Unlock instant, AI-driven research and patent intelligence for your innovation.

An image preprocessing method to improve the robustness of machine vision under natural lighting

An image preprocessing and natural lighting technology, which is applied in image data processing, image enhancement, image analysis, etc., can solve the problems of processing difficulty such as thresholding and large amount of calculation, and achieve small amount of calculation, high real-time performance, and eliminate influence Effect

Active Publication Date: 2021-09-07
SHANGHAI JIAOTONG UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The above color spaces can indeed reduce the influence of lighting conditions to a certain extent, but they are still three-color channels, and the three-channel calculation in the image processing algorithm is still relatively large, and it brings certain difficulties to thresholding and other processing.

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
  • An image preprocessing method to improve the robustness of machine vision under natural lighting
  • An image preprocessing method to improve the robustness of machine vision under natural lighting
  • An image preprocessing method to improve the robustness of machine vision under natural lighting

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The present invention will be described in detail below in conjunction with the accompanying drawings and embodiments.

[0036] figure 1 It is the flow chart of the image preprocessing method of the machine vision robustness that the present invention promotes under natural illumination, and concrete steps are as follows:

[0037] Step (1): Build the camera imaging model. First, hypothesis 1 is put forward: the point in the natural light scene is illuminated by a black body radiator, that is, it is assumed that the power spectral density of natural light obeys Planck's law.

[0038]

[0039] Here h, k, and c are Planck's constant, Boltzmann's constant, and the speed of light, respectively. T is the light temperature. λ is the wavelength of light. Furthermore, for an RGB three-color imaging camera, the R, G, and B color channel values ​​of an illuminated point x can be expressed as follows:

[0040]

[0041] where F x Indicates the vector product of the direc...

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 an image preprocessing method for improving the robustness of machine vision under natural light, including constructing a camera imaging model; performing sharpening processing on the original image, and converting the sharpened image to a logarithmic domain, thereby separating camera characteristics factor, the surface characteristic factor of the illuminated object and the illumination intensity and natural spectral characteristic factor; convert the three-channel color in the logarithmic domain to the N color space of a single channel, and the conversion factor is α; obtain the conversion according to the peak value of the spectral characteristic of the camera sensor Factor alpha. The invention converts the RGB three-channel image into a single-channel color space under the condition of reducing the influence of illumination conditions in the natural environment, eliminates the influence of illumination characteristics on the image, forms an illumination invariant graph, and has a smaller calculation amount and better real-time performance. High, and provides convenience for subsequent image processing algorithms such as thresholding.

Description

technical field [0001] The invention belongs to the field of machine vision, and in particular relates to an image preprocessing method for improving the robustness of machine vision under natural light. Background technique [0002] In recent years, machine vision has developed rapidly and has been widely used in robotics, unmanned driving and other fields. However, in the natural environment, the application of machine vision is limited to a certain extent. The main reason is that the natural environment is changeable. Different, and then, the image processing algorithm based on RGB three-color principle is less robust. [0003] Since the development of machine vision, there are a series of image processing algorithms aimed at the influence of lighting conditions. Including methods based on statistics, methods based on image decomposition, etc. These algorithms have high complexity and poor real-time performance, and are rarely used in practical applications. The most p...

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 Patents(China)
IPC IPC(8): G06T5/00G06T7/90
CPCG06T7/90G06T2207/10024G06T5/92
Inventor 刘成良蔡道清李彦明
Owner SHANGHAI JIAOTONG UNIV