOV D, ERHAN D, et al. SSD: Single Shot MultiBox Detector[J]. Proceedings of European Conference on Computer Vision ,2016:21-37.
 [5] REDMON J, DIVVALA S , GIRSHICK R, et al. You Only Look Once: Unified, Real-Time Object Detection[C]// Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2015:779-788.
[6] REDMON J, FARHADI A. YOLO9000: Better, Faster, Stronger[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR).Honolulu, USA, 2017:6517-6525.
 [7] REDMON J, FARHADI A. YOLOv3: An Incremental Improvement[J/OL]. arXiv,2018:1-8[2018-12-01]. https: //arxiv. org/pdf/1804.02767.pdf.
[8] HUANG J, RATHOD V, SUN C, et al. Speed/accuracy trade-offs for modern convolutional object detectors[C]// IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Honolulu, USA,2017:3296-3297.
 [9] IOFFE S, SZEGEDY C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C]// International Conference on International Conference on Machine Learning. Lille, France, 2015:448-456.
[10] SZEGEDY C, VANHOUCKE V, IOFFE S, et al. Rethinking the Inception Architecture for Computer Vision[C]//IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, USA,2016:2818-2826.
 [11] SZEGEDY C, IOFFE S, VANHOUCKE V, et al. Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning[C]//Proceedings of the Thirty-First AAAI Conference on Artificial Intelligence .California ,USA, 2016:4278-4284.
 [12] HOWARD A G, ZHU M, CHEN B, et al. MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications[J/OL]. arXiv, 2017: 1-9[2018-12-01]. https://arxiv. org/pdf/1704.04861.pdf.
 [13] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal Loss for Dense Object Detection[J]. IEEE Transactions on Pattern Analysis & Machine Intelligence, 2017(1):2999-3007.
 [14] GIRSHICK R. Fast R-CNN[C]//IEEE International Conference on Computer Vision (ICCV).Santiago, Chile, 2015:1440-1448.
 [15] TIELEMAN T,HINTON G. Double Sarsa and Double Expected Sarsa with Shallow and Deep Learning[J]. Journal of Data Analysis and Information Processing,2016,4(4):26-31.

相似文献/References:

[1]丁 驰,林 军,游 俊,等.基于深度学习的手势识别方法[J].控制与信息技术(原大功率变流技术),2018,(06):96.[doi:10.13889/j.issn.2096-5427.2018.06.016]
 DING Chi,LIN Jun,YOU Jun,et al.A Gesture Recognition Method Based on Deep Learning[J].High Power Converter Technology,2018,(03):96.[doi:10.13889/j.issn.2096-5427.2018.06.016]
[2]高 群,朱 均,王芊芊,等. 基于鱼眼图像的目标检测算法研究[J].控制与信息技术(原大功率变流技术),2019,(03):1.[doi:10.13889/j.issn.2096-5427.2019.03.100]
 GAO Qun,ZHU Jun,WANG Qianqian,et al.Research on the Object Detection Algorithm Based on Fisheye Image[J].High Power Converter Technology,2019,(03):1.[doi:10.13889/j.issn.2096-5427.2019.03.100]

备注/Memo

备注/Memo:
收稿日期:2018-12-13
作者简介:高群(1979—),男,高级工程师,主要从事信息、通信和电力传输研究工作。
基金项目:国家自然科学基金面上项目(61473253)
更新日期/Last Update: 2019-06-14