Image recognizing method and device for ruler reading

Abstract

The invention provides an image recognizing method and device for ruler reading. The method comprises the following steps of: acquiring an image of measuring a to-be-measured object by using a ruler;recognizing the length of a ruler image in the image of the to-be-measured object along the ruler measurement direction by using a ruler image detecting model and intercepting the ruler image from theimage of measuring the to-be-measured object; recognizing the length of a ruler scale image in the intercepted ruler image along the ruler measurement direction by using a ruler scale image detectingmodel; and calculating the reading of measuring the to-be-measured object by the ruler according to the length of the ruler image along the ruler measurement direction, the length of the ruler scaleimage along the ruler measurement direction, the actual total length of the ruler and the actual length of the ruler scale in the ruler. The measurement reading of the ruler is calculated through therecognized lengths of the ruler and the ruler scale, so that the adaptability of the scale image recognizing to complex environments can be increased.

Description

technical field [0001] The invention relates to the technical field of image recognition, in particular to an image recognition method and equipment for scale reading. Background technique [0002] With the development of the information age, the image data is getting bigger and bigger, and it is growing at a very fast speed. Through these images, we can discover people's thoughts and thoughts, and provide a basis for further decision-making. Before data mining, it is necessary to effectively classify and manage these huge image data. In the classification and management of image data, image recognition is a basic work, which has important significance and value in information extraction, information recognition and information retrieval. It is very difficult for a computer to perform image classification or image recognition on image data like a human being. These difficulties lie in the irregularity of image data, the representation methods of different image data, and t...

AI Technical Summary

Problems solved by technology

This algorithm uses the traditional image processing method, which only performs well on a small number of images, and cannot well adapt to various water gauge models (for example, water gauge color, scale, size, etc.) and complex environments (for example, lighting, angle, etc.)

The existing scale positioning algorithm based on the color of the scale scale uses the color segmentation method to extract the color channel features in the image and performs threshold segmentation to locate the scale, but this algorithm cannot be well adapted to complex environments, for example, when the scale is above The algorithm is not robust in the presence of smudges, spots, localized damage, or scale colors

Images