CN109446969A - A method of analysis and search threedimensional model - Google Patents
A method of analysis and search threedimensional model Download PDFInfo
- Publication number
- CN109446969A CN109446969A CN201811237227.2A CN201811237227A CN109446969A CN 109446969 A CN109446969 A CN 109446969A CN 201811237227 A CN201811237227 A CN 201811237227A CN 109446969 A CN109446969 A CN 109446969A
- Authority
- CN
- China
- Prior art keywords
- image
- search
- data
- analysis
- projected
- 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.)
- Withdrawn
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/64—Three-dimensional objects
- G06V20/647—Three-dimensional objects by matching two-dimensional images to three-dimensional objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/42—Global feature extraction by analysis of the whole pattern, e.g. using frequency domain transformations or autocorrelation
- G06V10/431—Frequency domain transformation; Autocorrelation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of methods of analysis and search threedimensional model, include the following steps: (1): analyzing data image respectively by global analysis and the local multiple data local features for obtaining multiple data global characteristics and multiple data images first;(2): obtaining search image;(3): and then by carrying out global analysis and partial analysis to search image respectively, obtain the local feature of search global characteristics and search search image;(4): based on search global characteristics, corresponding data global characteristics are obtained from data global characteristics, and corresponding data local feature is obtained from data local feature based on search local feature;(5): obtaining the image of the corresponding data from the local feature of image and corresponding data based on corresponding data global characteristics data.The present invention is used to solve the problems, such as to analyze due to using global characteristics input data and is compared with the model of storage in the database and causes search result inaccurate.
Description
Technical field
The present invention relates to technical field of image processing, more particularly to the method for analysis and search threedimensional model.
Background technique
Existing 3D pattern search system image or even can execute ratio by input 3D model by using sketch
Compared with search.Most of 3D pattern search system postulation object modules are rigid bodies.In addition, being input to the grass in 3D pattern search system
Figure and image are usually the form of front view and the side view perpendicular to front view.
However, not each object has rigid body characteristic.For example, there are many turning joints for human body.As user seeker
When body Model, if the arm of the human body of input or the position of leg are different from the modal position in database, or the figure of input
As not being front view and side view (for example, the image of input is perspective view), the search knot of this existing 3D pattern search system
Fruit is usually desired opposite with user.
The reason of difference discussed above be related to the prior art how by global characteristics analysis input data to execute
Compared with the model of storage in the database.Assuming that the model of input has moveable joint, even if they are identical
Model, when model is in different postures, their projection view is also different.Therefore, it is difficult to correct model is found,
And reduce the accuracy of search result.
In view of foregoing teachings, problems and disadvantages are related to the existing product required further improvement.However, art technology
Personnel not yet find solution.
Summary of the invention
The purpose of the present invention is to provide a kind of methods of analysis and search threedimensional model, to solve in above-mentioned background technique
The problem of proposition.
To achieve the above object, the invention provides the following technical scheme:
A method of analysis and search threedimensional model include the following steps:
(1) pass through global analysis and the local multiple data for obtaining multiple data global characteristics and multiple data images first
Local feature analyzes data image respectively;
(2) search image is obtained;
(3) then search global characteristics are obtained and are obtained by carrying out global analysis and partial analysis to search image respectively
Take the local feature of search image;
(4) corresponding data global characteristics are obtained from data global characteristics based on search global characteristics, and based on search
Local feature obtains corresponding data local feature from data local feature;
(5) it obtains from the corresponding of the local feature of image and corresponding data based on corresponding data global characteristics data
The image of data.
Preferably, in step (1), global analysis and partial analysis are carried out to the data image respectively first, obtained
The data local feature of the data local feature of the data global characteristics and the data image, the data image is included in
Multiple projected images of the data image are obtained and analyzed in different viewpoints;It is then based on the projected image of data image,
Correspondingly obtain the data global characteristics of data image;The projected image that will acquire data image later is divided into multiple Local maps
Picture and the topography based on data image correspondingly obtain the data local feature of data image.
Preferably, obtaining and analyze the projected image of data image in different points of view is the 3D mould for including by data image
Type is placed on the center of regular polygon, then in the different projected images of multiple apexes shooting 3D model of regular polygon.
Preferably, the projected image based on data image correspondingly obtains the data global characteristics of data image, datagram
The data global characteristics of picture include correspondingly obtaining the number of the projected image of data image by therefrom extracting feature and analysis
According to global characteristics, the data image projected image based on concentration gradient histogram and two dimension polarization Fourier transformation.
Preferably, it obtains the projected image of data image and is divided into topography, the topography is to pass through
The major part of each projected image of data image is obtained based on the analysis of the projected image of data image;By from datagram
Major part is removed in the projected image of picture, obtains the component of each projected image of data image.
Preferably, in step (3), image is searched for by global analysis and partial analysis first, it is global special to obtain search
It seeks peace and obtains the local feature of search image, the local feature of described search image includes analyzing search image in different points of view
Multiple projected images;It is then based on the projected image of search image, correspondingly obtains the search global characteristics of search image, is obtained
It searches for the projected image of image and is divided into multiple topographies;According to the topography of search image, correspondingly obtain
Search for the search local feature of image.
Preferably, the projected image of analysis search image is put by will search for the 3D model that image includes in different points of view
It sets at the center of regular polygon;Then in the different projected images of multiple apexes shooting 3D model of regular polygon;Based on searching
The search global characteristics that the projected image of rope image obtains search image are first by therefrom extracting feature and analyzing correspondingly
The search global characteristics for obtaining the projected image of search image, search for image projected image be based on concentration gradient histogram and
Two dimension polarization Fourier transformation.
Preferably, it obtains the projected image of search image and is divided into topography, the topography is to pass through
The major part of the projected image of search image is obtained based on the projected image of morphological analysis search image;By from search
Major part is removed in the projected image of image, obtains the component of the projected image of search image.
Preferably, the topography based on search image correspondingly obtains the search local feature of search image, described to search
Rope local feature is correspondingly to obtain the major part of search image and the search local feature of component by extracting.
Compared with prior art, beneficial effects of the present invention: the application of invention has many advantages, this disclosure relates to a kind of
Method for analyzing and searching for image, for solve due to using global characteristics analyze input data and be stored in data
Model in library is compared and causes the problem of search result inaccuracy.
Detailed description of the invention
Fig. 1 is flow diagram of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
Such as Fig. 1, it is proposed that following embodiments:
A method of analysis and search threedimensional model include the following steps:
(1): passing through global analysis and the local multiple numbers for obtaining multiple data global characteristics and multiple data images first
Data image is analyzed respectively according to local feature;Global analysis and partial analysis are carried out to the data image respectively first, obtained
The data local feature of the data local feature of the data global characteristics and the data image, the data image is included in
Multiple projected images of the data image are obtained and analyzed in different viewpoints;It is then based on the projected image of data image,
Correspondingly obtain the data global characteristics of data image;The projected image that will acquire data image later is divided into multiple Local maps
Picture and the topography based on data image correspondingly obtain the data local feature of data image;It is obtained in different points of view
It obtains and the projected image of analysis data image is that the 3D model for including is placed on the center of regular polygon by data image, then exist
The different projected images of multiple apexes shooting 3D model of regular polygon;Projected image based on data image correspondingly obtains
The data global characteristics of data image, the data global characteristics of data image include by extracting feature and analysis therefrom come corresponding
Ground obtains the data global characteristics of the projected image of data image, based on concentration gradient histogram and two dimension polarization Fourier transformation
Data image projected image;It obtains the projected image of data image and is divided into topography, the topography is
By the major part for obtaining each projected image of data image based on the analysis of the projected image of data image;By from number
According to major part is removed in the projected image of image, the component of each projected image of data image is obtained;
(2) search image is obtained;
(3) then search global characteristics are obtained and are obtained by carrying out global analysis and partial analysis to search image respectively
Take the local feature of search image;Image is searched for by global analysis and partial analysis first, obtain search global characteristics and is obtained
The local feature of search image is taken, the local feature of described search image includes analyzing multiple throwings of search image in different points of view
Shadow image;It is then based on the projected image of search image, correspondingly obtains the search global characteristics of search image, obtains search graph
The projected image of picture is simultaneously divided into multiple topographies;According to the topography of search image, search graph is correspondingly obtained
The search local feature of picture;The 3D model that the projected image that search image is analyzed in different points of view, which is by that will search for image, includes
It is placed on the center of regular polygon;Then in the different projected images of multiple apexes shooting 3D model of regular polygon;It is based on
The search global characteristics that the projected image of search image obtains search image are first by therefrom extracting feature and analyzing corresponding
Ground obtains the search global characteristics of the projected image of search image, and the projected image for searching for image is based on concentration gradient histogram
With two dimension polarization Fourier transformation;It obtains the projected image of search image and is divided into topography, the topography
It is the major part that the projected image of search image is obtained by searching for the projected image of image based on morphological analysis;Pass through
Major part is removed from the projected image of search image, obtains the component of the projected image of search image;Based on search
The topography of image correspondingly obtains the search local feature of search image, and described search local feature is corresponding by extracting
Ground obtains the major part of search image and the search local feature of component;
(4) corresponding data global characteristics are obtained from data global characteristics based on search global characteristics, and based on search
Local feature obtains corresponding data local feature from data local feature;
(5) it obtains from the corresponding of the local feature of image and corresponding data based on corresponding data global characteristics data
The image of data.
Working principle: application of the invention has many advantages, this disclosure relates to a kind of for analyzing and searching for image
Method, for solve to analyze input data due to using global characteristics and be compared with the model of storage in the database and
Lead to the problem of search result inaccuracy.
Although the present invention is described in detail referring to the foregoing embodiments, for those skilled in the art,
It is still possible to modify the technical solutions described in the foregoing embodiments, or part of technical characteristic is carried out etc.
With replacement, all within the spirits and principles of the present invention, any modification, equivalent replacement, improvement and so on should be included in this
Within the protection scope of invention.
Claims (9)
1. a kind of method of analysis and search threedimensional model, characterized by the following steps:
(1): passing through global analysis and the local multiple data offices for obtaining multiple data global characteristics and multiple data images first
Portion's feature analyzes data image respectively;
(2): obtaining search image;
(3): and then by carrying out global analysis and partial analysis to search image respectively, obtain search global characteristics and search is searched
The local feature of rope image;
(4): corresponding data global characteristics are obtained from data global characteristics based on search global characteristics, and based on search part
Feature obtains corresponding data local feature from data local feature;
(5): obtaining the respective counts from the local feature of image and corresponding data based on corresponding data global characteristics data
According to image.
2. a kind of method of analysis and search threedimensional model according to claim 1, it is characterised in that: in step (1),
Global analysis and partial analysis are carried out to the data image respectively first, obtain the data global characteristics and the datagram
The data local feature of picture, the data local feature of the data image include obtaining and analyzing the number in different viewpoints
According to multiple projected images of image;It is then based on the projected image of data image, the data for correspondingly obtaining data image are global
Feature;The projected image that will acquire data image later is divided into multiple topographies and the Local map based on data image
Picture correspondingly obtains the data local feature of data image.
3. a kind of method of analysis and search threedimensional model according to claim 2, it is characterised in that: in different points of view
It obtains and the projected image of analysis data image is that the 3D model for including is placed on the center of regular polygon by data image, then
In the different projected images of multiple apexes shooting 3D model of regular polygon.
4. a kind of method of analysis and search threedimensional model according to claim 3, it is characterised in that: be based on data image
Projected image correspondingly obtain the data global characteristics of data image, the data global characteristics of data image include by therefrom
Feature and analysis are extracted correspondingly to obtain the data global characteristics of the projected image of data image, is based on concentration gradient histogram
With the data image projected image of two dimension polarization Fourier transformation.
5. a kind of method of analysis and search threedimensional model according to claim 4, it is characterised in that: obtain data image
Projected image and be divided into topography, the topography be by the analysis of projected image based on data image come
Obtain the major part of each projected image of data image;By removing major part from the projected image of data image,
Obtain the component of each projected image of data image.
6. a kind of method of analysis and search threedimensional model according to claim 1, it is characterised in that: in step (3),
Image is searched for by global analysis and partial analysis first, obtains the local feature of search global characteristics and search search image,
The local feature of described search image includes analyzing multiple projected images of search image in different points of view;It is then based on search graph
The projected image of picture correspondingly obtains the search global characteristics of search image, obtains the projected image of search image and by its stroke
It is divided into multiple topographies;According to the topography of search image, the search local feature of search image is correspondingly obtained.
7. a kind of method of analysis and search threedimensional model according to claim 6, it is characterised in that: in different points of view
The projected image of analysis search image is that the center of regular polygon is placed on by the 3D model for including by search image;Then just
The different projected images of polyhedral multiple apex shooting 3D models;Projected image based on search image obtains search image
Search global characteristics be first by therefrom extract feature and analyze correspondingly obtain search image projected image search
Global characteristics, the projected image for searching for image is based on concentration gradient histogram and two dimension polarization Fourier transformation.
8. a kind of method of analysis and search threedimensional model according to claim 7, it is characterised in that: obtain search image
Projected image and be divided into topography, the topography be by based on morphological analysis search for image projection
Image come obtain search image projected image major part;By removing main portion from the projected image of search image
Point, obtain the component of the projected image of search image.
9. a kind of method of analysis and search threedimensional model according to claim 8, it is characterised in that: based on search image
Topography correspondingly obtain search image search local feature, described search local feature be by extract correspondingly obtain
The major part of image and the search local feature of component must be searched for.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811237227.2A CN109446969A (en) | 2018-10-23 | 2018-10-23 | A method of analysis and search threedimensional model |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811237227.2A CN109446969A (en) | 2018-10-23 | 2018-10-23 | A method of analysis and search threedimensional model |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109446969A true CN109446969A (en) | 2019-03-08 |
Family
ID=65548076
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811237227.2A Withdrawn CN109446969A (en) | 2018-10-23 | 2018-10-23 | A method of analysis and search threedimensional model |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109446969A (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102567465A (en) * | 2011-11-29 | 2012-07-11 | 西北工业大学 | Method for searching geometrical similarity of three-dimensional models on the basis of compound eye vision |
US20170147609A1 (en) * | 2015-11-19 | 2017-05-25 | National Chiao Tung University | Method for analyzing and searching 3d models |
CN107145487A (en) * | 2016-03-01 | 2017-09-08 | 深圳中兴力维技术有限公司 | Image search method and device |
CN108121806A (en) * | 2017-12-26 | 2018-06-05 | 湖北工业大学 | One kind is based on the matched image search method of local feature and system |
-
2018
- 2018-10-23 CN CN201811237227.2A patent/CN109446969A/en not_active Withdrawn
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102567465A (en) * | 2011-11-29 | 2012-07-11 | 西北工业大学 | Method for searching geometrical similarity of three-dimensional models on the basis of compound eye vision |
US20170147609A1 (en) * | 2015-11-19 | 2017-05-25 | National Chiao Tung University | Method for analyzing and searching 3d models |
TW201719572A (en) * | 2015-11-19 | 2017-06-01 | 國立交通大學 | Method for analyzing and searching 3D models |
CN107145487A (en) * | 2016-03-01 | 2017-09-08 | 深圳中兴力维技术有限公司 | Image search method and device |
CN108121806A (en) * | 2017-12-26 | 2018-06-05 | 湖北工业大学 | One kind is based on the matched image search method of local feature and system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102521595B (en) | Method for extracting image region of interest based on eye movement data and bottom-layer features | |
WO2015149302A1 (en) | Method for rebuilding tree model on the basis of point cloud and data driving | |
US20200242330A1 (en) | Method for object recognition | |
CN110135455A (en) | Image matching method, device and computer readable storage medium | |
CN108154104B (en) | Human body posture estimation method based on depth image super-pixel combined features | |
CN109446927A (en) | Double interbehavior recognition methods based on priori knowledge | |
Rallis et al. | Extraction of key postures from 3D human motion data for choreography summarization | |
Aiteanu et al. | Hybrid tree reconstruction from inhomogeneous point clouds | |
US20140300597A1 (en) | Method for the automated identification of real world objects | |
CN103871100A (en) | Method for rebuilding tree model based on point cloud and data driving | |
CN113343840B (en) | Object identification method and device based on three-dimensional point cloud | |
CN113449570A (en) | Image processing method and device | |
CN111950321A (en) | Gait recognition method and device, computer equipment and storage medium | |
CN109567839A (en) | Hip joint x-ray image automatic analysis method | |
CN108921062A (en) | A kind of gait recognition method for combining more gait feature collaboration dictionaries | |
TW201719572A (en) | Method for analyzing and searching 3D models | |
CN108734773A (en) | A kind of three-dimensional rebuilding method and system for mixing picture | |
CN107563366A (en) | A kind of localization method and device, electronic equipment | |
CN112906719A (en) | Standing tree factor measuring method based on consumption-level depth camera | |
CN109887009A (en) | A kind of point cloud local matching process | |
CN114240866B (en) | Tissue culture seedling grading method and device based on two-dimensional image and three-dimensional growth information | |
CN116152494A (en) | Building foot point identification segmentation method based on two-stage 3D point cloud semantic segmentation | |
Jiang et al. | Learned local features for structure from motion of uav images: A comparative evaluation | |
CN114689038A (en) | Fruit detection positioning and orchard map construction method based on machine vision | |
CN111914595B (en) | Human hand three-dimensional attitude estimation method and device based on color image |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20190308 |
|
WW01 | Invention patent application withdrawn after publication |