光学学报, 2014, 34 (s2): s210002, 网络出版: 2014-11-17  

彩色交通标志图像二维条形码抗干扰性能分析

Anti-Interference Analyzing of Two-Dimensional Barcode for Color Traffic Sign
王蒙军 1,2,*郝宁 1王霞 1
作者单位
1 河北工业大学信息工程学院, 天津 300401
2 天津市电子材料与器件重点实验室, 天津 300401
摘要
利用脉冲耦合神经网络(PCNN)模型融合点火像素点位置信息生成的二维条形码,实现镜像对称交通标志的分类识别。为了研究二维条形码的抗干扰性能,在待识别图像中加入椒盐噪声,并以不同分级数,采用R、G、B三个颜色空间进行处理,分析经过预处理前后形成的二维条形码抗干扰性能。针对国家标准道路交通标志图像库GB5768-1999中117个标志,其中45个黄色警告标志,43个红色禁令标志,29个蓝色指示标志进行实验,结果表明,二维条形码可以克服镜像对称标志无法区分的问题,并且抗干扰性能优于一维条形码,在R、G、B三个颜色空间分别处理时,1%的椒盐噪声影响下最高匹配率可达97%。
Abstract
Two-dimensional binary barcode are generated by pulse coupled neural network (PCNN) combined with position information of firing pixel can be used for mirror symmetry traffic sign images recognizing, in order to explore its character of anti-interference, salt and pepper noise is induced, anti-interference characters before and after pre-processing are tested at different division levels in RGB color space, respectively. Experiments are carried out based on 117 traffic signs images, including 45 yellow forbidden signs, 43 red warning signs, and 29 blue instruction signs in GB5768-1999. Experimental results show that two-dimensional binary barcode can distinguish mirror symmetry traffic signs and has better anti-interference characters than one-dimensional binary barcode. Highest matching rate is 97% when three channels in RGB color space are processed respectively under the influence of 1% salt and pepper noise.

王蒙军, 郝宁, 王霞. 彩色交通标志图像二维条形码抗干扰性能分析[J]. 光学学报, 2014, 34(s2): s210002. Wang Mengjun, Hao Ning, Wang Xia. Anti-Interference Analyzing of Two-Dimensional Barcode for Color Traffic Sign[J]. Acta Optica Sinica, 2014, 34(s2): s210002.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!