A Hough Transform-based Recognition Method for High-precision Mechanical Instrument Expression Numbers

Abstract

The invention discloses a high-precision mechanical instrument representation number recognition method based on Hough transform, which belongs to the field of image processing, and specifically includes the following steps; image preprocessing is performed on an image: the image is binarized into a grayscale image, and then the image is processed into a grayscale image. The image is filtered to remove noise points, and then edge detection is performed to remove unnecessary information while retaining the image edge information; the instrument center is detected by Hough transform; the line detection technology of Hough transform is used to detect the edge feature line of the pointer, The apex of the pointer is obtained by calculating the intersection of the line that detects the edge of the pointer, so as to obtain an accurate pointer ray in combination with the center of the meter; according to the vertex of the pointer and the angle of the line with the center of the meter as the apex, the specific number. The invention improves the accuracy of recognizing high-precision mechanical instrument numbers, and improves the stability of recognition.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to a high-precision mechanical instrument representation number recognition method based on Hough transform. Background technique [0002] Due to its simple structure, convenient maintenance, high precision, and strong anti-electromagnetic resistance, pointer-type instruments are widely used in the medical field. However, the pointer type instrument generally does not have the function of data line interface. The acquisition of the degree is mainly based on the reading of the naked eye, which is easily affected by many human factors. At the same time, if the scale of the instrument is not fine enough, the accuracy of the naked eye identification is very low. At the same time, the medical field needs to make an overall analysis of the changes before and after certain values, observe the changes, etc., and the small changes are difficult to be distinguished by the naked eye...

AI Technical Summary

Problems solved by technology

In the current research, there have been quite a lot of literature on the identification of meter representations. Most of the applicable ones are semi-circular voltmeters in the industry, and there are few full-circle meters, and because the pointers used by some meters Larger, different from the thinner pointer used by electrical instruments, so it cannot be determined by a simple and simplified method of detecting a single line. At the same time, due to the reality of shooting and other reasons, it is impossible to shoot from a completely frontal angle It is a meter, so the method of simply detecting a straight line will lose some accuracy in a fine meter

[0004] Accurate readings of blood pressure in medical treatment can be read with numerical blood pressure instruments, but due to the limitations of the numerical instrument itself, it is difficult to continuously display its changing curve, so it is necessary to display the blood pressure changing curve in a pointer-type way. On the contrary, its relative specific reading is not very important, what is important is the accuracy of its relative change curve

Images