欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
研究基于图像识别的CTCS-3级列控系统车载设备人机界面信息的识别方法。首先,根据人机界面信息具有位置固定、字体和大小确定、信息量有限等特点,将人机界面信息划分为图标、数字和字母、汉字3种类型。根据各类信息的特点,分别进行灰度、二值化、切分和归一化等预处理,其中采用改进的基于最大宽度回溯的字切分法提高切分单个汉字的准确度。然后,对图标采用改进的6主色法提取颜色特征,采用统计法提取面积特征;对数字和字母提取欧拉数和8等分面积特征;对汉字提取欧拉数、细化的面积特征和笔画复杂性指数。最后,针对3种类型信息,分别构造决策树对提取的特征进行分类,实现车载设备人机界面信息的识别。以图标和汉字为例的实验结果表明,本文的方法能够准确地实现DMI界面信息的识别。
英文摘要：
The image recognition based method for recognizing Driver Machine Interface （DMI） information of the onboard equipment in CTCS-3 was studied. Firstly,based on the fact that the DMI information was characterized by fixed position,predefined font style and font size,as well as limited information volume,the DMI information was classified into 3 types,namely icons,digit numbers and English letters,as well as Chinese characters. According to the characteristics of different information types,the following pre-processing steps,including gray processing,binary processing,segmentation and normalization were carried out,in which the improved Chinese character segmentation algorithm based on maximum width backtracking was adopted to enhance the accuracy for single Chinese character segmentation. Secondly,for icons,the color feature was extracted by the improved 6 dominant color method and the area feature was extracted by statistics method. For digital numbers and English letters,the Euler numbers and the 8 equal division area features were extracted. For Chinese characters,the Euler number,the detailed area feature,and the Chinese stroke complexity index were extracted. Finally,the decision trees corresponding to different information types were constructed respectively to classify the above mentioned features,thus realizing the recognition of DMI information. Taking the recognition of icons and Chinese characters as example,the results show that the method proposed in this paper can accurately recognize the DMI information of the on-board equipment.
同期刊论文项目
同项目期刊论文
Robust single-track train dispatching model under a dynamic and stochastic environment: a scenario-b
Admissibility in the General Growth Curve Model With Respect to Restricted Parameter Sets Under Matr
Generalized Ridge and Principal Correlation Estimator of the Regression Parameters and Its Optimalit
Admissible Estimators in the General Multivariate Linear Model with Respect to Inequality Restricted
位置：