基于稀疏取样和梯度分布特征的车标识别

doi:10.16182/j.issn1004731x.joss.201709021

系统仿真学报 ›› 2017, Vol. 29 ›› Issue (9): 2035-2042.doi: 10.16182/j.issn1004731x.joss.201709021

基于稀疏取样和梯度分布特征的车标识别

周斌斌^1,2, 高尚兵^1*, 潘志庚^1,2, 王亮亮¹, 王洪阳¹

1.淮阴工学院计算机与软件工程学院,淮安 223001;
2.杭州师范大学数字媒体与人机交互研究中心,杭州 311121

收稿日期:2017-04-29 发布日期:2020-06-02
作者简介:周斌斌(1994-),男,江苏南通,硕士生,研究方向为图像处理。
基金资助:
国家自然科学基金(61402192, 61332017),国家重点研发计划(2015BAK04B05),江苏省六大人才高峰资助项目(XYDXXJS-011),江苏省333工程资助项目(BRA2016454),江苏省青蓝工程资助

Vehicle Logo Recognition Based on Sparse Sampling and Gradient Distribution Features

Zhou Binbin^1,2, Gao Shangbing^1*, Pan Zhigeng^1,2, Wang Liangliang¹, Wang Hongyang¹

1. College of Computer and Software Engineering, Huaiyin Institute of Technology, Huaian 223001, China;
2. Virtual Reality and Human-Computer Interaction Research Center, Hangzhou Normal University, Hangzhou 311121, China

Received:2017-04-29 Published:2020-06-02

摘要/Abstract

摘要： 传统方法将车标定位与识别分开进行,定位的误差将会给后续的识别带来影响,并且车标图像具有低分辨率、低质量的特点。提出了一种新颖的车标定位和识别有机整合的方法。通过稀疏取样对样本图像进行取点采样,将点集分为邻近点集和非邻近点集,分别对其提取梯度特征和明暗特征,构建特征库,对车标粗定位区域进行多尺度扫描。实验结果表明,该方法在车标检测识别效率方面具有更大的优势,而且对于不同类型的车标图像都具有鲁棒性。

关键词: 图像处理, 灰度分布, 梯度分布, 多尺度检测, 车标定位, 车标识别

Abstract: The vehicle logo location and recognition are separated in the traditional method, the location errors will affect the subsequent recognition, at the same time the vehicle logo images are with low resolution and poor quality. Thus, a novel method was proposed which integrated the vehicle logo location and recognition organically. The sample images were sampled by sparse sampling, and then the point set was divided into adjacent point set and non adjacent point set, and the gradient feature and light and dark feature were extracted respectively, constructing the feature library. The logo coarse location area was multi-scale scanned. The experimental results show that the proposed method is superior to other advanced algorithms on the vehicle detection and recognition efficiency, and robust to the different types of logo images.

Key words: image processing, gray-scale distribution, gradient distribution, multi-scale detection, vehicle logo location, vehicle-logo recognition

中图分类号:

TP391.4

周斌斌, 高尚兵, 潘志庚, 王亮亮, 王洪阳. 基于稀疏取样和梯度分布特征的车标识别[J]. 系统仿真学报, 2017, 29(9): 2035-2042.

Zhou Binbin, Gao Shangbing, Pan Zhigeng, Wang Liangliang, Wang Hongyang. Vehicle Logo Recognition Based on Sparse Sampling and Gradient Distribution Features[J]. Journal of System Simulation, 2017, 29(9): 2035-2042.

参考文献

[1] 黄敏, 钮中铭, 张小兰, 等. 基于GIS-T的交通仿真路网构建研究[J]. 系统仿真学报, 2017, 29(3): 531-536.
(Huang Min, Niu Zhongming, Zhang Xiaolan, et al.Research of the Construction of GIS-T Traffic Simulation Road Network[J]. Journal of System Simulation (S1004-731X), 2017, 29(3): 531-536.)
[2] 魏平顺. 智能交通系统中车标图像识别技术研究 [D]. 南京: 南京理工大学, 2013.
(Wei Pingshun.Research on the recognition technology of vehicle logos in intelligent transportation system [D]. Nanjing, China: Nanjing University of Technology, 2013.)
[3] 李侠, 李文举, 孙娟红, 等. 基于模板匹配和边缘检测的车标定位方法[J]. 计算机工程与设计, 2011, 32(10): 3453-3456.
(Li Xia, Li Wenju, Sun Juanhong, et al.Approach for vehicle-logo location based on template matching and edge detection[J]. Computer Engineering and Design, 2011, 32(10): 3453-3456.)
[4] 周阳. 基于图像的车标定位与识别 [D]. 北京: 北方工业大学, 2015.
(Zhou Yang.Image based vehicle logo location and recognition [D]. Beijing, China: North China University of Technology, 2015.)
[5] 李熙莹, 吕硕, 袁敏贤. 基于散热器栅格背景精确分类的车标定位方法[J]. 计算机工程与应用, 2017, 53(2): 206-213.
(Li Xiying, Lv Shuo, Yuan Minxian.Vehicle logo location method based on accurate radiator grid back-ground classification[J]. Computer Engineering and Applications, 2017, 53(2): 206-213.)
[6] 朱资淘. 基于SIFT算子的车标识别软件 [D]. 广州: 华南理工大学, 2013.
(Zhu Zitao.Vehicle Logo recognition software based on SIFT operator [D]. Guangzhou, China: South China University of Technology, 2013.)
[7] 陈舜杰. 快速车标识别算法的研究 [D]. 上海: 上海交通大学, 2013.
(Chen Shunjie.Research on the rapid recognition algorithm of the vehicle logo [D]. Shanghai, China: Shanghai Jiaotong University, 2013.)
[8] 王建. 基于图像的智能交通系统中车标识别技术研究 [D]. 南京: 南京理工大学, 2014.
(Wang Jian.The research of vehicle logo recognition technology in intelligent transportation system [D]. Nanjing, China: Nanjing University of Technology, 2014.)
[9] 王永杰, 裴明涛, 贾云得. 多信息融合的快速车牌定位[J]. 中国图象图形学报, 2014, 19(3): 471-475.
(Wang Yongjie, Pei Mingtao, Jia Yunde.License plate detection based on multiple features[J]. Journal of Image and Graphics, 2014, 19(3): 471-475.)
[10] 周爱珍. 基于稀疏采样的医学成像方法研究 [D]. 广州: 南方医科大学, 2011.
(Zhou Aizhen.Research on medical imaging method based on sparse sampling [D]. Guangzhou, China: Southern Medical University, 2011.)
[11] Hongyin Shi, Baojing Jia.SAR imaging method based on coprime sampling and nested sparse sampling[J]. Journal of Systems Engineering and Electronics (S1004-4132), 2015, 6: 1222-1228.

基于稀疏取样和梯度分布特征的车标识别

Vehicle Logo Recognition Based on Sparse Sampling and Gradient Distribution Features

PDF (PC)

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 4

编辑推荐

Metrics

本文评价

[1]	袁彬淦, 钟铭恩, 倪晶鑫. 线束端子压接后外观缺陷的视觉检测算法研究[J]. 系统仿真学报, 2022, 34(5): 1152-1159.
[2]	王乐, 罗宇, 王海宽, 费敏锐. ToF深度相机测量误差校正模型[J]. 系统仿真学报, 2017, 29(10): 2323-2329.
[3]	李新利, 李楠, 孙愉佳, 卢钢, 闫勇, 刘石. 火焰自由基成像和极限学习机在NOx排放预测中的研究[J]. 系统仿真学报, 2016, 28(5): 1179-1185.
[4]	周建新, 吴建军, 薛均强, 林帅, 党岗, 程志全. 适用于密集人群目标检测的多尺度检测方法[J]. 系统仿真学报, 2016, 28(10): 2503-2509.