Monocular infrared image depth estimating method on basis of SVM (Support Vector Machine) model

An infrared image and depth estimation technology, applied in image analysis, image data processing, computing and other directions, can solve the problem of inaccurate categories without learning, and achieve the effect of improving the accuracy.

Inactive Publication Date: 2012-10-03
DONGHUA UNIV
View PDF2 Cites 23 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Although this method is suitable for a series of pictures with simple scenes and achieves the expected effect, it is often inaccurate for categories that have not been learned in the scene.

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
  • Monocular infrared image depth estimating method on basis of SVM (Support Vector Machine) model
  • Monocular infrared image depth estimating method on basis of SVM (Support Vector Machine) model
  • Monocular infrared image depth estimating method on basis of SVM (Support Vector Machine) model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] In order to make the present invention more comprehensible, a preferred embodiment is described in detail below with accompanying drawings.

[0029] Such as figure 1 As shown, this embodiment discloses a method for estimating the depth of a monocular infrared image based on the SVM model, the steps of which are:

[0030] Step 1. Obtain any monocular infrared image I(x, y) and the depth map corresponding to the monocular infrared image I(x, y);

[0031]Step 2. Set feature areas on three different scales for each pixel in the monocular infrared image I(x, y). The three different scales are from small to large: the first scale, the second scale and The third scale, wherein the feature region of each scale includes an image block at the center and image blocks adjacent to the image block above, below, left, and right, and the i-th pixel point at the center on the first scale The image block is the i-th pixel itself, the image block at the center of the i-th pixel on the s...

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 method for carrying out depth estimation on an infrared image. A support vector regression theory is utilized to establish a depth estimating model for depth characteristic extracted from the infrared image so as to utilize the depth model to estimate a new depth map of the infrared image. The method comprises the following steps of: firstly, selecting a characteristic vector in the infrared image; searching a characteristic vector related to the infrared depth characteristic by utilizing a stepwise linear regression method, carrying out independent component analysis on the screened characteristic vectors to obtain independent characteristic vectors and establishing an infrared depth training set; then on the basis of a nonlinear support vector regression theory with a kernel function, calculating the training set to obtain the depth estimating model; and finally, after carrying out extraction on an infrared image which is newly introduced according to a characteristic extraction method in the training set, carrying out depth estimation on the infrared image which is newly introduced so as to obtain the depth map of the infrared image. An experimental result proves that the depth map of the infrared image can be effectively estimated by the method.

Description

technical field [0001] The invention relates to a method capable of estimating the depth of a monocular infrared image, through which the spatial position information of the scene in the infrared image can be estimated. Background technique [0002] Image depth estimation is to obtain depth and distance information from images, which is essentially a problem of depth perception. The spatial position information constructed by depth perception represents the relative distance from the observer to the detected surfaces in the scene. There is an ideal algorithm for restoring the depth and distance information in color images, but for infrared images, because it reflects the temperature distribution of the scene and has defects such as low signal-to-noise ratio and low contrast, the depth algorithm for restoring the image Still blank. If the depth information of the infrared image can be restored, it will greatly improve the human eye's understanding of the image. [0003] Th...

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): G06T7/00
Inventor 席林孙韶媛
Owner DONGHUA UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap