基于多级定向执行长度编码快速识别物体特征

王欢欢; 卞越新; 王向军

doi:10.3788/OPE.20142207.1976

您当前的位置：

首页 >

文章列表页 >

基于多级定向执行长度编码快速识别物体特征

信息科学 | 更新时间：2020-08-13

- 基于多级定向执行长度编码快速识别物体特征
- Fast test and recognition of object characters based on multilevel orientation run length coding
- 光学精密工程 2014年22卷第7期页码：1976-1983
- 作者机构：
  
  1. 天津大学微光机电系统技术教育部重点实验室天津,300072
  2. 天津科技大学机械工程学院天津,300222
- 作者简介：
- 基金信息：
  
  国家教育部支撑项目（No.625010110）();天津科技大学引进人才科研启动基金资助项目（No.201304003）();微光机电系统技术教育部重点实验室（天津大学
- DOI：10.3788/OPE.20142207.1976
  中图分类号： TP391
- 收稿日期：2014-04-22，
  
  修回日期：2014-05-15，
  
  纸质出版日期：2014-07-25
- 稿件说明：
移动端阅览
王欢欢, 卞越新, 王向军. 基于多级定向执行长度编码快速识别物体特征[J]. 光学精密工程, 2014,22(7): 1976-1983

WANG Huan-huan, BIAN Yue-xin, WANG Xiang-jun. Fast test and recognition of object characters based on multilevel orientation run length coding[J]. Editorial Office of Optics and Precision Engineering, 2014,22(7): 1976-1983
王欢欢, 卞越新, 王向军. 基于多级定向执行长度编码快速识别物体特征[J]. 光学精密工程, 2014,22(7): 1976-1983 DOI： 10.3788/OPE.20142207.1976.

WANG Huan-huan, BIAN Yue-xin, WANG Xiang-jun. Fast test and recognition of object characters based on multilevel orientation run length coding[J]. Editorial Office of Optics and Precision Engineering, 2014,22(7): 1976-1983 DOI： 10.3788/OPE.20142207.1976.

摘要

提出了一种新的快速物体检测与识别方法，用于检测和识别在运动过程中图像尺寸不断变化的物体。该方法基于多级定向执行长度编码法（MORLC）生成一种新的多级字节编码和物体特征样本形式，其属于字符数据样本，具有数据量小、占用存储空间少、构造过程简单、可根据不同的应用需求选择不同级次的样本形式等特点。对MORLC方法中二级字节编码和样本的系统理论进行了分析，给出了利用MORLC二级字节编码和样本检测和识别运动物体的实验数据。对静态背景中的行人及位置进行了检测和识别实验，结果显示多级字节编码和物体特征样本形式匹配处理速度快，匹配和识别的鲁棒性好，不易产生错误匹配。该方法特别适用于场景中由于被探测物体运动产生投影图像比例发生变化的情况，即对图像大小不敏感。该方法的另一个突出优点是高的执行处理效率。

Abstract

A new rapid detection and recognition method was proposed for those objects whose image sizes would change in movement processing.Based on Multilevel Orientation Run Length Coding (MORLC) method

the method can generate a new multilevel byte code and object feature templet forms.It is a byte coding and templets and characterized by smaller data amounts

smaller storage addresses

easy constructing

and can choose different classes of sample forms.According to different application requirements

two level byte code in MORLC method and system theory of templets were analyzed

and the experimental data from detection and recognition of moving objects using MORLC two level byte code were given.A detection and recognition experiment on a pedestrian and its position in a static background was performed.The experimental results show that the multilevel byte code and object feature matching templet form have higher processing speeds

robust matching and identification

and not easily get wrong matching.This method is especially suitable for changed images caused by detecting object movement in the scene projection

which means it is not sensitive to image sizes.Another advantage of this method is high processing efficiency.

关键词

Keywords

references

SUGATA G,MEHROTRA R S. Detection of composite edges[J]. IEEE Transactions on Image Processing,1994,3(1):14-25.

NARENDRA A. Transform for multiscale image segmentation by integrated edge and region detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1996,18(12):1211-1235.

MOSALAM E,MAYOL C,WALTERIO W. Adaptive sampling for feature detection,tracking,and recognition on mobile platforms[J]. IEEE Transactions on Circuits and Systems for Video Technology,Special Issue on Video Analysis on Resource-Limited Systems,2011,21(10):1467-1475.

MARK S H. Geometric moments and their invariants[J]. Journal of Mathematical Imaging and Vision,2012,44(3):223-235.

SAYYOURI M,ABDESLAM H. Improving the performance of image classification by Hahn moment invariant[J]. Journal of the Optical Society of America A:Optics and Image Science,and Vision,2013,30(11):2381-2394.

KARAKASIS E G,PAPAKOSTAS G A,KOULOURIOTIS D E. Generalized dual Hahn moment invariants[J]. Pattern Recognition,2013,46(7):1998-2014.

赵立荣,朱玮,曹永刚. 改进的SURF算法在特征匹配中的应用[J]. 光学精密工程,2013,21(12):3263-3271. ZHAO L R,ZHU W,CAO Y G. Application of improved SURF algorithm to feature matching[J]. Opt. Precision Eng. ,2013,21(12) 3263-3271. (in Chinese)

王欢欢,杨清平,王向军. 物体特征的快速检测与识别方法[J]. 红外与激光工程,2014,43(3):1002-1008. WANG H H,YANG Q P,WANG X J. A fast test and recognition method for object character[J]. Infrared and Laser Engineering,2014,43(3):1002-1008. (in Chinese)

WEN P C,WANG X J,WEI H. Modified level set method with Canny operator for image noise removal[J]. Chinese Optics Letter,2010,8(12):1127-1132.

QIU P H,CHEN XING C. Feature based image registration using non-degenerate pixels[J]. Signal Processing,2013,93(4):706-720.

VRIGKAS M,NIKOU C,KONDI L P. Accurate image registration for MAP image super-resolution[J]. Signal Processing:Image Communication,2013,28(5):494-508.

朱娟娟,郭宝龙. 复杂场景中基于变块差分的运动目标检测[J]. 光学精密工程,2011,19(1):183-189. ZHU J J,GUO B L. Moving object detection based on variable block differene in complex scenes[J]. Opt. Precision Eng. ,2011,19(1):183-189.

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

基于游程和扩展指数哥伦布编码的任意形状感兴趣区域图像编码

AC预测改进图像编码计算机实时仿真