Road Target Depth Estimation Method Based on Ground Clue Propagation
A technology of target depth and clues, applied in computing, image analysis, image enhancement, etc., can solve problems such as unsatisfactory depth estimation results, inconsistency of actual depth information, and insufficient fineness of the global depth relationship, etc., to achieve reliable scene depth estimation.
Active Publication Date: 2021-09-28
HEFEI UNIV OF TECH
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the global depth relationship obtained by using the layout structure and semantic information is not fine enough, and often does not match the actual depth information.
[0005] Aiming at the problem of unsatisfactory depth estimation results above, this paper proposes a road target depth estimation based on ground clue propagation.
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0149] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. The present invention is a road target depth estimation based on ground clue propagation, and the specific process is as follows figure 1 As shown, the implementation scheme of the present invention is divided into the following steps:
[0150] Step S1: For the input road image, extract over-segmented superpixels and combine them to generate the target area. The specific operation steps include:
[0151] Step S1-1: For the input image, use the watershed algorithm to obtain the over-segmented superpixel set SP, denoted as SP={sp i},i=1,...,n sp , where sp i is the superpixel numbered i, n sp is the number of superpixels in the road image;
[0152]Step S1-2: Calculate the average RGB color of all superpixels, and extract the average RGB color of all superpixels C={c i},i=1,...,n sp ;
[0153] Step S1-3: Mark all superpixels as untraversed...
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
Login to View More Abstract
The invention discloses a method for estimating the depth of a road target based on ground clue propagation. By acquiring the scene depth distribution of vertical objects in contact with the ground and using the edge occlusion orientation relationship between regions, the scene target depth reasoning and estimation are realized. First, obtain the over-segmented superpixels of the road image, traverse the over-segmented superpixels in turn, and merge regions according to the RGB average color to generate the target region; secondly, extract the vanishing points and image surface markers of the road image, and use Gaussian probability distribution is used to estimate the depth of the segmented area; then, according to the shared boundary of the segmented area, an edge occlusion orientation detector is constructed to extract the edge occlusion relationship between the segmented areas; finally, the ground clues are used to propagate until the scene depth relationship is stable, and the target depth is output. Obtain a scene object depth estimate.
Description
technical field [0001] The invention belongs to the technical field of scene depth estimation, and more specifically relates to a road target depth estimation method based on ground clue propagation. Background technique [0002] Depth in the vision system is a measure of the vertical distance between the observed object and the imaging plane, and is an important 2.5-dimensional information representation that converts the two-dimensional plane captured by ordinary sensor equipment into a three-dimensional structure description. In computer vision, depth estimation is helpful to guide humans in 3D realistic simulation of scenes, and it also has broad application prospects in road monitoring equipment, robot navigation and other fields. Due to the complexity of real road scenes, there are multiple challenges in recovering the depth information of road objects from a single image. [0003] Monocular image depth information estimation mainly relies on some monocular depth cues...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/11G06T7/13G06T7/187G06T7/50G06T7/90
CPCG06T2207/10024G06T2207/20081G06T2207/30256G06T7/11G06T7/13G06T7/187G06T7/50G06T7/90
Inventor 张鹿鸣谢昭吴克伟韩娜高扬童赟
Owner HEFEI UNIV OF TECH