Method of character recognition for LKJ human-machine interface unit display
-
摘要: 人机界面(DMI)单元是列车运行监控装置(LKJ)的用户交互设备。通过对图像字符识别方法的研究,设计了基于点阵对比方式的字符识别方法,来对DMI图像中的字符做识别,识别的结果提供给系统校验程序使用。通过将界面中的字符点阵信息与字库点阵信息作对比,得到界面中所显示的字符信息。经过大量图像识别测试后得到识别正确率可达到98%,对整幅图像识别时间在1 s以内。结果表明,基于点阵识别的字符识别方法可以满足系统校验程序的使用要求。Abstract: Driver-machine interface (DMI) unit is user interface of the train operation monitoring device (LKJ). Through the research of image character recognition method, this article designed a character recognition method based on "Dot Matrix Contrast" to recognize the characters in the DMI unit image. The result of identification was provided to the system verification program. By comparing the character lattice information in the interface with that in the font library, the character information displayed in the interface could be obtained. After a large number of image recognition tests, the recognition accuracy can reach 98%, and the recognition time for the whole image is less than 1 second. The results show that the character recognition method based on "Dot Matrix Contrast" can meet the requirements of the system verification program.
-
Keywords:
- LKJ /
- driver-machine interface (DMI) /
- character recognition
-
-
[1] 李晓杰,钱雪军. LKJ2000型列车运行监控系统仿真研究[J]. 铁路计算机应用, 2015, 24(5):38-41. [2] 付强,袁磊. 基于聚类分析及SVM的DMI机车信号自动识别[J]. 铁路计算机应用, 2015, 24(8):46-49. [3] 瞿洋,杨利平. Hough变换OCR图象倾斜矫正方法[J]. 中国图象图形学报(A辑), 2001, 6(2):178-181. [4] Xu W, Zhang L. A geometric reasoning based algorithm for point pattern matching[J]. Science in China. Series F:Information Sciences, 2001, 44(6):445-452.
[5] 秦进,陈笑蓉,汪维家,等. 文本分类中的特征抽取[J]. 计算机应用, 2003, 23(2):45-46. [6] 高永英,张利,吴国威. 一种基于灰度期望值的图象二值化算法[J]. 中国图象图形学报, 2018, 4(6):524-528. [7] 匡纲要. 图像纹理特征提取方法综述[J]. 中国图象图形学报, 2009, 14(4):622-635. [8] 张云刚,张长水. 利用Hough变换和先验知识的车牌字符分割算法[J]. 计算机学报, 2004(1):130-135. [9] 焦慧,刘迁,贾惠波. 基于词平台汉字编码的自动标引研究[J]. 计算机工程与应用, 2007, 43(15):173-175. [10] 陕耀,张雁桥. 图像处理技术在铁路车号自动识别中应用的探讨[J]. 铁路计算机应用, 2006, 15(7):31-33. -
期刊类型引用(2)
1. 廖小春,华莹珂. 一种图像识别系统与远传技术的应用研究. 铜业工程. 2020(06): 83-85+100 . 百度学术
2. 王来兵. 基于卷积神经网络与GPU计算的名片文本识别算法研究. 长春工程学院学报(自然科学版). 2020(03): 100-104 . 百度学术
其他类型引用(1)
计量
- 文章访问数: 56
- HTML全文浏览量: 0
- PDF下载量: 15
- 被引次数: 3