Online wheel identification character recognition method and online wheel identification character detection system

Abstract

The invention provides an online wheel identification character recognition method and an online wheel identification character detection system. The detection system consists of a communication module, a control module, an image acquiring and processing module, a roller bed, a high-definition camera, a capturing camera, a camera stand, an LED (light emitting diode) lamp strip, a studio and a circuit. The online wheel identification character recognition method comprises the following steps: extracting surface images of a wheel by virtue of a high-definition SLR (Single Lens Reflex) camera; extracting identification character rings; extracting distribution shape parameters of the identification character rings on the ring by adopting a binarization algorithm; amplifying a strip-shaped identification character image, carrying out smoothing algorithm, Laplace sharpening, smoothing algorithm, binarization, image denoising, image spreading and image inclination correction processing; then sending to an OCR (Optical Character Recognition) system for recognition; and finally obtaining codes of the identification characters, and carrying out data transmission of codes of the identification characters with a production line management system, thus realizing wheel tracking management.

Description

technical field [0001] The invention belongs to the technical field of detection, and in particular relates to an online automatic recognition method and a detection system for identification characters on the surface of a wheel. Background technique [0002] After the wheels are produced, they have to go through multiple inspections. Before testing, it is necessary to mark the logo, that is, the hollow word code. After the logo is marked, the wheels go through several inspection processes on the production line. Wheels with various logos are easily mixed together. Manual identification is inefficient and error-prone. [0003] At present, the resolution of the highest-definition SLR camera is about 20 million pixels. Due to the large shape of the wheel, the recognition rate of characters with a resolution of 20 million pixels is not enough, so simply using this SLR camera cannot meet the recognition requirements. There are two ways to improve the resolution, image stitching...

AI Technical Summary

Problems solved by technology

Before the inspection, it is necessary to mark the logo, that is, the hollow word code. After the logo is marked, the wheels go through several inspection processes on the production line. The wheels with various logos are easily mixed together. The manual method of identification is inefficient and error-prone.

[0003] At present, the resolution of the highest-definition SLR camera is about 20 million pixels. Due to the large shape of the wheel, the recognition rate of characters with a resolution of 20 million pixels is not enough, so simply using this SLR camera cannot meet the recognition requirements.

Images