Thermal Instrumentation, Machine Readable

Dial indicator is a kind of measuring meter which is widely applied to the current producing process because it is reliable, low-priced. Presently, the reading work of the dial indicator in our country still adopts the manual method. This method is restricted by such subjective factors as observing angel and distance as well as fatigue strength and so on, which results in not only large error and low reliability but also low efficiency and great labor costs. On this point, this design aims at developing an auto-reading system to solve some key problems in the development of the auto-reading technology of dial indicators. It is a valuable and useful study which is full of market prospects.The development of digital image processing technology as well as its application to industry gives the inspiration to us that replaces man's eyes with untiring computer's "eyes" to do the pointer reading work. This design mainly focuses on the auto reading of the pointer-meter, proposes to pick-up the pointer and scale number's information from the collected dial images by the image processing technology so as to judge the relative position between the pointer and the scale numbers. At the same time, I use the swatch images to train the Neural Network and utilize the trainerd Neural Network to recognize the scale numbers. The goal is to realize the auto-reading of pointer-meter with the help of the digital image processing technology.This paper takes pressure pointer-meter as an example and develops the digital image processing system. It is finally proved that this system is superior to the manual method in both working efficiency and testing precision. This method of auto reading can be widely applied to the auto-reading of dial indicators.