CN113034664A - Automatic inspection method and device based on 3D vision - Google Patents
Automatic inspection method and device based on 3D vision Download PDFInfo
- Publication number
- CN113034664A CN113034664A CN201911342309.8A CN201911342309A CN113034664A CN 113034664 A CN113034664 A CN 113034664A CN 201911342309 A CN201911342309 A CN 201911342309A CN 113034664 A CN113034664 A CN 113034664A
- Authority
- CN
- China
- Prior art keywords
- images
- model
- product
- variance
- field
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Quality & Reliability (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention provides an automatic inspection method and device based on 3D vision, wherein the method comprises the steps of obtaining at least two 3D images of a product to be detected, wherein the shooting angle of the first 3D image and the shooting angle of the second 3D image form a preset angle; fitting two of the 3D images into a 3D model; comparing the 3D model with a pre-stored 3D standard model to obtain a 3D variance between the 3D model and the 3D standard model; and judging whether the product is qualified or not based on the 3D variance. By acquiring the 3D image of the product, the image precision is greatly improved, and the detection accuracy can be effectively improved for the product with a complex product structure.
Description
Technical Field
The invention relates to the technical field of 3D vision equipment, in particular to an automatic inspection method and device based on 3D vision.
Background
At present, in the field of automatic inspection of product defects, two-dimensional photographing is generally adopted, and then the photographed picture is compared with a qualified image stored in a database to judge whether the size of a product exceeds a preset range.
However, the inventor finds that the method is only suitable for the case that the product structure is simple, and for the product with a complex product structure, the error is large, and the using effect is not ideal.
Disclosure of Invention
According to the defects of the prior art, the invention provides an automatic inspection method and device based on 3D vision, which are suitable for automatic inspection of complex products.
The technical problem to be solved by the invention is realized by adopting the following technical scheme:
as a first aspect of the present invention, there is provided a 3D vision-based automated inspection method, comprising
Acquiring at least two 3D images of a product to be detected, wherein the shooting angle of the first 3D image and the shooting angle of the second 3D image form a preset angle;
fitting two of the 3D images into a 3D model;
comparing the 3D model with a pre-stored 3D standard model to obtain a 3D variance between the 3D model and the 3D standard model;
and judging whether the product is qualified or not based on the 3D variance.
As an alternative embodiment, before acquiring the at least two 3D images of the product to be detected, an auto-focusing step is further included.
As an alternative embodiment, the automatic focusing comprises
Pre-shooting two 3D images;
respectively acquiring the front depth of field and the rear depth of field in the two 3D images;
respectively acquiring the definition of the preset marks in the two 3D images;
and focusing is carried out through the analysis of the acquired definition of the mark and the mark position.
As an alternative embodiment, the focusing includes analyzing the acquired definition of the mark and the mark position
And adjusting the focal length until the marks all fall into the depth of field, wherein the front depth of field and the rear depth of field respectively contain at least one mark, and the sum of the distances between the marks and the focal point is minimum.
As an alternative embodiment, the fitting of the two 3D images into one 3D model includes filtering the two 3D images separately;
and fitting the two filtered 3D images into a 3D model.
As an alternative embodiment, the determining whether the product is qualified based on the 3D variance includes
Comparing the 3D variance with a preset threshold value, judging whether the 3D variance exceeds the threshold value, and if the 3D variance exceeds the threshold value, determining that the product size is unqualified; and if the 3D variance does not exceed the threshold value, the product size is qualified.
As a second aspect of the present invention, there is provided a 3D vision-based automated inspection apparatus, comprising
The acquisition module is used for acquiring at least two 3D images of a product to be detected, wherein the shooting angle of the first 3D image and the shooting angle of the second 3D image form a preset angle;
the fitting module is used for fitting the two 3D images into a 3D model;
the comparison module is used for comparing the 3D model with a pre-stored 3D standard model to obtain a 3D variance between the 3D model and the 3D standard model;
and the judging module is used for judging whether the product is qualified or not based on the 3D variance.
As an optional implementation manner, the device further comprises a focusing module, and the focusing module is used for focusing
Pre-shooting two 3D images;
respectively acquiring the front depth of field and the rear depth of field in the two 3D images;
respectively acquiring the definition of the preset marks in the two 3D images;
and focusing is carried out through the analysis of the acquired definition of the mark and the mark position.
As an optional implementation manner, the focusing module is further configured to adjust the focal length until the identifiers both fall into the depth of field, at least one of the identifiers is included in each of the front depth of field and the rear depth of field, and a sum of distances between the identifier and the focal point is minimum.
As an alternative embodiment, the fitting module is configured to filter the two 3D images respectively; and fitting the two filtered 3D images into a 3D model.
The invention has the beneficial effects that:
in the embodiment of the invention, at least two 3D images are obtained for a product to be detected, the at least two 3D images are fitted to obtain the 3D model so as to reduce errors generated by shooting, and then the fitted 3D model is compared with the pre-stored 3D standard model to obtain the 3D variance so as to judge whether the product is qualified. In the embodiment of the invention, the 3D image is acquired for the product, so that the image precision is greatly improved, and the detection accuracy rate can be effectively improved for the product with a complex product structure.
Drawings
The invention is further illustrated with reference to the following figures and examples.
FIG. 1 is a logic diagram of the method of this embodiment;
fig. 2 is a schematic view of the apparatus of this embodiment.
Detailed Description
The following embodiments are provided to describe the embodiments of the present invention, and to further describe the detailed description of the embodiments of the present invention, such as the shapes, configurations, mutual positions and connection relationships of the components, the functions and operation principles of the components, the manufacturing processes and operation methods, etc., so as to help those skilled in the art to more fully, accurately and deeply understand the inventive concept and technical solutions of the present invention.
To achieve the above object, as shown in fig. 1, there is provided an automated inspection method based on 3D vision, comprising
S100, obtaining at least two 3D images of a product to be detected, wherein the shooting angle of the first 3D image and the shooting angle of the second 3D image form a preset angle;
s200, fitting the two 3D images into a 3D model;
s300, comparing the 3D model with a pre-stored 3D standard model to obtain a 3D variance between the 3D model and the 3D standard model;
s400, judging whether the product is qualified or not based on the 3D variance.
In the embodiment of the invention, at least two 3D images are obtained for a product to be detected, the at least two 3D images are fitted to obtain the 3D model so as to reduce errors generated by shooting, and then the fitted 3D model is compared with the pre-stored 3D standard model to obtain the 3D variance so as to judge whether the product is qualified. In the embodiment of the invention, the 3D image is acquired for the product, so that the image precision is greatly improved, and the detection accuracy rate can be effectively improved for the product with a complex product structure.
As an optional implementation manner, before acquiring at least two 3D images of the product to be detected, an automatic focusing step is further included. In this way, the accuracy of the obtained 3D image is ensured.
Optionally, the automatic focusing comprises
Pre-shooting two 3D images;
respectively acquiring the front depth of field and the rear depth of field in the two 3D images;
respectively acquiring the definition of the preset marks in the two 3D images;
and focusing is carried out through the analysis of the acquired definition of the mark and the mark position.
So, to focus on 3D camera, it is simple and convenient.
As an alternative embodiment, the focusing includes analyzing the acquired definition of the mark and the mark position
And adjusting the focal length until the marks all fall into the depth of field, wherein the front depth of field and the rear depth of field respectively contain at least one mark, and the sum of the distances between the marks and the focal point is minimum.
As an alternative embodiment, the fitting of the two 3D images into one 3D model comprises
Filtering the two 3D images respectively;
and fitting the two filtered 3D images into a 3D model.
As an optional implementation mode, the judging whether the product is qualified or not based on the 3D variance comprises
Comparing the 3D variance with a preset threshold value, judging whether the 3D variance exceeds the threshold value, and if the 3D variance exceeds the threshold value, determining that the product size is unqualified; and if the 3D variance does not exceed the threshold value, the product size is qualified.
As shown in FIG. 2, corresponding to the method, there is provided a 3D vision-based automatic inspection apparatus, which comprises
The acquisition module 100 is configured to acquire at least two 3D images of a product to be detected, where a shooting angle of a first 3D image and a shooting angle of a second 3D image form a preset angle;
a fitting module 200 for fitting the two 3D images into a 3D model;
a comparison module 300, configured to compare the 3D model with a pre-stored 3D standard model, and obtain a 3D variance between the 3D model and the 3D standard model;
a decision module 400 for deciding whether the product is qualified based on the 3D variance.
In the embodiment of the invention, at least two 3D images are obtained for a product to be detected, the at least two 3D images are fitted to obtain the 3D model so as to reduce errors generated by shooting, and then the fitted 3D model is compared with the pre-stored 3D standard model to obtain the 3D variance so as to judge whether the product is qualified. In the embodiment of the invention, the 3D image is acquired for the product, so that the image precision is greatly improved, and the detection accuracy rate can be effectively improved for the product with a complex product structure.
As an optional implementation manner, the device further comprises a focusing module, and the focusing module is used for focusing
Pre-shooting two 3D images;
respectively acquiring the front depth of field and the rear depth of field in the two 3D images;
respectively acquiring the definition of the preset marks in the two 3D images;
and focusing is carried out through the analysis of the acquired definition of the mark and the mark position.
As an optional implementation manner, the focusing module is further configured to adjust the focal length until the identifiers both fall into the depth of field, at least one of the identifiers is included in each of the front depth of field and the rear depth of field, and a sum of distances between the identifier and the focal point is minimum.
As an alternative embodiment, the fitting module is configured to filter the two 3D images respectively; and fitting the two filtered 3D images into a 3D model.
The invention has been described in an illustrative manner, and it is to be understood that the invention is not limited to the precise form disclosed, and that various insubstantial modifications of the inventive concepts and solutions, or their direct application to other applications without such modifications, are intended to be covered by the scope of the invention. The protection scope of the present invention shall be subject to the protection scope defined by the claims.
Claims (10)
1. An automated inspection method based on 3D vision is characterized in that: comprises that
Acquiring at least two 3D images of a product to be detected, wherein the shooting angle of the first 3D image and the shooting angle of the second 3D image form a preset angle;
fitting two of the 3D images into a 3D model;
comparing the 3D model with a pre-stored 3D standard model to obtain a 3D variance between the 3D model and the 3D standard model;
and judging whether the product is qualified or not based on the 3D variance.
2. The automated inspection method based on 3D vision according to claim 1, characterized in that: before the at least two 3D images of the product to be detected are obtained, the method further comprises the step of automatic focusing.
3. The automated inspection method based on 3D vision according to claim 2, characterized in that: the automatic focusing comprises
Pre-shooting two 3D images;
respectively acquiring the front depth of field and the rear depth of field in the two 3D images;
respectively acquiring the definition of the preset marks in the two 3D images;
and focusing is carried out through the analysis of the acquired definition of the mark and the mark position.
4. The automated inspection method based on 3D vision according to claim 3, characterized in that: the focusing includes analyzing the acquired definition and position of the mark
And adjusting the focal length until the marks all fall into the depth of field, wherein the front depth of field and the rear depth of field respectively contain at least one mark, and the sum of the distances between the marks and the focal point is minimum.
5. The automated inspection method based on 3D vision according to claim 1, characterized in that: said fitting of two of said 3D images into one 3D model comprises
Filtering the two 3D images respectively;
and fitting the two filtered 3D images into a 3D model.
6. The automated inspection method based on 3D vision according to claim 1, characterized in that: the determining whether the product is qualified based on the 3D variance includes
Comparing the 3D variance with a preset threshold value, judging whether the 3D variance exceeds the threshold value, and if the 3D variance exceeds the threshold value, determining that the product size is unqualified; and if the 3D variance does not exceed the threshold value, the product size is qualified.
7. An automatic inspection device based on 3D vision, its characterized in that: comprises that
The acquisition module is used for acquiring at least two 3D images of a product to be detected, wherein the shooting angle of the first 3D image and the shooting angle of the second 3D image form a preset angle;
the fitting module is used for fitting the two 3D images into a 3D model;
the comparison module is used for comparing the 3D model with a pre-stored 3D standard model to obtain a 3D variance between the 3D model and the 3D standard model;
and the judging module is used for judging whether the product is qualified or not based on the 3D variance.
8. The 3D vision-based automated inspection device of claim 7, wherein: further comprises a focusing module for
Pre-shooting two 3D images;
respectively acquiring the front depth of field and the rear depth of field in the two 3D images;
respectively acquiring the definition of the preset marks in the two 3D images;
and focusing is carried out through the analysis of the acquired definition of the mark and the mark position.
9. The 3D vision-based automated inspection device of claim 7, wherein: the focusing module is further used for adjusting the focal length until the marks all fall into the depth of field, the front depth of field and the rear depth of field respectively contain at least one mark, and the sum of the distances between the marks and the focal point is minimum.
10. The 3D vision-based automated inspection device of claim 7, wherein: the fitting module is used for filtering the two 3D images respectively; and fitting the two filtered 3D images into a 3D model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911342309.8A CN113034664A (en) | 2019-12-23 | 2019-12-23 | Automatic inspection method and device based on 3D vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911342309.8A CN113034664A (en) | 2019-12-23 | 2019-12-23 | Automatic inspection method and device based on 3D vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113034664A true CN113034664A (en) | 2021-06-25 |
Family
ID=76451300
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911342309.8A Pending CN113034664A (en) | 2019-12-23 | 2019-12-23 | Automatic inspection method and device based on 3D vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113034664A (en) |
-
2019
- 2019-12-23 CN CN201911342309.8A patent/CN113034664A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109300161B (en) | Positioning method and device based on binocular vision | |
US10698308B2 (en) | Ranging method, automatic focusing method and device | |
CN109859272B (en) | Automatic focusing binocular camera calibration method and device | |
CN105657282B (en) | A kind of visual identity method that brightness of image actively optimizes | |
EP3416370A1 (en) | Photography focusing method, device, and apparatus for terminal | |
CN111932636B (en) | Calibration and image correction method and device for binocular camera, storage medium, terminal and intelligent equipment | |
CN106990646A (en) | Many lens systems, its method of work and portable electron device | |
CN106031148B (en) | Imaging device, method of auto-focusing in an imaging device and corresponding computer program | |
JP2014057303A (en) | System and method for utilizing enhanced scene detection in depth estimation procedure | |
TW201530493A (en) | Image capturing device and method for calibrating image deformation thereof | |
KR101566619B1 (en) | Apparatus and method for estimating distance using dual off-axis color filter aperture | |
CN110087049A (en) | Automatic focusing system, method and projector | |
CN109636857B (en) | Alignment method and calibration system | |
CN108156359A (en) | Intelligent industrial camera | |
KR101913705B1 (en) | Method and apparatus for measuring depth of materials attached to cylinder using line laser | |
CN116433780A (en) | Automatic calibration method for laser structured light based on machine vision | |
US8169532B2 (en) | Portable electronic device and method for measuring distance by performing auto focus function | |
US8547439B2 (en) | Testing an optical characteristic of a camera component | |
CN116563391B (en) | Automatic laser structure calibration method based on machine vision | |
KR101578029B1 (en) | Apparatus and method for correcting image distortion | |
CN106303174B (en) | Digital adjustment method for inclination of micro-lens array in light field camera | |
CN113034664A (en) | Automatic inspection method and device based on 3D vision | |
JP2011095131A (en) | Image processing method | |
CN114964032B (en) | Blind hole depth measurement method and device based on machine vision | |
CN106934786B (en) | Method for realizing image processing software of mold monitor |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210625 |
|
WD01 | Invention patent application deemed withdrawn after publication |