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

Binocular disparity map enhancement method based on confidence fusion

A technology of binocular disparity and disparity map, which is applied in the field of stereo matching, can solve the problems that the accuracy of disparity value in low-texture areas cannot be improved, and high-quality disparity maps cannot be generated, so as to achieve the effect of improving accuracy

Active Publication Date: 2020-04-24
NORTHWESTERN POLYTECHNICAL UNIV
View PDF11 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a binocular disparity map enhancement method based on confidence fusion to solve the problem that the prior art cannot improve the accuracy of disparity values ​​in low-texture regions and discontinuous depth regions, and cannot generate high-quality disparity maps question

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
  • Binocular disparity map enhancement method based on confidence fusion
  • Binocular disparity map enhancement method based on confidence fusion
  • Binocular disparity map enhancement method based on confidence fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0108] In order to verify the effectiveness of the method in this paper, we select the Tsukuba map and compare it with the binocular disparity map enhancement method based on confidence fusion through the traditional method SGBM of the present invention.

[0109] First, image correction is carried out according to the method in the above-mentioned step S2 for the input Fig. 2 (a) left Tsukuba figure and Fig. 2 (b) right Tsukuba figure; then the improved convolutional neural network method described in step S3 Corrected left Tsukuba image and right Tsukuba image for disparity estimation: We use the KITTI Stereo 2012 dataset as the training set, use the Tsukuba image as the test set, calculate the convolutional layer, update the weight by optimizing the cost function, and finally obtain the disparity map I c , as shown in Figure 2(d). Then we use the left Tsukuba map and right Tsukuba map corrected by the method of step S2 to obtain the optical flow field through the LK optical...

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 binocular disparity map enhancement method based on confidence fusion, and the method comprises the following steps: S1, placing a picture right ahead the center of a binocular camera, photographing a target through the binocular camera, and obtaining a left image a and a right image b of a binocular map; S2, performing stereogram distortion correction on the left image aand the right image b to obtain a left image a1 and a right image b1 after stereogram correction; S3, performing parallax estimation on the left image a1 and the right image b1 by adopting a convolutional neural network algorithm to obtain a dense parallax image a2; performing disparity map optimization processing on the dense disparity map a2 to obtain an optimized disparity map Ic; S4, performing parallax estimation on the corrected left image a1 and right image b1 of the stereogram by adopting an improved optical flow estimation algorithm to obtain a parallax image IO; and S5, according tothe confidence level, fusing the obtained disparity map IO and disparity map Ic to obtain a final disparity map I (u).

Description

【Technical field】 [0001] The invention belongs to the technical field of stereo matching in image processing, and in particular relates to a binocular disparity map enhancement method based on confidence fusion. 【Background technique】 [0002] Binocular stereo vision is an important form of machine vision, which obtains the three-dimensional information of objects through the principle of parallax and combining two images. Stereo matching is the main technical means to obtain three-dimensional object information from two-dimensional image information. It refers to the projection points of any point on two or more cameras on the spatial scene. These points are called corresponding points. Corresponding points between the left and right image planes. The process of obtaining stereo corresponding points is called the process of disparity estimation, so disparity estimation is an important basis for depth estimation. [0003] Traditional disparity estimation methods generally ...

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): G06T5/50G06N3/04
CPCG06T5/50G06T2207/20221G06T2207/20084G06N3/045
Inventor 赵春晖苏梅梅刘慧霞胡劲文田雪涛
Owner NORTHWESTERN POLYTECHNICAL UNIV