Research Of Automatic Recognition And Remote Management System For Dial Instruments

The dial instrument, as an important tool to measure the signal, is widely used in industrial production and scientific research. It has the advantages of simple structure, convenient maintenance, low cost, anti-EMI, dustproof and waterproof, and so on. However, with the development of technology, there are some problems when using dial instruments for measuring, verification or calibration, such as low efficiency, easy fatigability, and hard to manage. And this has been unable to meet the actual production and work requirements.Therefore, the domestic and foreign research institutions have launched the research on the recognition of dial instrument. In this paper, an automatic recognition and remote management system for dial instruments is disigned and realised, based on machine vision technology. The recognition terminal acquires the image of dial instrument in real-time by its camera. Then after recognition, the recognized information will be transmitted to the remote server for monitoring and management. The core algorithm consists of a pointer extraction algorithm based on region growing method and a pointer recognition algorithm based on hit-miss transform method. Besides, the seeds needed by region growing method can automatically obtained by fuzzy clustering based on difference image.The paper first puts forward the region growing method of image segmentation algorithm can be applied to the field of dial instruments recognition, and so does the morphological hit-miss transform method applied to the pointer recognition algorithm. These methods enrich the application method of dial instruments recognition, and achieved good recognition effect. Experiments show that, the algorithms can adapt well to many types of dial instruments for recognition. The algorithms is verified on Matlab platform, then transplanted to OpenCV platform, and finally worked on android devices for extensive use.