Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Image processing and character recognition algorithm based on near-infrared imaging

A technology of character recognition and near-infrared light, which is applied in the field of image processing and computer vision, can solve the problems of high noise, low contrast of near-infrared light imaging, and low accuracy of character recognition in near-infrared light images, achieving high recognition accuracy, Strong adjustability and enhanced effects

Inactive Publication Date: 2020-01-07
HARBIN INST OF TECH
View PDF3 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide an image processing and character recognition algorithm based on near-infrared light imaging, so as to solve the problems of low contrast and more noise in the existing near-infrared light imaging, and the low accuracy of character recognition in near-infrared light images. high problem

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Image processing and character recognition algorithm based on near-infrared imaging
  • Image processing and character recognition algorithm based on near-infrared imaging
  • Image processing and character recognition algorithm based on near-infrared imaging

Examples

Experimental program
Comparison scheme
Effect test

specific Embodiment approach 1

[0026] Specific embodiment one: an image processing and character recognition algorithm based on near-infrared light imaging, comprising the following steps:

[0027] Step 1. Perform filtering and smoothing processing on the acquired near-infrared license plate image to achieve a denoising effect;

[0028] Step 2, utilize the OpenCV function library to realize that the near-infrared light license plate image after filtering and de-drying processing is preprocessed; said preprocessing includes binarization and edge enhancement (the purpose of this step is to extract characters in the near-infrared light license plate image Effective information, so that the subsequent character recognition, this step mainly utilizes the OpenCV function library, which greatly facilitates the realization of image processing);

[0029] Step 3, using Matlab programming to implement a template matching algorithm, and matching the extracted characters of the preprocessed license plate image with the ...

specific Embodiment approach 2

[0031] Specific embodiment two: this embodiment is a further description of specific embodiment one. In step 1, median filtering is used to filter and smooth the near-infrared light image (median filtering belongs to nonlinear filtering, and the Noise has an obvious suppression effect. The basic idea of ​​median filtering is to replace the gray value at the pixel with the median value of the gray value of the neighborhood of the pixel. Simply put, it is to select a matrix window, and the center of the neighborhood window The pixel value at the point is the median value of all pixel gray values ​​in the window), and the filtering formula of median filtering is:

[0032]

[0033] Among them: f(x, y) is the gray value of the selected target pixel (arrange all the gray values ​​of the domain window, and the selected intermediate value G is the gray value of the filtered pixel), and the surrounding area of ​​the target pixel is After the gray value of the pixels in the nine-squa...

specific Embodiment approach 3

[0034] Specific embodiment three: this embodiment is a further description of specific embodiment one. In step 2, the image after the binarization process is filtered and smoothed (binarization is to convert the image into one containing only 0 and 1) Pixel value image, the purpose is to separate the target character from the irrelevant background for subsequent character segmentation and recognition), the calculation formula of the binarization threshold is:

[0035] W(x,y)=0.5×(maxf(x+x 1 ,y+y 1 )+min(x+x 1 ,y+y 1 )) (2)

[0036] Among them: f(x, y) is the gray value of the selected target pixel, and binarization converts the gray image obtained from the near-infrared light image into a binarized image that is easy for computer processing; W(x, y) is the calculation Threshold in the area, x, y are the coordinates of any pixel in the image, x 1 、y 1 To get the coordinates of all pixels in the window around the pixel point (x, y);

[0037] The purpose of the edge enhanc...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a character recognition algorithm based on a near-infrared imaging technology. The invention belongs to the field of computer vision and the field of image processing, and solves the technical problems that a near-infrared image obtained under a near-infrared band is not high in contrast ratio and is seriously influenced by noise, and obtained license plate characters are fuzzy and are not easy to identify. According to the technical scheme of the invention, the method mainly comprises the steps: obtaining a near-infrared license plate image through a high-resolution silicon-based CMOS chip; carrying out smooth filtering processing on the near-infrared image by adopting an OpenCV function library; performing character recognition on the filtered license plate imagethrough a template matching algorithm; a CNN convolutional neural network and a deep learning algorithm are used to train a near-infrared image character library to carry out more accurate character recognition. Compared with infrared thermal imaging, the near-infrared imaging can obtain more image details and information, and has a wide application prospect in the fields of license plate recognition, night reconnaissance and the like.

Description

technical field [0001] The invention belongs to the fields of computer vision and image processing, and in particular relates to an image processing and character recognition algorithm based on near-infrared light imaging. Background technique [0002] Near-infrared imaging technology can directly obtain the shape and image details of the detected object, and identify and distinguish the key information of the object. At the same time, in the case of insufficient lighting conditions, near-infrared imaging can still obtain high-resolution images, so it has a very wide application background in the fields of security detection, pattern detection and recognition. In addition, in recent years, with the rapid development of science and technology and the rapid improvement of urban traffic, the concept of Intelligent Traffic System (Intelligent Traffic System, ITS) emerged as the times require, and has been widely used in the field of traffic management. As an indispensable part ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/34G06K9/36G06N3/04G06N3/08
CPCG06N3/08G06V10/20G06V30/153G06V20/625G06V30/10G06N3/045
Inventor 孙芳魁李效欣丁卫强
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com