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

A Hybrid Tone Mapping Approach for Machine Vision

A tone mapping, machine vision technology, applied in instrumentation, image data processing, computing, etc.

Active Publication Date: 2017-11-03
深圳市三宝创新机器人有限公司
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

There is no universal tone mapping method in the world that can satisfy all applications

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
  • A Hybrid Tone Mapping Approach for Machine Vision
  • A Hybrid Tone Mapping Approach for Machine Vision
  • A Hybrid Tone Mapping Approach for Machine Vision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0034] The technical solution of this patent will be further described in detail below in conjunction with specific embodiments.

[0035] see Figure 1-5 , a hybrid tone mapping method that can be used in machine vision, the specific steps are as follows:

[0036] (1) Perform tone mapping on the input HDR image according to the tone mapping empirical model to obtain the mapped image LDR 1 , the tone mapping empirical model includes a gradient domain tone mapping method.

[0037] (2) Use the visual saliency map calculation model to calculate the visual saliency map S of the HDR image HDR , the visual saliency map calculation model includes an image saliency map calculation method;

[0038] (3) The visual saliency map S of the HDR image HDR Perform logarithmic transformation to obtain the visual saliency map S HDR1 :

[0039] S HDR1 =ln(S HDR ) Formula 1)

[0040] (4) The visual saliency map S in formula (1) HDR1 Quantify to get the visual saliency map S HDR2 , the ca...

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 hybrid tone mapping method that can be used in machine vision. The specific steps are as follows: perform tone mapping on the input HDR image according to the tone mapping empirical model to obtain the mapped image LDR1; use the visual saliency map calculation model to calculate the HDR The visual saliency map SHDR of the image; logarithmically transform the visual saliency map SHDR of the HDR image to obtain the visual saliency map SHDR1: quantify the visual saliency map SHDR1 to obtain the visual saliency map SHDR2; transform the visual saliency map SHDR2 from the logarithmic domain Back to the luminance domain, the improved visual saliency map S'HDR is obtained; finally, the improved visual saliency map S'HDR and the image LDR1 obtained after tone mapping are multiplied and synthesized to obtain the final rendering result image LDR. The invention can switch between the empirical tone mapping method and the perceptual tone mapping method through user adjustment parameters, simulates the visual attention mechanism of human eyes to natural scenes, and can meet the requirements of different application occasions.

Description

technical field [0001] The invention relates to the field of digital image processing, in particular to a hybrid tone mapping method applicable to machine vision. Background technique [0002] The research goal of machine vision is to make it have the same environmental perception ability as the human eye. For the autonomous mobile robot system working in typical unstructured scenes, how to make the robot better understand the working environment in which it is able to autonomously Preconditions for running. [0003] High Dynamic Range (HighDynamicRange, HDR) image is a powerful tool that can record the brightness values ​​of real natural scenes. Currently, advanced digital cameras or video capture devices can be used to obtain HDR image content, but ordinary display devices still cannot directly process and output HDR image data. [0004] In order to solve the problem of the mismatch between the real scene and the dynamic range of the display device, many foreign scholars...

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): G06T5/50
Inventor 庄永军梁磊文康益兰兵华徐东群
Owner 深圳市三宝创新机器人有限公司