Research On Intelligent Identification System Of Railway Container Number And Vehicle Type

Container is an important transportation unit for freight transportation,and it is a standardized container for supporting the multimodal logistics system nodes such as ports,transfer stations,loading stations,etc.In the process of container transportation,the efficiency and process of cargo handling operations have become one of the important bottlenecks to improve the timeliness of cargo transportation.At present,most of the data entry of container loading and unloading operations still need to be manually completed by operators,which not only restricts the loading and unloading efficiency of transportation process,but also increases the cost of operation.It also affects the accuracy of data information due to the transcript's error input.Therefore,how to apply modern intelligent equipment and information technology to solve the problem of loading and unloading has aroused widespread concern.Under this background,the general process,method and principle of image processing,character recognition and vehicle type recognition are compared and analyzed in this paper.This paper summarizes the general characteristics and laws of image of containers,container numbers and railway vehicles.According to the characteristics,this paper improved the location and segmentation method of container number characters in three stages,and a railway vehicle identification method based on vehicle projection peak characteristics is also designed.This paper takes the demand for software development of an intelligent handling system in a loading and unloading equipment company as an example.On the research content of the container number identification,the location and segmentation method is improved according to the length width ratio of container character area,and the accuracy of character recognition is improved by using the three segment segmentation and training process of font store.On the research content of railway vehicle type recognition,a vertical projection method based on vehicle profile is designed to identify vehicle types according to the characteristics of overlooking images of railway vehicles,which can effectively distinguish railway flat or gondola.On the basis of the research results,this paper realizes the development of a software system that can automatically identify container numbers and identify railway vehicle types.According to the characteristics and rules of the actual acquisition image,this paper adopts the appropriate image processing technology and methods to preprocess,locate,segment and match the pictures.In order to ensure the reliability of database entry information,the system increases the function of self-checking and manual correction of recognition results.Finally,a relatively complete intelligent recognition software system for container numbers and railway vehicle types is developed,which is capable of image acquisition,data storage and human-machine interaction.After actual testing,the accuracy and recognition speed of container number recognition and railway vehicle identification function are able to meet the application requirements,and the reliability of the method is verified.