基于卷积神经网络的候选区域优化算法
Region proposal optimization algorithm based on convolutional neural networks
查看参考文献24篇
|
文摘
|
在目标检测中,通常使用候选区域提高目标的检测效率。为解决当前候选区域质量较低的问题,本文将卷积边缘特征、显著性及目标位置信息引入到候选区域算法中。首先,利用卷积神经网络将待检测图像生成更富有语义信息的边缘特征,并通过边缘点聚合及边缘组相似性策略,获取每个滑动窗口的边缘信息得分;其次,利用显著性目标的局部特征,统计每个滑动窗口中的目标显著性得分;第三,根据目标可能出现的位置,计算每个滑动窗口中的目标位置信息得分;最后,利用边缘信息、显著性及位置信息的分数确定候选区域。在PASCAL VOC 2007验证集上进行实验,给定10 000个候选区域,交并比取0.7时,所提算法的召回率为90.50%,较Edge Boxes算法提高了3%。每张图像的运行时间大约为0.76 s。结果表明,本文算法可快速产生较高质量的候选区域。 |
|
其他语种文摘
|
Region proposals are usually used to efficiently detect objects in object detection. In order to solve the problem that the region proposals have low quality, the convolutional edge features, object saliency and position information of objects are introduced into the region proposals algorithm. Firstly, the edge features with semantically meaningful information are generated from the images to be detected using the convolutional neural networks, and the score of edge information for per sliding window is obtained through the strategy of edge clustering and the similarities between the edge groups. Then, the salient object scores of each sliding window are computed using the local features of salient objects. Thirdly,the scores of object position information are calculated according to the location where objects may occur. Finally, the region proposals are determined by three components including edge information scores, salient object scores and the object positions scores. The experimental results in PASCAL VOC 2007 validation set show that given just 10 000 region proposals, the object recall of the proposed algorithm is 90.50%,that is increased by 3% comparing with Edge Boxes with intersection over union threshold of 0.7. The run time of the proposed method is about 0.76 seconds for processing one image,and this demonstrates that our approach can yield a set of region proposals with higher quality at a fast speed. |
|
来源
|
中国光学（中英文）
,2019,12(6):1348-1361 【核心库】
|
|
DOI
|
10.3788/co.20191206.1348
|
|
关键词
|
计算机视觉
;
目标检测
;
候选区域
;
卷积神经网络
;
显著性目标
|
|
地址
|
1.
中国科学院国家空间科学中心, 中国科学院复杂航天系统电子信息技术重点实验室, 北京, 100190
2.
中国科学院大学, 北京, 100049
3.
长春大学, 吉林, 长春, 130022
|
|
语种
|
中文 |
|
文献类型
|
研究性论文 |
|
ISSN
|
2095-1531 |
|
学科
|
自动化技术、计算机技术 |
|
基金
|
国家自然科学基金
|
|
文献收藏号
|
CSCD:6639781
|
参考文献 共
24
共2页
|
|
1.
梁华. 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
|
被引
18
次
|
|
|
|
|
2.
李艳荻. 基于超像素时空特征的视频显著性检测方法.
光学学报,2019,39(1):1-8
|
被引
1
次
|
|
|
|
|
3.
Fattal A. Saliency-guided region proposal network for CNN based object detection.
IEEE Conference on Intelligent Transportation Systems,2017:1-8
|
被引
1
次
|
|
|
|
|
4.
Uijlings J. Selective search for object recognition.
International Journal of Computer Vision,2013,104(2):154-171
|
被引
418
次
|
|
|
|
|
5.
Alexe B. Measuring the objectness of image windows.
IEEE Transactions on Software Engineering,2012,34(11):2189-2202
|
被引
69
次
|
|
|
|
|
6.
Cheng M M. BING: Binarized Normed Gradients for Objectness Estimation at 300fps.
2014 IEEE Conference on Computer Vision and Pattern Recognition. 1,2014:3286-3293
|
被引
1
次
|
|
|
|
|
7.
Zitnick C L. Edge boxes:locating object proposals from edges.
Proceedings of 13th European Conference on Computer Vision. 8689,2014:391-405
|
被引
1
次
|
|
|
|
|
8.
Jiang S. Class agnostic image common object detection.
EEE Transactions on Image Processing,2019,28(6):2836-2846
|
被引
1
次
|
|
|
|
|
9.
He K M. Mask R-CNN.
2017 IEEE International Conference on Computer Vision, Venice,Italy,2017:2980-2988
|
被引
1
次
|
|
|
|
|
10.
丁鹏. 基于视觉显著性的海面舰船检测技术.
电子学报,2018,46(1):127-134
|
被引
13
次
|
|
|
|
|
11.
李宇. 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
|
被引
29
次
|
|
|
|
|
12.
Liu Y. Richer convolutional features for edge detection.
2017 IEEE Conference on Computer Vision and Pattern Recognition. 1,2017:5872-5881
|
被引
1
次
|
|
|
|
|
13.
Ishikura K. Saliency detection based on multiscale extrema of local perceptual color differences.
IEEE Transactions on Image Processing,2018,27(2):703-717
|
被引
6
次
|
|
|
|
|
14.
Kuang P J. Improved edge boxes with object saliency and location awards.
IEICT Transactions on Information and Systems,2016,E99D(2):488-495
|
被引
1
次
|
|
|
|
|
15.
Carreira J. CPMC: automatic object segmentation using constrained parametric min-cuts.
IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(7):1312-1328
|
被引
32
次
|
|
|
|
|
16.
邝沛江.
基于改进Edge Boxes的物体检测算法的研究,2017
|
被引
2
次
|
|
|
|
|
17.
Manen S. Prime object proposals with randomized prims algorithm.
Proceedings of the 2013 IEEE International Conference on Computer Vision,2013:2536-2543
|
被引
2
次
|
|
|
|
|
18.
Rantalankila P. Generating object segmentation proposals using global and local search.
2014 IEEE Conference on Computer Vision and Pattern Recognition,2014:2417-2424
|
被引
1
次
|
|
|
|
|
19.
Pont-Tuset J. Multiscale combinatorial grouping for image segmentation and object proposal generation.
IEEE Transactions on Pattern Analysis and Machine Intelligence,2017,39(1):128-140
|
被引
34
次
|
|
|
|
|
20.
Endres I. Category-independent object proposals with diverse ranking.
IEEE Transactions on Pattern Analysis and Machine Intelligence,2014,36(2):222-234
|
被引
13
次
|
|
|
|
|
了解气象卫星更多信息，请访问