View two-dimensional attitude template construction method, view two-dimensional attitude template identification method and positioning and grabbing system

A construction method and a two-dimensional model technology, applied in the field of image processing, can solve problems such as perspective distortion, difficulty in changing the relative relationship between cameras and objects, and inability to solve perspective distortion

Active Publication Date: 2020-04-28
SHENZHEN HUAHAN WEIYE TECH
View PDF8 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

According to template feature matching, there are still some disadvantages: it is difficult to find features in the image with certainty due to changes in the relative relationship between the camera and the object target, resulting in features being occluded and perspective distortion of features, and the template matching method cannot solve this perspective distortion
[0007] When using the view to recognize a 3D object, there is a problem in projecting a 3D target model from different viewpoints: many 2D projections are required to cover the range of viewpoint changes of the target object
To deal with this problem, pose clustering techniques have been applied, however, the number of 2D projections that must be compared with 2D images is still very large, so these view-based recognition cannot meet the real-time requirements
Often, calculations are accelerated by reducing the number of views generated, but the camera is always pointed at the center of the three-dimensional target, which easily leads to perspective distortion, so that the target object that is not in the center of the image cannot be found. Based on view recognition technology, another An open issue is the application of 2D projections to 2D template matching in order to produce 2D templates

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
  • View two-dimensional attitude template construction method, view two-dimensional attitude template identification method and positioning and grabbing system
  • View two-dimensional attitude template construction method, view two-dimensional attitude template identification method and positioning and grabbing system
  • View two-dimensional attitude template construction method, view two-dimensional attitude template identification method and positioning and grabbing system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0044] Please refer to figure 1 , this application discloses a visual Figure II A method for constructing a three-dimensional posture template, which includes steps S110-S140, which will be described separately below.

[0045] Step S110, acquiring template information of a 3D model corresponding to the target object.

[0046] It should be noted that the target objects here can be products on the industrial assembly line, mechanical parts in the object box, tools on the operating table, etc., for example Figure 5 Irregularly shaped mechanical parts as shown in . Then, the three-dimensional graphic information of the target objects can be obtained by using a contact or non-contact scanning device such as a three-dimensional drawing tool, and the graphic information can be part of the appearance shape data of the target object surface.

[0047] For example, the 3D model corresponding to the target object can be expressed by a CAD model or a similar 3D description, for exampl...

Embodiment 2

[0137] Please refer to Figure 12 , in a disclosed view of the embodiment Figure II On the basis of the method for constructing a three-dimensional pose template, this embodiment discloses a method for recognizing a three-dimensional pose of a target object, which includes steps S210-S230, which will be described separately below.

[0138] Step S210, acquiring a two-dimensional image of the target object.

[0139] It should be noted that the target objects here can be products on the industrial assembly line, mechanical parts in the object box, tools on the operating table, etc., for example Figure 5 Irregularly shaped mechanical parts as shown in . Then, the digital image of the target object can be collected by the camera device arranged around the target object, and then the two-dimensional image of the target object can be obtained according to the digital image.

[0140] Step S220, combining the two-dimensional image with the view constructed in step S140 Figure II d...

Embodiment 3

[0188] Please refer to Figure 15 , on the basis of the three-dimensional pose recognition method of the target object disclosed in the second embodiment, this embodiment discloses a positioning and grasping system for the target object, which mainly includes a camera 31, a processor 32, a controller 33 and a motion mechanism 34, respectively explained below.

[0189] The imaging device 31 is arranged around the target object and is used to collect digital images of the target object. The target object here can be a product on an industrial assembly line, a mechanical part in a case, a tool on an operating table, etc., for example Figure 5 Irregularly shaped mechanical parts as shown in . Then, the camera device 31 may be a digital video camera or a digital camera, which has the function of real-time collection of digital images. It can be understood that the camera device 31 can be a specific application of the target camera device or virtual camera mentioned in the first ...

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 view two-dimensional attitude template construction method, a view two-dimensional attitude template identification method and a positioning and grabbing system. The identification method comprises: acquiring a two-dimensional image of a target object, matching the two-dimensional image with a view two-dimensional attitude template obtained by a construction method, searching from the view two-dimensional attitude template to obtain two-dimensional attitude data with the highest matching degree, and determining a three-dimensional pose of the target object according to the two-dimensional attitude data with the highest matching degree. The two-dimensional models distributed in the adjacent levels in the view two-dimensional attitude template are associated with each other through the parent-child view relationship of the corresponding view model, so when the view two-dimensional attitude template is used for matching the two-dimensional image, matching from the high level to the low level of the pyramid one by one is facilitated, and the efficiency of two-dimensional image matching is improved while the two-dimensional attitude data with the highest matching degree is obtained through rapid searching.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to video Figure II The construction method and recognition method of the three-dimensional posture template, and the positioning and grasping system. Background technique [0002] The invention mainly solves the problems of visual recognition of three-dimensional objects and determination of three-dimensional poses in image processing. Target recognition is the first problem that needs to be solved in the vision-guided problem. In many fields, it is assumed that the target is planar and the target in the image is limited to a certain extent, such as affine transformation or similar transformation. However, because the target moves in 3D space in front of a fixed camera, the camera moves around a fixed target, or the target and camera move simultaneously, with 6 degrees of freedom (3 rotations and 3 translations), in many applications the recognized target is not planar, b...

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 & AuthorityApplications(China)
IPC IPC(8): G06K9/00G06T3/00G06T7/73
CPCG06T7/73G06V20/647G06T3/067
Inventor杨洋
OwnerSHENZHEN HUAHAN WEIYE TECH