CN103207994B - A kind of motion object kind identification method based on multi-project mode key morphological characteristic - Google Patents
A kind of motion object kind identification method based on multi-project mode key morphological characteristic Download PDFInfo
- Publication number
- CN103207994B CN103207994B CN201310154940.1A CN201310154940A CN103207994B CN 103207994 B CN103207994 B CN 103207994B CN 201310154940 A CN201310154940 A CN 201310154940A CN 103207994 B CN103207994 B CN 103207994B
- Authority
- CN
- China
- Prior art keywords
- morphological characteristic
- crucial
- motion
- perspective plane
- crucial morphological
- 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.)
- Active
Links
Landscapes
- Investigating Or Analysing Materials By Optical Means (AREA)
- Image Analysis (AREA)
Abstract
A kind of motion object kind identification method based on multi-project mode key morphological characteristic, relates to Manufacturing Automation Technology field, belongs to object monitor in real time and information gathering。The present invention first threedimensional model according to dissimilar object, sets up the crucial morphological characteristic on its multiple interrelated perspective planes respectively；Gather this motion object shape information from multiple perspective planes that are mutually related when motion object passes through；Automatically identifying of motion type of item is carried out in the difference of the crucial morphological characteristic on multiple interrelated perspective planes according to different objects。The present invention is a kind of motion object mixed running that can effectively solve the problem that number of different types universal method automatically identifying difficult problem when passing through。
Description
Technical field:
The present invention relates to Manufacturing Automation Technology field, belong to object monitor in real time and information gathering；It is specifically related to a kind of motion object form code recognition methods based on multi-project mode key morphological characteristic。
Technical background:
Recognition methods for motion object at present has a lot, and what be most widely used has bar code (including bar code, Quick Response Code, metallic barcode etc.) and electronic tag (RFID)；It addition, the method for image acquisition and pattern recognition also has more application。
Bar code and electronic tag can solve the problem that most of object identification problem in motor process, such as CN201828928U discloses " a kind of oil field based on planar bar code technology identification system is managed object ", and the method will be managed object and be implanted into Quick Response Code carrier and be used for article identification；And for example EP1271378-A is disclosed adopts bar code mode for identifying manufaturing data and the price tag of food。When adopting method identified above, what first should solve is the installation question of bar code or electronic tag, and under different application occasion, the position of its installation and mode are distinguishing, and from the expenditure installed, the recognition methods of bar code and electronic tag does not possess versatility；And in some industrial process, object needs in the way of motion through spray treatment or heat treatment, the recognition methods of bar code or electronic tag is now adopted to lose efficacy, because the figure of bar code is it cannot be guaranteed that can also be correctly validated after spraying, electronic tag cannot guarantee that and remains intact under the high temperature more than 200 degrees Celsius。Therefore, the recognition methods of existing motion object also cannot meet spraying, high temperature etc. particular application, it is necessary to a kind of more general recognition methods。
The method of image acquisition and pattern recognition is the comparatively conventional recognition methods of one, such as CN1529276 open " a kind of motor vehicles trade mark automatic identification equipment and recognition methods " and CN1506246 disclosed " a kind of vehicle license plate reading system ", is all the identification that carries out object moving of the mode by electronic image capture and image procossing。
But, the method of existing image acquisition and pattern recognition is mainly used in the identification of Text region and picture material, picture quality (includes light), and the impact of the process of identification is bigger, the algorithm of image steganalysis is generally complex, in motion article identification process, the error rate of identification can't be completely eliminated。
Summary of the invention:
For prior art above shortcomings, present invention aim to address the motion object of number of different types, the problem automatically identifying difficulty when certain operation is flow through in mixing, it is particularly well-suited to the particular application needs such as spraying, high temperature, it is proposed to a kind of comparatively general motion object kind identification method based on multi-project mode key morphological characteristic。
Realize above-mentioned purpose, the present invention adopts the following technical scheme that a kind of motion object kind identification method based on multi-project mode key morphological characteristic, it is characterized in that, carrying out automatically identifying of type of item according to different objects in the difference of the crucial morphological characteristic on multiple interrelated perspective planes, its process is as follows:
Step 1: set up the crucial morphological characteristic on multiple interrelated perspective plane for different objects respectively。
By the threedimensional model of object, set up the space geometry restriction relation between shape information and the perspective plane of its different projection (angle mainly having between perspective plane and the incidence relation in orientation), and choose crucial morphological characteristic on this basis。The view outline information that the shape information of object different projection is mainly under different angles object, the choosing of perspective plane and crucial morphological characteristic is primarily to the type to multiple object and distinguishes, so, need the form of all objects to be identified is analyzed, find out the perspective plane that wherein can distinguish all type of item to be identified and crucial morphological characteristic above thereof。In order to improve the accuracy rate of identification, it should select perspective plane to set up the shape information of object as much as possible more。
Step 2: the extraction of motion object key morphological characteristic。
1, on selected perspective plane, carry out the extraction of moving object part morphological characteristic。
Install object shape information harvester at object additional by the side in path, the process that object passes through carries out object shape information collection from different angles。In order to improve the accuracy rate of identification, it should the shape information of object is set up on more options perspective plane as much as possible, and under different application scenarios, harvester should dynamically regulate the angle on perspective plane as required, thus strengthening its versatility。
2, the information from objective pattern extracted it is analyzed and calculates, obtaining motion object key morphological characteristic。
Step 3: the identification of object。
The object shape information collected in step 2 is mated with the crucial morphological characteristic set up in step 1, is judged the type of this object by the difference of different object key morphological characteristics。This identification process is a synthetic determination process, and wherein, the incidence relation of the angle between perspective plane should as a crucial judgment basis in matching process。
Compared to existing technology, there is advantages that
1, the present invention flows through the process of certain operation (identification of the different object as passed through on conveyer belt in hybrid precast process) suitable in the mixing of multiple object, particularly process to the morphology influence of object bigger time, it is achieved the automatic accurate identification to type of item。
2, the present invention realizes the automatic identification of motion object according to the difference of the different types of object crucial morphological characteristic on different projection, utilize these crucial morphological characteristics to define to be mutually authenticated and synthetic determination, discrimination is greatly improved, error correcting capability is greatly improved。
3, the present invention takes full advantage of the feature that the crucial morphological characteristic of object is not affected by some special operation condition (such as spraying, high temperature, high pressure etc.) in certain moment, it is achieved that the correct identification of object。
4, the perspective plane number gathered in step 2 of the present invention is generally less than the perspective plane number chosen in step 1, increase along with the perspective plane quantity chosen in step 1, in actual decision process, the error of object form will reduce, and adding is the synthetic determination of multi-project mode key morphological characteristic。So, under certain conditions, object any attitude can flow through information collecting device, and the present invention further enhancing the versatility of the method。
Accompanying drawing illustrates:
Fig. 1 is the multi-project mode information from objective pattern exemplary plot of object in the embodiment of the present invention；
Fig. 2 is the present invention identification process figure based on the motion object kind identification method of multi-project mode key morphological characteristic；
Fig. 3 is the crucial morphological characteristic storehouse of object in Fig. 1；
Fig. 4 is that schematic diagram disposed by embodiment of the present invention middle infrared (Mid-IR) harvester。
Detailed description of the invention:
Below in conjunction with the drawings and specific embodiments, the present invention is described in further detail。
The implementation case being assumed, all objects are all put (it is suitable for the situation of random reverses direction by the above-below direction shown in Fig. 1, but need to consider more spatial shape) by conveyer belt, select the side projection form of object is carried out type identification and judgement。
In embodiment, the flow process of type of item identification as shown in Figure 2, specifically comprises the following steps that
The first step: respectively obtain the morphological characteristic on its multiple interrelated perspective planes for the geometric model of different objects, and choose crucial morphological characteristic。Wherein, interrelated perspective plane and crucial morphological characteristic need the morphological characteristic according to object to be identified to choose, having dynamic, its principle is: by united for the crucial morphological characteristic on these interrelated perspective planes, it is possible to type of item is uniquely identified。
Below the perspective plane of object in Fig. 1 and crucial choosing of morphological characteristic are illustrated:
(1) perspective plane is chosen: according to the morphological characteristic of object in Fig. 1, by turning clockwise of the top view of object, often rotate an angle a and can obtain a side projection form, as shown in Figure 1, the angle a selected in figure is 30 degree, and (this angle is more little, the error caused due to the putting phase place difference of motion object will be more little, therefore discrimination will be more high)。
(2) choose interrelated perspective plane: from figure 1 it appears that the morphological characteristic every 60 degree has notable difference, the type of object can be made a distinction, therefore the angle between the perspective plane in identification process is set as 60 degree。
(3) choose crucial morphological characteristic: owing in Fig. 1, object cross-sectional length in the level height of different projection there are differences, therefore choose from rising the ratio of horizontal cross-section length at 1/3 and 2/3 place with the height of object as crucial morphological characteristic。(length in the cross section collected can be produced impact by the difference of the speed v of object motion, therefore adopts the ratio of horizontal cross-section length on different projection to eliminate；May be otherwise and choose from rising the ratio of horizontal cross-section length at 1/4 and 3/4 place with the height of object as crucial morphological characteristic。)
(4) the interrelated relation between crucial morphological characteristic and the perspective plane of polytype object is built up crucial morphological characteristic storehouse, as shown in Figure 3, wherein crucial morphological characteristic one is the ratio from top 1/3 and 2/3 place's cross-sectional length, and crucial morphological characteristic two is object height。
Second step: the extraction of motion object key morphological characteristic。
(1) collection of motion object morphological characteristic。When motion object passes through, harvester is installed in its side and realizes the collection of motion object key morphological characteristic。
Adopting infrared induction device to gather the form on object perspective plane in the implementation case, as it is shown on figure 3, wherein, the angle between two groups of infrared induction device AA ' and BB ' is 60 degree。At object when moving through AA ' and BB ', infrared induction device can collect object profile information (often organize infrared induction device precision can choose according to the needs of object to be identified)。
(2) extraction of crucial morphological characteristic。
Crucial morphological characteristic selected by (3) in the first step, is analyzed above profile information by the method for mathematics and calculates, obtaining the crucial morphological characteristic value on two perspective planes of AA ' and BB ' place。
3rd step: the identification of type of item。
Crucial morphological characteristic in the first step and the object key morphological characteristic extracted in second step are compared, the crucial morphological characteristic value at AA ' and BB ' place is mated in crucial morphological characteristic storehouse, namely can determine that the type of this object。
Last explanation:
(1) in order to increase discrimination, the angle a chosen should be little as much as possible；And in concrete matching process, owing to the crucial morphological characteristic value that phase place may make to calculate of putting of motion object has certain deviation with the value in crucial morphological characteristic storehouse, it is possible to adopt error concealment mechanism to eliminate。
(2) above case study on implementation is only in order to illustrate technical scheme and unrestricted, technical scheme can be modified or equivalent replacement, without deviating from objective and the scope of technical solution of the present invention, it all should be encompassed in the middle of scope of the presently claimed invention。
Claims (1)
1. the motion object kind identification method based on multi-project mode key morphological characteristic, it is characterised in that specifically comprise the following steps that
Step 1: respectively obtain the morphological characteristic on its multiple interrelated perspective planes for the geometric model of different objects, and choose crucial morphological characteristic；Wherein, interrelated perspective plane and crucial morphological characteristic need the morphological characteristic according to object to be identified to choose, having dynamic, its principle is: by united for the crucial morphological characteristic on these interrelated perspective planes, it is possible to type of item is uniquely identified；
Wherein, the choosing of the perspective plane of object and crucial morphological characteristic:
(1) choosing perspective plane: the morphological characteristic of object turns clockwise by the top view of object, often rotate an angle [alpha] and can obtain a side projection form, the angle [alpha] of selection is 30 degree；
(2) interrelated perspective plane is chosen: the angle between the perspective plane in identification process is set as 60 degree；
(3) crucial morphological characteristic is chosen: choose from rising the ratio of horizontal cross-section length at 1/3 and 2/3 place with the height of object as crucial morphological characteristic；
(4) the interrelated relation between crucial morphological characteristic and the perspective plane of polytype object being built up crucial morphological characteristic storehouse, wherein crucial morphological characteristic one is the ratio from top 1/3 and 2/3 place's cross-sectional length, and crucial morphological characteristic two is object height；
Step 2: the extraction of motion object key morphological characteristic:
(1) collection of motion object morphological characteristic: when motion object passes through, installs harvester and realizes the collection of motion object key morphological characteristic in its side；
Adopting infrared induction device to gather the form on object perspective plane, wherein, the angle between two groups of infrared induction device AA ' and BB ' is 60 degree；At object when moving through AA ' and BB ', infrared induction device collects the profile information of object；
(2) extraction of crucial morphological characteristic:
Crucial morphological characteristic selected by (3) in step 1, is analyzed above profile information by the method for mathematics and calculates, obtaining the crucial morphological characteristic value on two perspective planes of AA ' and BB ' place；
Step 3: the identification of type of item:
Crucial morphological characteristic in step 1 and the object key morphological characteristic extracted in step 2 are compared, the crucial morphological characteristic value at AA ' and BB ' place is mated in crucial morphological characteristic storehouse, namely can determine that the type of this object。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310154940.1A CN103207994B (en) | 2013-04-28 | 2013-04-28 | A kind of motion object kind identification method based on multi-project mode key morphological characteristic |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310154940.1A CN103207994B (en) | 2013-04-28 | 2013-04-28 | A kind of motion object kind identification method based on multi-project mode key morphological characteristic |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103207994A CN103207994A (en) | 2013-07-17 |
| CN103207994B true CN103207994B (en) | 2016-06-22 |
Family
ID=48755211
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310154940.1A Active CN103207994B (en) | 2013-04-28 | 2013-04-28 | A kind of motion object kind identification method based on multi-project mode key morphological characteristic |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103207994B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP3213292A4 (en) * | 2014-10-28 | 2018-06-13 | Hewlett-Packard Development Company, L.P. | Three dimensional object recognition |
| CN106846462B (en) * | 2016-12-30 | 2019-12-17 | 北京农业信息技术研究中心 | insect recognition device and method based on three-dimensional simulation |
| CN106951843A (en) * | 2017-03-09 | 2017-07-14 | 深圳创维－Rgb电子有限公司 | Dust mite detection method and device |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102928431A (en) * | 2012-10-24 | 2013-02-13 | 浙江工业大学 | Device for automatically grading pearls on line according to size and shape on basis of monocular multi-view machine vision |
| CN102968100A (en) * | 2012-11-07 | 2013-03-13 | 上海意利信息科技有限公司 | System and method for controlling automatic production line |
-
2013
- 2013-04-28 CN CN201310154940.1A patent/CN103207994B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102928431A (en) * | 2012-10-24 | 2013-02-13 | 浙江工业大学 | Device for automatically grading pearls on line according to size and shape on basis of monocular multi-view machine vision |
| CN102968100A (en) * | 2012-11-07 | 2013-03-13 | 上海意利信息科技有限公司 | System and method for controlling automatic production line |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103207994A (en) | 2013-07-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zheng et al. | Improved lane line detection algorithm based on Hough transform | |
| Du et al. | Vision-based approach towards lane line detection and vehicle localization | |
| CN103123722B (en) | Road object detection method and system | |
| CN106529443B (en) | The method for improving Hough variation detection lane line | |
| CN104217427B (en) | Lane line localization method in a kind of Traffic Surveillance Video | |
| CN106934817B (en) | Multi-attribute-based multi-target tracking method and device | |
| CN105513349B (en) | Mountainous area highway vehicular events detection method based on double-visual angle study | |
| CN106447697B (en) | A kind of specific moving-target fast tracking method based on moving platform | |
| CN109871745A (en) | Identify method, system and the vehicle of empty parking space | |
| CN103207994B (en) | A kind of motion object kind identification method based on multi-project mode key morphological characteristic | |
| CN102654902A (en) | Contour vector feature-based embedded real-time image matching method | |
| CN103383733A (en) | Lane video detection method based on half-machine study | |
| CN103345628A (en) | Target recognition and shape retrieval method based on hierarchical description | |
| CN109523790A (en) | A kind of license plate error correction method based on license plate similarity | |
| CN106650608A (en) | Identification method for rectangle locating frame in test paper without locating points | |
| CN103593695A (en) | Method for positioning DPM two-dimension code area | |
| CN105844227A (en) | Driver identity authentication method for school bus safety | |
| CN115144399B (en) | Assembly quality detection method and device based on machine vision | |
| CN105469380A (en) | Method and device for detecting shielding against object | |
| CN103425966B (en) | The method and apparatus that code is associated with part | |
| CN103886296A (en) | Fingerprint identification method and device based on feedback | |
| CN109948436A (en) | The method and device of vehicle on a kind of monitoring road | |
| CN104200675B (en) | Vehicle speed measurement method based on invariant feature matching | |
| CN106408593A (en) | Video-based vehicle tracking method and device | |
| CN103777253B (en) | Method for performing human-body detection through use of human-body sensor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant |