基于OpenCV的车站信号平面图图像识别  

Image identification of station signal plan based on OpenCV

作　　者：李立 杜文文 李鹏 李逸 LI Li;DU Wenwen;LI Peng;LI Yi(CRRC Urban Transportation Planning and Design Research Institute Co.Ltd.,Nanjing 210031,China)

机构地区：[1]中车城市交通规划设计研究院有限公司,南京210031

出　　处：《铁路计算机应用》2024年第7期7-13,共7页Railway Computer Application

摘　　要：为实现异构车站信号平面图图纸数据格式的统一,结合车站信号平面图(简称:平面图)中各模块的形态特征分析与逻辑关系,提出基于OpenCV的图像识别方法以对平面图中主要的站场信号设备进行图像识别。将CAD平面图统一转化为相同分辨率及尺寸的PNG格式图像,采用灰度化和边缘检测算法进行图像预处理,得到站场的轮廓;分析平面图结构,利用Hough圆检测与直线检测实现平面图元的特征提取;根据识别到的圆与直线段,结合每种站场信号设备的属性特点及其之间物理连接的逻辑关系,分别设计识别方法。试验结果表明:基于OpenCV的图像识别方法可较为准确、完整地识别出各类站场信号设备,并获得每个设备相关的站场数据。To implement the standardization of data format for heterogeneous station signal plan drawings,this paper proposed an image identification method based on OpenCV by analyzing the morphological characteristics and logical relationships of each module in the station signal plan(referred to as the plan)to identify the main station signal equipment in the plan.The paper converted CAD floor plans into PNG format images of the same resolution and size,and used grayscale and edge detection algorithms for image preprocessing to obtain the contour of the station,analyzed the structure of the floor plan and use Hough circle detection and line detection to extract features of the floor plan elements,and based on the identified circles and straight line segments,combined with the attribute characteristics of each station yard signal equipment and the logical relationship of the physical connections between attribute characteristics to design identification methods separately.The experimental results show that the OpenCV based image identification method can accurately and completely identify various types of station yard signal equipment,and obtain station yard data related to each equipment.

关 键 词：车站信号平面图 OPENCV 图像识别 特征提取 Hough检测 

分 类 号：U284.7[交通运输工程—交通信息工程及控制] TP39[交通运输工程—道路与铁道工程]