Research on automatic indication values of pointer gauges based on computer vision
Xin Shen, Min Cao, Yong Hong Lu, Linshan Zhang · 2016
This paper mainly discusses the application of computer visual technology in embedded image recognition system. We propose and realize a new system called automatic recognition system of analog display instruments using computer vision. After the brief introduction of the characteristics of computer visual technology and embedded system, this paper analyzes the principles and the developing status of computer visual technology used in embedded system, proposes the automatic recognition system of analog display instruments using computer vision, which is on the basis of the application characteristics. Its working principles are just like the principles of the visual system of human being, “analyzing” the images “viewed” by the system, and obtaining the information needed. This paper designs the software and hardware system through focusing on the digital image processing technology and the application circumstances of the embedded system. The following key problems have been studied: how to realize automatic recognition of analog display instruments, how to solve automatic judgment of different kinds of indicator dial, how to eliminate disturbance cased by light and circumstances while exact result, how to realize the image collecting with high quality while low cost, how to improve the efficiency of data collection and image recognition (speed and accuracy) and how to correspond the each module design in hardware and software to improve the system's performance and stabilization. At the end, this paper designs automatic recognition program of analog display instruments.