视觉注意机制下的粒子窗快速目标检测

徐超; 高敏; 杨锁昌; 方丹; 卢志才

doi:10.3788/OPE.20152311.3227

您当前的位置：

首页 >

文章列表页 >

视觉注意机制下的粒子窗快速目标检测

信息科学 | 更新时间：2020-08-13

- 视觉注意机制下的粒子窗快速目标检测
- Visual attention mechanism-aided fast target detection by particle window
- 光学精密工程 2015年23卷第11期页码：3227-3237
- 作者机构：
  
  军械工程学院导弹工程系,河北石家庄,050003
- 作者简介：
- 基金信息：
  
  军内科技创新项目(No.2013562)();军械工程学院科研基金资助项目(No.YJJXM11018)()
- DOI：10.3788/OPE.20152311.3227
  中图分类号： TP391.4
- 收稿日期：2015-08-10，
  
  修回日期：2015-10-10，
  
  纸质出版日期：2015-11-25
- 稿件说明：
移动端阅览
徐超, 高敏, 杨锁昌等. 视觉注意机制下的粒子窗快速目标检测[J]. 光学精密工程, 2015,23(11): 3227-3237

XU Chao, GAO Min, YANG Suo-chang etc. Visual attention mechanism-aided fast target detection by particle window[J]. Editorial Office of Optics and Precision Engineering, 2015,23(11): 3227-3237
徐超, 高敏, 杨锁昌等. 视觉注意机制下的粒子窗快速目标检测[J]. 光学精密工程, 2015,23(11): 3227-3237 DOI： 10.3788/OPE.20152311.3227.

XU Chao, GAO Min, YANG Suo-chang etc. Visual attention mechanism-aided fast target detection by particle window[J]. Editorial Office of Optics and Precision Engineering, 2015,23(11): 3227-3237 DOI： 10.3788/OPE.20152311.3227.

摘要

针对传统滑动窗目标检测方法需要在全图像范围内穷举搜索的缺点

提出了一种基于视觉注意机制的粒子窗检测方法

旨在保持较高检测精度的同时减少计算量。该方法将目标显著性作为先验知识引入搜索过程

采用"图像签名"方法生成显著图

然后通过阈值门限提取出包含有目标真实位置的局部区域。利用蒙特卡洛采样在显著目标对应的图像范围内均匀生成粒子窗

并依据分类器的响应对粒子进行重采样

以凸显真实目标区域、避免滑动窗方法对搜索步长的依赖。建立了Adaboost+类Harr特征(HLF)和支持向量机(SVM)+方向梯度直方图(HOG)的多级分类器结构

前级分类器用于大范围目标的快速筛选

后级分类器用于小范围目标的精确定位。将本文目标检测模型与传统滑动窗法和粒子窗法进行了比较

结果表明本文方法的受试者工作特征曲线(ROC)包含的面积更大

耗时仅为滑动窗法的1/3到1/4

粒子窗法的1/2

在保持较高检测精度的条件下显著提升了检测速度

实现了快速准确的目标检测。

Abstract

As traditional sliding window detectors need to search the whole image by exhaustive method

a visual attention mechanism-aided target detection model by the particle window is proposed to reduce the calculational load while containing high detection accuracy. This model takes the target saliency as prior information of searching process

and then extracts the region of interest containing true target position by the "Image Signature" saliency map generator and entropy threshold. By uniformly drawing particle windows in an image range corresponding to the saliency targets with Monte Carlo sampling

the local region is treated as candidate detection points

thus resampling is carried out according to corresponding particle windows' response. This strategy only focuses on the areas where the objects are potentially present and avoiding the tradeoff between accuracy and efficiency resulting from searching steps. A multi-stage classifier with Adaboost+HLF and SVM+ HOG is established

the former is applied to once-over and the latter is used to locate precisely. The target detection model proposed is compared with the traditional sliding window method and particle window method

and the results show that the Receiver Operating Characteristic(ROC) curve by proposed method contains the area to be larger than that of the other methods and the time consuming is only 1/3 to 1/4 that of the sliding window method and 1/2 that of the particle window method. It increases significantly detection speeds at maintaining high precision detection speed and achieves fast and accurate target detection.

关键词

Keywords

references

ZHANG Y X, DU B, ZHANG L P. A sparse representation-based binary hypothesis model for target detection in hyperspectral images[J]. IEEE Transactions on Geoscience and Remote Sensing, 2015, 53(3):1346-1354.

VAZQUEZ D, LOPEZ A M, MARIN J, et al.. Virtual and real world adaption for pedestrian detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2014, 36(4):797-809.

黄炎, 范赐恩, 朱秋平, 等. 联合梯度直方图和局部二值模式特征的人体检测[J]. 光学精密工程, 2013, 21(4):1047-1053. HUANG Y, FAN C E, ZHU Q P, et al.. HOG-LBP pedestrian detection[J].Opt. Precision Eng., 2013, 21(4):1047-1053.(in Chinese)

PEDERSOLI M, GONZALEZ J, HU X, et al.. Toward real-time pedestrian detection based on a deformable template model[J]. IEEE Transactions on Intelligence Transportation Systems, 2014, 15(1):355-364.

LAMPERT C H, BLASCHKO M B, HOFMANN T. Efficient subwindow search:a branch and bound framework for object localization[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(12):2129-2142.

YANG B, LEI Y Q. Vehicle detection and classification for low-speed congested traffic with anisotropic magnetoresistive sensor[J].IEEE Sensors Journal, 2015, 15(2):1132-1138.

PEDERSOLI M, GONZALEZ J, BAGDANOV A D, et al.. Recursive coarse-to-fine localization for fast object detection[C]. Proceedings of 11th European Conference on Computer Vision, 2010:280-293.

GUALDI G, PRATI A, CUCCHIARA R. Multistage particle windows for fast and accurate object detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(6):1589-1604.

杨扬, 李善平. 分割位置提示的可变形部件模型快速目标检测[J]. 自动化学报, 2012, 38(4):540-548. YANG Y, LI SH P. Fast object detection with deformable part models and segment locations' hint[J]. Acta Automatic Sinca, 2012, 38(4):540-548.(in Chinese)

MUNDER S, GAVRILA D. An experimental study on pedestrian classification[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006, 28(11):1863-1868.

ENZWEILER M, GAVRILA D. Monocular pedestrian detection:survey and experiments[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2009, 31(12):2179-2195.

孙锐, 陈军, 高隽. 基于显著性检测与HOG-NMF特征的快速行人检测方法[J]. 电子与信息学报, 2013, 35(8):1921-1926. SUN R, CHEN J, GAO J. Fast pedestrian detection based on saliency detection and HOG-NMF features[J].Journal of Electronics & Information Technology, 2013, 35(8):1921-1926.(in Chinese)

SUN W, ZHAO C, SUN M. Learning based particle filtering object tracking for visible-light systems[J]. International Journal for Light and Electron Optics, 2015, 126(19):1830-1837.

杨利平, 辜小花. 用于人脸识别的相对梯度直方图特征描述[J]. 光学精密工程, 2014, 22(1):152-159. YANG L P, GU X H. Relative gradient histogram features for face recognition[J].Opt. Precision Eng., 2014, 22(1):152-159.(in Chinese)

钱生, 陈宗海, 林名强, 等. 基于条件随机场和图像分割的显著性检测[J]. 自动化学报, 2015, 41(4):711-724. QIAN SH, CHEN Z H, LIN M Q, et al.. Saliency detection based on conditional random field and image segmentation[J]. ACTA AUTOMATIC SINCA, 2015, 41(4):711-724.(in Chinese)

修春波, 魏世安. 显著性直方图模型的Camshift跟踪方法[J]. 光学精密工程, 2015, 23(6):1749-1757. XIU CH B, WEI SH A. Camshift tracking with saliency histogram[J]. Opt. Precision Eng., 2015, 23(6):1749-1757.(in Chinese)

黎万义, 王鹏, 乔红. 引入视觉注意机制的目标跟踪方法综述[J]. 自动化学报, 2014, 40(4):561-576. LI W Y, WANG P, QIAO H. A survey of visual attention based methods for object tracking[J].ACTA AUTOMATIC SINCA, 2014, 40(4):561-576.

GUO C, ZHANG L. A novel multiresolution spatiotemporal saliency detection model and its applications in image and video compression[J]. IEEE Transactions on Image Processing, 2010, 19(1):185-198.

ACHANTA R, HEMAMI S, ESTRADA F, et al.. Frequency-tuned salient region detection[C]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition, Miami:FL, 2009:1597-1604.

HOU X D, HAREL J, KOCH C, et al.. Image signature:highlighting sparse salient regions[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2012, 34(1):194-201.

浏览量

614

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

跨层注意力交互下的多特征交叉无人机图像检测

适应大范围星场密度变化的恒星辨识与抑制

利用光谱解混合的目标检测

强化特征提取能力的下颌骨骨折检测3M-YOLOv5网络

面向无人机平台的轻量化目标检测网络