基于MPSOC的航空图像目标检测系统设计

任彬; 王宇庆; 丛振; 聂海涛; 杨航

doi:10.37188/CJLCD.2020-0310

您当前的位置：

首页 >

文章列表页 >

基于MPSOC的航空图像目标检测系统设计

图像处理 | 更新时间：2021-07-13

- 基于MPSOC的航空图像目标检测系统设计
- Design of aerial image target detection systembased on MPSOC
- 液晶与显示 2021年36卷第7期页码：1006-1017
- 作者机构：
  
  1.中国科学院长春光学精密机械与物理研究所, 吉林长春 130033
  2.中国科学院大学, 北京 100049
  3.中国人民解放军联勤保障部队第946医院医学工程科, 吉林长春 130033
- 作者简介：
  
  [ "任彬(1996-), 男, 山东淄博人, 硕士研究生, 2018年于哈尔滨理工大学获得学士学位, 主要从事计算机视觉, 数字图像处理方面的研究。E-mail:Ren1393138227@163.com" ]
  [ "王宇庆(1979-), 男, 吉林长春人, 副研究员, 2008年于中科院长春光学精密机械与物理研究所获得博士学位, 主要研究方向: 基于人工智能的快速目标检测方法、人工智能芯片的研究与应用、面阵三维成像、嵌入式数字图像处理、图像质量评价等。E-mail: wyq7903@163.com" ]
- 基金信息：
  
  国家自然科学基金青年基金(61401425)
- DOI：10.37188/CJLCD.2020-0310
  中图分类号： TP391.4
- 收稿日期：2020-11-22，
  
  修回日期：2020-12-16，
  
  纸质出版日期：2021-07
- 稿件说明：
移动端阅览
任彬, 王宇庆, 丛振, 等. 基于MPSOC的航空图像目标检测系统设计[J]. 液晶与显示, 2021,36(7):1006-1017.

Bin REN, Yu-qing WANG, Zhen CONG, et al. Design of aerial image target detection systembased on MPSOC[J]. Chinese journal of liquid crystals and displays, 2021, 36(7): 1006-1017.
任彬, 王宇庆, 丛振, 等. 基于MPSOC的航空图像目标检测系统设计[J]. 液晶与显示, 2021,36(7):1006-1017. DOI： 10.37188/CJLCD.2020-0310.

Bin REN, Yu-qing WANG, Zhen CONG, et al. Design of aerial image target detection systembased on MPSOC[J]. Chinese journal of liquid crystals and displays, 2021, 36(7): 1006-1017. DOI： 10.37188/CJLCD.2020-0310.

摘要

近年来，航空光学成像技术快速发展，机载图像处理系统对于目标检测精度和检测速度的要求越来越高，传统的目标检测算法已经无法满足要求。与此同时，基于深度学习的目标检测算法凭借更优的性能表现得到了学术界的广泛关注。但这类算法往往参数较多，时间复杂度高且移动端移植困难。针对上述问题，本文提出了一种基于Yolo V3算法的MPSOC平台实现方案。利用改进的k均值聚类算法获取新的初始锚框，之后通过改变特征图的大小提高算法对小目标的检测精度，通过基于敏感度的剪枝方法压缩算法大小，最后利用VISDRONE数据集在MPSOC平台进行了验证。实验结果表明：改善的Yolo算法的MAP提高了1.3%，误检率也得到了极大降低。算法经过压缩后，检测速度提高了1倍，体积仅为原来的37%，基本满足了对航空图像目标检测的设计要求，同时为深度学习算法在MPSOC中实现提供了可行的解决方案。

Abstract

In recent years

the traditional aerial image target detection algorithms have been unable to meet the requirements of detection accuracy and speed

while the rapid development of target detection algorithms based on deep learning provides a new idea for target detection. However

this kind of algorithm is often accompanied by large scale and highly dependent on GPU devices

which makes the migration of the mobile end of the algorithm difficult. Aiming at the above problems

this paper proposes a MPSOC platform implementation scheme based on Yolo V3 algorithm. Firstly

the anchor frame of the original network is re-selected by means of k-means clustering

the detection accuracy of the algorithm is increased by adjusting the convolutional layer

and then the model scale is compressed by sensity-based pruning operation. Finally

VISDRONE data set is used to verify the Xilinx ZYNQ series MPSOC platform. The experimental results show that MAP of the improved Yolo algorithm increases by 1.3%

and the false detection rate is also greatly reduced. After the model is compressed

the detection speed is doubled and the volume becomes 37% of the original. It basically meets the design requirements of aerial image target detection

and provides a feasible solution for the implementation of deep learning algorithm in MPSOC.

关键词

Keywords

references

梁华 , 宋玉龙 , 钱锋 , 等 . 基于深度学习的航空对地小目标检测 . 液晶与显示 , 2018 . 33 ( 9 ): 793 - 800 . http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20183309.0793&graphicAbstract=0 http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20183309.0793&graphicAbstract=0 .

H LIANG , Y L SONG , F QIAN , 等 . Detection of small target in aerial photography based on deep learning . Chinese Journal of Liquid Crystals and Displays , 2018 . 33 ( 9 ): 793 - 800 . http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20183309.0793&graphicAbstract=0 http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20183309.0793&graphicAbstract=0 .

马永杰 , 宋晓凤 . 基于YOLO和嵌入式系统的车流量检测 . 液晶与显示 , 2019 . 34 ( 6 ): 613 - 618 . http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20193406.0613&graphicAbstract=0 http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20193406.0613&graphicAbstract=0 .

Y J MA , X F SONG . Vehicle flow detection based on YOLO and embedded system . Chinese Journal of Liquid Crystals and Displays , 2019 . 34 ( 6 ): 613 - 618 . http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20193406.0613&graphicAbstract=0 http://cjlcd.lightpublishing.cn/thesisDetails#10.3788/YJYXS20193406.0613&graphicAbstract=0 .

裴信彪 , 吴和龙 , 马萍 , 等 . 基于无人机遥感的不同施氮水稻光谱与植被指数分析 . 中国光学 , 2018 . 11 ( 5 ): 832 - 840 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201805015.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201805015.htm .

X B PEI , H L WU , P MA , 等 . Analysis of the spectrum and vegetation index of rice under different nitrogen levels based on unmanned aerial vehicle remote sensing . Chinese Optics , 2018 . 11 ( 5 ): 832 - 840 . https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201805015.htm https://www.cnki.com.cn/Article/CJFDTOTAL-ZGGA201805015.htm .

裴伟 , 许晏铭 , 朱永英 , 等 . 改进的SSD航拍目标检测方法 . 软件学报 , 2019 . 30 ( 3 ): 738 - 758 . https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201903015.htm https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201903015.htm .

W PEI , Y M XU , Y Y ZHU , 等 . The target detection method of aerial photography images with improved SSD . Journal of Software , 2019 . 30 ( 3 ): 738 - 758 . https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201903015.htm https://www.cnki.com.cn/Article/CJFDTOTAL-RJXB201903015.htm .

BOYKOV Y, HUTTENLOCHER D P. A bayesian framework for model based tracking[R]. Place Ithaca, NY, United States: Cornell University, 1999.

ALI S, SHAH M. COCOA: tracking in aerial imagery[C]// Proceedings of SPIE 6209, Airborne Intelligence , Surveillance , Reconnaissance ( ISR ) Systems and Applications Ⅲ. Orlando (Kissimmee), Florida, United States: SPIE, 2006: 62090D.

IBRAHIM A W N, CHING P W, SEET G L G, et al . Moving objects detection and tracking framework for UAV-based surveillance[C]// Proceedings of 2010 Fourth Pacific-Rim Symposium on Image and Video Technology . Singapore: IEEE, 2010.

谭熊 , 余旭初 , 刘景正 , 等 . 基于无人机视频的运动目标快速跟踪 . 测绘通报 , 2011 . ( 9 ): 32 - 34, 41 . https://www.cnki.com.cn/Article/CJFDTOTAL-CHTB201109011.htm https://www.cnki.com.cn/Article/CJFDTOTAL-CHTB201109011.htm .

X TAN , X C YU , J Z LIU , 等 . Object fast tracking based on unmanned aerial vehicle video . Bulletin of Surveying and Mapping , 2011 . ( 9 ): 32 - 34, 41 . https://www.cnki.com.cn/Article/CJFDTOTAL-CHTB201109011.htm https://www.cnki.com.cn/Article/CJFDTOTAL-CHTB201109011.htm .

汤轶 , 周鹏程 , 肖璇 , 等 . 基于无人机平台的运动目标检测与跟踪算法研究 . 机器人技术与应用 , 2017 . ( 3 ): 35 - 37 . DOI: 10.3969/j.issn.1004-6437.2017.03.012 http://doi.org/10.3969/j.issn.1004-6437.2017.03.012 .

Y TANG , P C ZHOU , X XIAO , 等 . Research on moving target detection and tracking algorithm based on UAV platform . Robot Technique and Application , 2017 . ( 3 ): 35 - 37 . DOI: 10.3969/j.issn.1004-6437.2017.03.012 http://doi.org/10.3969/j.issn.1004-6437.2017.03.012 .

李航 , 朱明 . 基于深度卷积神经网络的小目标检测算法 . 计算机工程与科学 , 2020 . 42 ( 4 ): 649 - 657 . DOI: 10.3969/j.issn.1007-130X.2020.04.011 http://doi.org/10.3969/j.issn.1007-130X.2020.04.011 .

H LI , M ZHU . A small object detection algorithm based on deep convolutional neural network . Computer Engineering & Science , 2020 . 42 ( 4 ): 649 - 657 . DOI: 10.3969/j.issn.1007-130X.2020.04.011 http://doi.org/10.3969/j.issn.1007-130X.2020.04.011 .

范丽丽 , 赵宏伟 , 赵浩宇 , 等 . 基于深度卷积神经网络的目标检测研究综述 . 光学精密工程 , 2020 . 28 ( 5 ): 1152 - 1164 . https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM202005019.htm https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM202005019.htm .

L L FAN , H W ZHAO , H Y ZHAO , 等 . Survey of target detection based on deep convolutional neural networks . Optics and Precision Engineering , 2020 . 28 ( 5 ): 1152 - 1164 . https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM202005019.htm https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM202005019.htm .

陈科峻 , 张叶 . 基于YOLO-V3模型压缩的卫星图像船只实时检测 . 液晶与显示 , 2020 . 35 ( 11 ): 1168 - 1176 . DOI: 10.37188/YJYXS20203511.1168 http://doi.org/10.37188/YJYXS20203511.1168 .

K J CHEN , Y ZHNAG . Real-time ship detection in satellite images based on YOLO-v3 model compression . Chinese Journal of Liquid Crystals and Displays , 2020 . 35 ( 11 ): 1168 - 1176 . DOI: 10.37188/YJYXS20203511.1168 http://doi.org/10.37188/YJYXS20203511.1168 .

谢晓竹 , 薛帅 . 改进YOLOv3模型对航拍汽车的目标检测 . 装甲兵工程学院学报 , 2019 . 33 ( 3 ): 84 - 89 . DOI: 10.3969/j.issn.1672-1497.2019.03.015 http://doi.org/10.3969/j.issn.1672-1497.2019.03.015 .

X Z XIE , S XUE . Target detection of aerial photographic vehicles based on improved YOLOv3 model . Journal of Academy of Armored Force Engineering , 2019 . 33 ( 3 ): 84 - 89 . DOI: 10.3969/j.issn.1672-1497.2019.03.015 http://doi.org/10.3969/j.issn.1672-1497.2019.03.015 .

李宇 , 刘雪莹 , 张洪群 , 等 . 基于卷积神经网络的光学遥感图像检索 . 光学精密工程 , 2018 . 26 ( 1 ): 200 - 207 . https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201801025.htm https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201801025.htm .

Y LI , X Y LIU , H Q ZHANG , 等 . Optical remote sensing image retrieval based on convolutional neural networks . Optics and Precision Engineering , 2018 . 26 ( 1 ): 200 - 207 . https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201801025.htm https://www.cnki.com.cn/Article/CJFDTOTAL-GXJM201801025.htm .

HU H Y, PENG R, TAI Y W, et al . Network trimming: a data-driven neuron pruning approach towards efficient deep architectures[J]. arXiv : 1607.03250, 2016.

LI H, KADAV A, DURDANOVIC I, et al . Pruning filters for efficient ConvNets[C]// Proceedings of the International Conference on Learning Representations . Toulon, France: ICLR, 2017.

Xilinx. Zynq UltraScale+ MPSoC product tables and product selection guide[R]. Amrerican: Xilinx, 2016.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于改进YOLOv5s的手术器械检测与分割方法

基于注意力机制的复杂背景下红外弱小目标检测方法研究

基于级联特征和图卷积的三维手部姿态估计算法

复杂海背景下的自适应舰船目标检测