With the rapid development of science and technology, image processing is applied to all aspects of industry and daily life. Number recognition technology in seven-segment numeric instrument, a branch of image processing, can be used in many areas. With this technology, we can collect numeric instrument data automatically and rapidly.This paper is divided into two parts. In the first part, history of the number recognition technology and algorithms commonly used are reviewed. Number recognition procedure is mainly composed of three parts:image pre-processing, numeric character positioning and digital character recognition. In this paper, multi-thresholds method and improved sewing method are proposed. The decimal point is also detected. Finally, the number in instrument can be recognized effectively.In the second part, implementation of number recognition system in two different platforms is introduced. In windows platform, an image which contains numeric instrument is input and the numbers in it is output. In Android platform, we achieve taking a photo by a mobile device and detecting the numbers. The results show that both can work properly. |