基于改进模板匹配算法的物料计数方法研究

doi:10.3969/j.issn.1000-1158.2022.07.05

摘要
图/表
参考文献(24)
相关文章 (15)

全文: PDF (1056 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要针对生产过程中物料计数方法存在的缺陷,研究了一种改进的模板匹配算法进行物料数量检测。通过基于双模板搜索方式和位置约束系数r改进传统模板匹配算法(template matching algorithm,TMA),优化寻找最佳匹配图像的过程获得最优解,并以此进行图像拼接,完成物料计数。与传统模板匹配算法(TMA)和高斯金字塔优化的模板匹配算法(Gaussian pyramid transformation,GPT)进行实验对比,证明所提算法较其他2种算法在处理每一张样本图片的时间上分别由0.013s、0.011s缩短到0.008s,在计数准确度上分别由77.32%、84.94%提高到97.36%以上,并且在拼接精度上也优于其他2种算法,实验结果验证了所提算法的有效性。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	韩硕
	陈晓荣
	张彩霞
	郭蓉蓉
	王晓龙

关键词 ：计量学, 物料计数, 图像拼接, 模板匹配, 双模板搜索, 位置约束

Abstract：In order to solve the defects of the material counting method in the production process, an improved template matching algorithm is studied for material quantity detection, and the improved template matching algorithm (TMA) is optimized based on the dual template search method and the position constraint coefficient r. The process of finding the best matching image, obtaining the optimal solution, and then performing image stitching to complete the material count. Through experimental comparison with traditional TMA and Gaussian pyramid transformation (GPT) optimization, it proves that compared with the other two algorithms, the time to process each sample image is shortened from 0.013s, 0.011s to 0.008s, and the counting accuracy is improved from 77.32%, 84.94% to more than 97.36% respectively. The stitching accuracy is also better than the other two algorithms, and the experimental results verify the effectiveness of the algorithm.

Key words： metrology material counting image mosaic template matching dual template search position constraints

收稿日期: 2021-03-15 发布日期: 2022-07-18

PACS:	TB96
	TB938.1

基金资助:国家自然科学基金(52175513)

作者简介: 韩硕(1996-),女,新疆吐鲁番鄯善人,上海理工大学硕士研究生,主要研究方向为图像处理与模式识别。Email:sure_han@163.com

引用本文:

韩硕,陈晓荣,张彩霞,郭蓉蓉,王晓龙. 基于改进模板匹配算法的物料计数方法研究[J]. 计量学报, 2022, 43(7): 863-868.
HAN Shuo,CHEN Xiao-rong,ZHANG Cai-xia,GUO Rong-rong,WANG Xiao-long. Research on Material Counting Method Based on Improved Template Matching Algorithm. Acta Metrologica Sinica, 2022, 43(7): 863-868.

链接本文:

http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2022.07.05 或 http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2022/V43/I7/863

［1］姚莹, 蔡锦达, 刘倩. 基于机器视觉的平板式数粒机检测方法［J］. 光学仪器, 2018, 40 (4): 9-14.
Yao Y, Cai J D, Liu Q. Detection Method of Tablet Counting Machine Based on Machine Vision ［J］. Optical Instruments, 2018, 40 (4): 9-14.
［2］Lowe D. Distinctive Image Features from Scale-Invariant Keypoints ［J］. International Journal of Computer Vision, 2004, 60 (2): 91-110.
［3］Brown M, Lowe D. Automatic Panoramic Image Stitching Using Invariant Features ［J］. International Journal of Computer Vision, 2007, 74: 59-73.
［4］张见双, 张红民, 罗永涛, 等. 一种改进的Harris角点检测的图像配准方法［J］. 激光与红外, 2017, 47 (2): 230-233.
Zhang J S, Zhang H M, Luo Y T, et al. An Improved Harris Corner Detection Image Registration Method ［J］. Laser and Infrared, 2017, 47 (2): 230-233.
［5］Hu J W. Fast Image Matching Algorithm Based on Pixel Gray Value［C］//FMSMT 2017. Paris, 2017.
［6］Gong M G, Wu Y, Cai Q, et al. Discrete Particle Swarm Optimization for High-order Graph Matching ［J］. Information Sciences, 2016, 328: 158-171.
［7］Zhang X Q, Wang T, Wang J X, et al. Pyramid Channelbased Feature Attention Network for Image De-hazing ［J］. Computer Vision and Image Understanding, 2020: 197-198.
［8］Hamid H, Ahmed F, Ahmad J. Robust Image Hashing Scheme Using Laplacian Pyramids ［J］. Computers & Electrical Engineering, 2020, 84: 106648.
［9］Lyu W, Zhou Z, Chen L, et al. A Survey on Image and Video Stitching ［J］. Virtual Reality & Intelligent Hardware, 2019, 1 (1): 55-83.
［10］Gonzalez R C, Woods R E. 数字图像处理 (英文版)［M］. 北京: 电子工业出版社, 2010.
［11］Alessandro B, Massimo B, Massimiliano G, et al. Exploiting Multi-Level Parallelism for Stitching Very Large Microscopy Images ［J］. Frontiers in Neuroinformatics, 2019, 13: 235-246.
［12］Wang L Y, Shen Y, Liu H D, et al. An Accurate and Efficient Multi-category Edge Detection Method ［J］. Cognitive Systems Research, 2019, 58: 160-172.
［13］邢雪亮, 甘文波, 蒋朝根. 基于机器视觉的航空铆钉尺寸检测技术［J］. 计量学报, 2020, 41 (5): 518-523.
Xing X L, Gan W B, Jiang C G. Aeronautical Rivet Size Detection Technology Based on Machine Vision ［J］. Acta Metrologica Sinica, 2020, 41 (5): 518-523.
［14］Steger C, 等. 机器视觉算法与应用(双语版)［M］. 杨少荣, 等译. 北京: 清华大学出版社, 2019.
［15］Wang Y L, Liu W, Li F, et al. A Fast Template Matching Method Based on Improved Ring Projection Transformation and Local Dynamic Time Warping ［J］. Optik, 2020, 216: 164954.
［16］Lai J X, Lei L, Deng K Y, et al. Fast and Robust Template Matching with Majority Neighbour Similarity and Annulus Projection Transformation ［J］. Pattern Recognition, 2020, 98: 107029.
［17］Xiang T Z, Xia G S, Bai X, et al. Image Stitching by Line-guided Local Warping with Global Similarity Constraint ［J］. Pattern Recognition, 2018, 83: 481-497.
［18］Wang Z B, Cui Z J, Zhu Y. Multi-modal Medical Image Fusion by Laplacian Pyramid and Adaptive Sparse Representation ［J］. Comput Biol Med, 2020, 123: 103823.
［19］Shen D H, Zareapoor M, Yang J. Multimodal Image Fusion Based on Point-wise Mutual Information ［J］. Image and Vision Computing, 2020,105: 104047.
［20］包能胜, 方海涛. 连续运动螺纹尺寸自适应机器视觉检测［J］. 计量学报, 2020, 41 (9): 1062-1069.
Bao N S, Fang H T. Continuous Motion Thread Size Detection with Adaptive Machine Vision ［J］. Acta Metrologica Sinica, 2020, 41 (9): 1062-1069.
［21］杨璐. 一种基于人体轮廓特征的教室人数统计算法［J］. 计量学报, 2021, 42 (2): 178-183.
Yang L. An Algorithm for Counting People in Classroom Based on Human Contour Features ［J］. Acta Metrologica Sinica, 2021, 42 (2): 178-183.
［22］薛震,于莲芝,胡婵娟. 基于图像序列的运动目标检测识别关键技术研究［J］. 计量学报, 2020, 41(12): 1475-1481.
Xue Z, Yu L Z, Hu C J. Research on Key Techniques of Moving Target Detection and Recognition Based on Image Sequence［J］. Acta Metrologica Sinica, 2020, 41(12): 1475-1481.
［23］王一丁. 工业图像处理［M］. 西安: 西安电子科技大学出版社, 2018.
［24］万国挺, 王俊平, 李锦, 等. 图像拼接质量评价方法［J］. 通信学报, 2013, 34 (8): 76-81.
Wan G T, Wang J P, Li J, et al. Image Mosaic Quality Evaluation Method ［J］. Journal of Communications, 2013, 34 (8): 76-81.