A method and device for extracting vehicle surface graphics and text based on laser ranging

Abstract

Embodiments of the present invention provide a method and device for extracting image and text on a vehicle surface based on laser ranging. The method includes: determining a lattice distance on the vehicle surface according to distance information from a laser pulse start position to each laser scanning point on the vehicle surface matrix; according to the intensity information of the echo pulses of each laser scanning point on the vehicle surface, determine the lattice intensity image of the vehicle surface; according to the preset gradient interval, divide the lattice distance matrix by row or column gradient to determine the point Multiple distance gradients of the matrix distance matrix; according to the multiple distance gradients, the rows or columns corresponding to the lattice intensity image and the lattice distance matrix are binarized with multiple thresholds under the same distance gradient to determine the vehicle surface image and text. Compared with conventional OCR recognition, the present invention has the advantages of long detection distance and large format, can break through the limitation of conventional short-range image and text recognition, and can realize not only the overall three-dimensional modeling of the target vehicle, but also long-distance large-size patterns and images. Extraction of text.

Description

technical field [0001] The embodiments of the present invention relate to the field of intelligent transportation, and in particular to a method and device for extracting images and texts on a vehicle surface based on laser ranging. Background technique [0002] Optical Character Recognition (Optical Character Recognition; OCR for short) is the current mainstream technology for image and text recognition. It mainly acquires images on the medium through scanning equipment or imaging equipment, and then extracts images and texts through image processing technology to identify semantic patterns. Mainly used in license plate recognition, scanned text recognition, electronic paper books, etc. [0003] Conventional OCR recognition, the recognition object is roughly A4 paper size graphic carrier. Recognition objects have limitations in terms of pixels and image frames. In addition, the distance between the recognition device and the recognition object is required to be short, whi...

AI Technical Summary

Problems solved by technology

[0006] In order to solve the problems of short recognition distance and low background resolution existing in the existing vehicle surface image and text recognition methods, the present invention provides a vehicle surface image and text extraction method and device

Images