Road Target Detection Method Based on 3D Model
A technology for 3D model and target detection, applied in biological neural network models, character and pattern recognition, instruments, etc., can solve problems such as low robustness, occlusion and shadows, and achieve the effect of improving robustness and reducing costs
Active Publication Date: 2022-04-22
HANGZHOU DIANZI UNIV
View PDF4 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] In order to overcome the problems of occlusion and shadows and low robustness in the two-dimensional detection process existing in the prior art, the present invention provides a method that can improve the detection robustness and solve the problems that occur in the two-dimensional detection process. 3D model-based road object detection method for occlusion and shadow problems
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
[0049] The present invention will be further described below in conjunction with accompanying drawing and specific embodiment:
[0050] Such as figure 1 The illustrated embodiment is a three-dimensional model-based road target detection method, comprising the following steps:
[0051] Step 100, acquire two road sampling images synchronously, calculate the disparity images of the two road sampling images, and obtain the depth features of the disparity images
[0052] Step 101, acquire two road sampling images synchronously, establish a coordinate system with the optical center of the left camera as the coordinate origin, and obtain the relationship between the visual image point p and the binocular vision measurement system:
[0053]
[0054]
[0055]
[0056] Among them, A 1 , A 2 Respectively represent the internal parameters of the left camera and the right camera,
[0057] R is the rotation matrix,
[0058] T is the translation matrix, (u 1 , v 1 ), (u ...
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 road target detection method based on a three-dimensional model, comprising the following steps: synchronously acquiring two road sampling images, calculating the parallax images of the two road sampling images, and obtaining the depth features of the parallax images; according to the depth features of the parallax images Generate candidate areas, and map the coordinate points of the 3D bounding box of the candidate areas to the 2D image; use the convolutional neural network of the multi-scale pooling network layer to extract the shape features of the 2D image, and use the 3D model to obtain the road target. The invention has the following beneficial effects: the algorithm of the invention solves the problems of occlusion and shadow in the two-dimensional detection process, improves the robustness and reduces the cost.
Description
technical field [0001] The invention relates to the technical field of machine vision, in particular to a three-dimensional model-based road target detection method capable of improving detection robustness and solving occlusion and shadow problems in the two-dimensional detection process. Background technique [0002] A stable and reliable vehicle detection process is the first step in traffic analysis. Related vehicle counting, vehicle tracking, vehicle classification, assisted driving, accident detection and road behavior judgment are all based on accurate vehicle detection. In the process of traffic detection, the traffic information that traditional detectors can obtain is relatively simple, and it usually needs to combine multiple sensors to assist in completing a detection task. However, due to the characteristics of multi-source and heterogeneous detection data, the integration and fusion of various traffic detection data has become a bottleneck. This makes the vis...
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): G06V20/52G06V10/44G06V10/82G06N3/04
CPCG06V20/588G06V10/44G06N3/045
Inventor 陈婧许文强彭伟民
Owner HANGZHOU DIANZI UNIV