首页 > 论文 > 激光与光电子学进展 > 55卷 > 6期(pp:61002--1)

基于局部特征的多目标图像分割算法

A Multi-Object Image Segmentation Algorithm Based on Local Features

王琳   刘强  
  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

近年来, 智能机器人技术逐步成熟, 以目标识别为代表的机器视觉技术是其核心。现有目标识别算法通常先根据颜色信息分割场景, 再提取特征以识别目标。但对于颜色信息比较复杂的场景, 往往存在过分割现象, 影响后续目标识别。针对这一问题, 提出一种基于局部特征的多目标图像分割算法。该算法使用双目摄像头采集场景图像, 对场景图像进行预处理, 同时通过立体匹配得到场景的深度信息;结合深度信息确定目标区域;设计动态阈值的尺度不变特征变换(SIFT)算法以提取目标区域的局部特征, 将局部特征转化为特征约束;基于区域约束、特征约束和空间信息组成的特征向量进行聚类分割, 得到最终分割结果, 同时实现对每个目标区域的识别。实验结果表明, 在颜色特征复杂的场景中, 本文算法的整体误差率小于10%, 与已有算法相比降低了15%以上。

Abstract

Intelligent robot has recently matured in industry, whose core technology is machine vision, especially object recognition. In existing object recognition methods, scenes are segmented based on color, and features are then extracted to recognize objects. However, over segmentation exists for scenes with complex color features, which influences subsequent object recognition process. To deal with the over segmentation problem, a multi-object image segmentation algorithm based on local features is proposed, which uses binocular camera to collect scene images. Firstly, the scene image is preprocessed. The depth information of the scene is then obtained by stereo matching, and is used to determine the target area. Secondly, the local features of the target region are extracted by a scale-invariant feature transform (SIFT) algorithm with dynamic threshold, and the local features are then transformed into feature constraints. Finally, the feature vectors, which are based on region constraint, feature constraint and spatial information, are used for clustering segmentation to obtain the final segmentation result. Simultaneously, each object region is recognized. The experiment results show that the overall error rate of the proposed algorithm is less than 10% for a scene with complex color features, and is reduced by 15% compared with those of the existing algorithms.

Newport宣传-MKS新实验室计划
补充资料

中图分类号:TP391.4

DOI:10.3788/LOP55.061002

所属栏目:图像处理

基金项目:国家自然科学基金(61574099)

收稿日期:2017-11-02

修改稿日期:2017-11-24

网络出版日期:--

作者单位    点击查看

王琳:天津大学微电子学院, 天津 300072天津市成像与感知微电子技术重点实验室, 天津 300072
刘强:天津大学微电子学院, 天津 300072天津市成像与感知微电子技术重点实验室, 天津 300072

联系人作者:刘强(qiangliu@tju.edu.cn)

备注:王琳(1992-), 男, 硕士研究生, 主要从事图像处理、硬件加速方面的研究。E-mail: devil@tju.edu.cn

【1】Bai L B. Research on vision guidance method of industrial robot[D]. Shenzhen: Shenzhen University, 2016.
白龙彪. 工业机器人的视觉引导方法研究[D]. 深圳: 深圳大学, 2016.

【2】Wu Q Q. Application of mean shift algorithm in color image filtering and segmentation[D]. Xi′an: Shaanxi Normal University, 2015.
吴琴琴. Mean Shift算法在彩色图像滤波与分割中的应用[D]. 西安: 陕西师范大学, 2015.

【3】Feng J M, Lu Z M, Yang P, et al. Color image segmentation algorithm based on local priority[J]. Journal of Huazhong University of Science and Technology (Natural Science Edition), 2014, 42(9): 44-48.
冯进玫, 卢志茂, 杨朋, 等. 基于局部占优度的彩色图像分割算法[J]. 华中科技大学学报(自然科学版), 2014, 42(9): 44-48.

【4】Lie J, Wang L H, He Y Q, et al. Image segmentation method for robot vision[J]. System Engineering and Electronics, 2017, 39(7): 1653-1659.
雷俊, 王立辉, 何芸倩, 等. 适用于机器人视觉的图像分割方法[J]. 系统工程与电子技术, 2017, 39(7): 1653-1659.

【5】Wang Z Y, Li G, Hou Z Q, et al. Color image segmentation by fusion of mean shift and region saliency[J]. Computer Engineering and Applications, 2010, 46(28): 181-184.
王朝英, 李光, 侯志强, 等. 融合mean shift和区域显著性的彩色图像分割算法[J]. 计算机工程与应用, 2010, 46(28): 181-184.

【6】Shelhamer E, Long J, Darrell T. Fully convolutional networks of semantic segmentation[J]. IEEE Transactions on Pattern Recognition and Machine Intelligence, 2017, 39(4): 640-651.

【7】Feng J W, Zhang L M, Deng X Y.Image segmentation based on multi-source fusion FCN[J/OL]. (2017-10-10)[2017-10-20]. http:∥kns.cnki.net/kcms/detail/51.1196.TP.20171010.1732.110.html.
冯家文, 张立民, 邓向阳. 基于多源融合FCN的图像分割[J/OL]. (2017-10-10) [2017-10-20]. http:∥kns.cnki.net/kcms/detail/51.1196.TP.20171010.1732.110.html.

【8】Zhang Q, Lu S Q, Li H B, et al. Research on underwater stereo matching method based on color segmentation[J]. Acta Optica Sinica, 2016, 36(8): 0815001.
张强, 卢士强, 李海滨, 等. 基于色彩分割的水下立体匹配算法的研究[J]. 光学学报, 2016, 36(8): 0815001.

【9】Huo X Y, Liu Y. A stereovision measurement method using epipolar constraint to correct digital image correlation matching[J]. Chinese Journal of Lasers, 2017, 44(8): 0804003
霍晓洋, 刘洋. 一种极线约束修正数字图像相关匹配的立体视觉测量方法[J]. 中国激光, 2017, 44(8): 0804003

【10】Bai T Z, Hou X B. An improved image matching algorithm based on SIFT[J]. Transactions of Beijing Institute of Technology, 2013, 33(6): 622-627
白廷柱, 侯喜报. 基于SIFT算子的图像匹配算法研究[J]. 北京理工大学学报, 2013, 33(6): 622-627.

【11】Yang S, Xia M H, Zheng Z S. Medical image registration algorithm based on polynomial deterministic matrix and SIFT transform[J]. Laser & Optoelectronics Progress, 2016, 53(8): 081002.
杨飒, 夏明华, 郑志硕. 基于多项式确定性矩阵的SIFT医学图像配准算法[J]. 激光与光电子学进展, 2016, 53(8): 081002.

【12】Yu Z J, Wang S B. Improved PCA-SIFT algorithm for matching stereo system[J]. Laser & Optoelectronics Progress, 2016, 53(3): 031501.
于之靖, 王韶彬. 改进PCA-SIFT 算法的立体匹配系统[J]. 激光与光电子学进展, 2016, 53(3): 031501.

【13】Zhang X, Jin Y X, Xue D. Image matching algorithm based on SICA-SIFT and particle swarm optimization[J]. Laser & Optoelectronics Progress, 2017, 54(9): 091002.
张鑫, 靳雁霞, 薛丹. SICA-SIFT和粒子群优化的图像匹配算法[J]. 激光与光电子学进展, 2017, 54(9): 091002.

【14】Pan C, Cui F. Color image segmentation based on spatial and temporal difference sampling[J]. Computer Engineering, 2012, 38(13): 199-201.
潘晨, 崔凤. 基于空间和时间差别采样的彩色图像分割[J]. 计算机工程, 2012, 38(13): 199-201.

引用该论文

Wang Lin,Liu Qiang. A Multi-Object Image Segmentation Algorithm Based on Local Features[J]. Laser & Optoelectronics Progress, 2018, 55(6): 061002

王琳,刘强. 基于局部特征的多目标图像分割算法[J]. 激光与光电子学进展, 2018, 55(6): 061002

被引情况

【1】徐超,平雪良. 基于改进随机Hough变换的直线检测算法. 激光与光电子学进展, 2019, 56(5): 51001--1

【2】路文超,庞彦伟,何宇清,王建. 基于可分离残差模块的精确实时语义分割. 激光与光电子学进展, 2019, 56(5): 51005--1

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF