Global point cloud description method based on point cloud projection contour signature and distribution matrix

A distribution matrix and cloud projection technology, applied in the field of pattern recognition, can solve problems such as mixed noise and achieve the effect of strong expressive ability

Inactive Publication Date: 2018-07-06
深圳慎始科技有限公司
View PDF5 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, currently commonly used point cloud acquisition sensors such as kinect and depth cameras based on time-of-flight technology will be affected by various error sources, and the depth images obtained by using these devices will inevitably be mixed with various noises after being converted into point cloud data.

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
  • Global point cloud description method based on point cloud projection contour signature and distribution matrix
  • Global point cloud description method based on point cloud projection contour signature and distribution matrix
  • Global point cloud description method based on point cloud projection contour signature and distribution matrix

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0109] Embodiment 1 Evaluation of the recognition accuracy of the invention descriptor on the public data set

[0110] The global point cloud description method based on the point cloud projection contour signature and distribution matrix provided by the present invention includes the following steps in order:

[0111] 1) Preprocessing stage

[0112] 1-1) Map the depth image into 3D point cloud data. In this embodiment, the Washington RGB-D ObjectDataset (http: / / rgbd-dataset.cs.washington.edu / ) is used as the test data set, which provides the depth of different viewing angles corresponding to 300 common household daily necessities in 51 categories image set.

[0113] Such as figure 1 As shown, taking the plate_1_4_236_crop.png in the data set as an example, it shows the schematic process of generating the corresponding CSDM descriptor. First, use the interface function depthToCloud provided by the data set to convert the depth image into 3D point cloud data. For any point ...

Embodiment 2

[0147] Embodiment 2 Invention Descriptor Robustness Evaluation to Noise

[0148] In order to evaluate the robustness of the global point cloud descriptor proposed by the present invention to noise, 10 groups of noise test experiments were designed:

[0149] Add different levels of Gaussian noise to each point in the target point cloud in the data set in its three coordinate directions, and the standard deviations of the corresponding noise are 1, 2, 3, 4, 5, 6, 7, 8, 9 and 10, the unit is mm. The corresponding result is as image 3 shown.

Embodiment 3

[0150] Example 3 Robustness Evaluation of Invented Descriptors to Point Cloud Density Changes

[0151] In order to evaluate the robustness of the global descriptor proposed by the present invention to point cloud density changes, five sets of test experiments were designed: using the VoxelGrid filter provided in the point cloud library PCL (http: / / pointclouds.org / documentation / tutorials / voxel_grid.php) performs different degrees of downsampling on the target point cloud. The VoxelGrid filter achieves different degrees of downsampling by controlling the size of the Voxel. Five sets of design experiments use Voxel sizes of 1mm, 5mm, 10mm, 15mm and 20mm. The corresponding experimental results are as Figure 4 shown.

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 relates to a point cloud description method based on projection contour signature and a distribution matrix. Firstly a local reference coordinate system having rotation and translation invariance is constructed by using the spatial coordinate information of the target point cloud, and then the target point cloud is transferred to the reference coordinate system. A space boundary boxof the target point cloud after attitude normalization is constructed, and the target point cloud is respectively projected to the three adjacent planes of the boundary box. The projection contour signature and the distribution matrix of projection of the point cloud on each projection plane are calculated for enhancing the expression ability of the descriptor for the visible part of the target point cloud, and the related series rules are determined according to the statistical features of point cloud projection. The sub-features of the target point cloud on the three projection planes are connected in series in the form of a histogram so as to obtain the final CSDM descriptor.

Description

technical field [0001] The invention belongs to the field of pattern recognition, and in particular relates to a global point cloud description method based on point cloud projection contour signatures and distribution matrices. Background technique [0002] Object recognition is one of the most challenging tasks in robotics. In order to ensure the physical interaction between the robot and the surrounding environment, it is necessary to use the relevant object representation method to provide the robot with relevant information about the interactive objects in the surrounding environment in real time. [0003] Although there have been many object recognition methods designed based on object 2D information and 3D information, it is still a challenging task to recognize 3D objects in the presence of noise and varying point cloud resolution. Compared with 2D information, 3D information such as point cloud contains more target space information, which will be beneficial to ach...

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): G06K9/46G06K9/62G06K9/00
CPCG06V20/64G06V10/507G06V10/44G06F18/253
Inventor 付明亮冷雨泉韩小宁任利学占志鹏马维斯
Owner 深圳慎始科技有限公司
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