应用级联分类器检测安瓿内弱小运动目标

秦垚; 王伯雄; 李伟; 杨春毓

doi:10.3788/OPE.20122001.0190

您当前的位置：

首页 >

文章列表页 >

应用级联分类器检测安瓿内弱小运动目标

信息科学 | 更新时间：2020-08-12

- 应用级联分类器检测安瓿内弱小运动目标
- Inspection of small moving foreign substances in ampoule based on cascade classifiers
- 光学精密工程 2012年20卷第1期页码：190-196
- 作者机构：
  
  清华大学精密仪器与机械学系精密测试技术及仪器国家重点实验室北京,100084
- 作者简介：
- 基金信息：
  
  国家863高技术研究发展计划资助项目(No.2008AA042207)()
- DOI：10.3788/OPE.20122001.0190
  中图分类号： TP391.41
- 收稿日期：2011-07-20，
  
  修回日期：2011-08-18，
  
  网络出版日期：2012-01-25，
  
  纸质出版日期：2012-01-25
- 稿件说明：
移动端阅览
秦垚, 王伯雄, 李伟, 杨春毓. 应用级联分类器检测安瓿内弱小运动目标[J]. 光学精密工程, 2012,20(1): 190-196

QIN Yao, WANG Bo-xiong, LI Wei, YANG Chun-yu. Inspection of small moving foreign substances in ampoule based on cascade classifiers[J]. Editorial Office of Optics and Precision Engineering, 2012,20(1): 190-196
秦垚, 王伯雄, 李伟, 杨春毓. 应用级联分类器检测安瓿内弱小运动目标[J]. 光学精密工程, 2012,20(1): 190-196 DOI： 10.3788/OPE.20122001.0190.

QIN Yao, WANG Bo-xiong, LI Wei, YANG Chun-yu. Inspection of small moving foreign substances in ampoule based on cascade classifiers[J]. Editorial Office of Optics and Precision Engineering, 2012,20(1): 190-196 DOI： 10.3788/OPE.20122001.0190.

摘要

针对序列图像内具有低信噪比和低对比度特征的运动目标

提出了一种基于级联分类器的弱小目标检测算法。该算法从安瓿瓶序列图像内提取绝对差分值、局部差分对比度和局部相关系数3个图像特征。每个图像特征对应一个分类器

通过三层级联形式实现序列图像中的小目标检测。第一个节点与传统帧间差分法类似

主要去除大量背景图像并检测出大颗粒运动目标

后两个节点则用于检测弱小目标、排除光流和瓶身污渍产生的噪声点。实验结果显示

相对于传统的帧间差分法

本文算法具有高检测精度和高抗干扰能力等特点

不仅可以检测出图像中弱小运动目标

同时也消除了复杂背景下的噪声影响

弱小目标的检出率达到99.3%

并且满足安瓿在线检测的实时性要求。

Abstract

An inspection algorithm based on cascade classifiers is presented for detecting small moving foreign substances with low Signal and Noise Ratio (SNR) and low contrast in sequential images. The algorithm obtains three features of absolute difference

local difference contrast and neighborhood correlation from the sequential images of an ampoule. Each feature corresponds to a classifier

and small foreign substances are inspected by using three-layer cascade classifiers. The first layer corresponds to a traditional frame differencing method

which is used to remove the background and detect the large moving foreign substances. The next two layers are used to inspect small foreign substances and remove the noises generated by optical flow and the stain of bottle. Experiment results show that compared with the traditional frame differencing method

this algorithm has higher detection precision and higher anti-interference ability in inspecting small substances with the interference of a complex background

and the detection rate of small foreign substance is 99.3%. This algorithm can meet the requirement of real-time detection of ampoules for medicine production.

关键词

Keywords

references

ELGAMMAL A, DURAISWMI R, HARWOOD D. Background and foreground modeling using nonparametric kernel density estimation for visual surveillance[J]. IEEE, 2002, 7(90): 1152-1163.[2] 邱道尹,张文静,顾波,等. 帧差法在运动目标实时跟踪中的应用 [J].华北水利水电学院学报, 2009,30(3):45-46. QIU D Y,ZHANG W J,GU B, et al.. Application of frame differece methods in real-time moving target tracking[J]. Journal of North China Institute of Water Conservancy and Hydroelectirc Power, 2009,30(3):45-46. (in Chinese)[3] SMITH S M. ASSET-2:Real-time motion segmentation and shape tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence. 1995, 17(8):237-244.[4] 张长城,杨德贵,王宏强. 红外图像中弱小目标检测前跟踪算法研究综述 [J]. 激光与红外, 2007,37(2):104-107. ZHANG CH CH,YANG D G,WANG H Q. Algorithm surveys for dim targets track-before-detect in infrared image[J]. Laser & Infrared, 2007,37(2):104-107. (in Chinese)[5] AKIRA I,TAKAYUKI M, et al.. Detection of foreign substances mixed in a plastic bottle of medicinal solution using real-time video image processing . Pattern Recognition, Proc 14th International Conf, 1998,2:1646-1650.[6] 温江涛,王伯雄. 应用小波包能量谱及支持向量机实现安瓿内浮类异物的识别 [J].光学精密工程, 2009,17(11): 2794-2799. WEN J T,WANG B X. Recognition of floating particles in ampoules by wavelet packet energy spectrum and SVM[J].Opt. Precision Eng., 2009,17(11): 2794-2799. (in Chinese)[7] 周博文,王耀南,葛继,等. 基于机器视觉的医药注射剂智能检测系统研究 [J]. 机器人,2009,31(1):52-60. ZHOU B W, WANG Y N, GE J, et al.. A Machine-Vision-Based intelligent inspection system for pharmaceutical injections[J]. ROBOT, 2009,31(1):52-60. (in Chinese)[8] 李杨果,王耀南,王威. 基于机器视觉的大输液智能灯检机研究 [J]. 光电工程,2006,33(11):69-74. LI Y G, WANG Y N, WANG W. Intelligent transfusion liguor inspector based on machine-vision[J].Opt. Precision Eng.,2006,33(11):69-74. (in Chinese)[9] 丁雪梅,王维雅,黄向东. 基于差分和特征不变量的运动目标检测与跟踪 [J].光学精密工程, 2007,15(4): 570-576. DING X M, WANG W Y, HUANG X D. New method for detecting and tracking of moving target based on difference and invariant[J]. Opt. Precision Eng., 2007,15(4): 570-576. (in Chinese)[10] VIOLA P, JONES M. Fast and robust classification using asymmetric adaboost and a detector cascade[J]. Advances in Neural Information Processing System, 2001,2:1311-1318.

浏览量

458

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于序列图像提高光斑质心定位精度

基于双目标模型的时空上下文跟踪算法