Design Of Character Recognition System Of Digital Dial Based On Machine Vision

Due to the fact that the traditional digital detection instruments are not equipped with the corresponding communication serial interface,the data measurement or calibration by manual or semi-automatic way always lead to low efficiency and error prone in the current measurement and verification process.In addition,there are certain risks in the individual environment,which is not suitable for long-term manual operation.Therefore,it is very important to realize the automation of detection system and measurement process.This thesis designs a character recognition system of digital dial based on machine vision,which has important practical significance for the realization of the automation of character recognition and measurement process of digital dial,as well as the production of intelligent and precise measurement.At the same time,it has certain theoretical significance and practical value for improving the level of measurement and verification technology in the field of military scientific research.In this thesis,the functional requirements of the character recognition system of digital dial are analyzed in detail.According to the functional requirements and design scheme,the corresponding demonstration is carried out,and the hardware design,software design and system detection interface design of the system are completed.The main contents include the following three aspects:Firstly,according to the functional requirements of the digital dial character recognition system,the overall scheme is designed,and the anti-light interference camera module and automatic shift module are demonstrated.Secondly,the hardware,software and detection interface design of the system are introduced.The hardware mainly includes the anti-light interference camera module and automatic shift adjustment module.The anti-light interference camera module adopts the design of shielding the external light,supplementing the light needed for camera shooting,and equipped with a camera suitable for the system to complete the image acquisition of the digital dial.And the automatic shift adjustment module uses the device designed and processed by the machine to fix the digital multimeter,controls the dial knob with the shift clamp,and rotates the fixed angle in a certain direction by sending the shift rotation instruction to the STM32F103ZET6 chip development board,so as to realize the rotation of the dial knob.For the system software,Matlab programming language is adopted on the host computer to complete the acquisition and location of the dial area image.Then,the image is processed by character tilt correction,character segmentation and character feature extraction to obtain the dial character features.After that,the dial character features are recognized by back propagation(BP)neural network algorithm and the recognition results are obtained.The system detection interface uses Lab VIEW language to develop a digital dial character recognition system software interface,and the Matlab program is packaged as a dynamic link library to call directly.Finally,on-line debugging is carried out after the completion of the system design and implementation.The test results show that the performance of the character recognition system based on machine vision designed in this thesis can meet the design requirements,and it can effectively complete the measurement and verification of digital instruments.