CN106407997A - System for identifying object through mobile device and identification method thereof - Google Patents
System for identifying object through mobile device and identification method thereof Download PDFInfo
- Publication number
- CN106407997A CN106407997A CN201610551918.4A CN201610551918A CN106407997A CN 106407997 A CN106407997 A CN 106407997A CN 201610551918 A CN201610551918 A CN 201610551918A CN 106407997 A CN106407997 A CN 106407997A
- Authority
- CN
- China
- Prior art keywords
- mobile device
- image
- pretreatment
- digital picture
- device identification
- 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
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/94—Hardware or software architectures specially adapted for image or video understanding
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a system for identifying an object through a mobile device and an identification method thereof. The system comprises the mobile device for capturing digital images, a preprocessing module for receiving information of the digital images and reducing image resolution, and a template feature matrix module for receiving the information of the preprocessing module and carrying out comparison and classification on feature vectors. According to the system for identifying the object through the mobile device and the identification method thereof, by reducing the image resolution to an appropriate level, burden of a processor of the mobile device is reduced to the maximum degree, and calculating efficiency is optimized; and the system has very high robustness.
Description
Technical field
The present invention relates to mobile image capture and process field, particularly a kind of system of use mobile device identification object
And its recognition methodss.
Background technology
It is known that in digital image processing field, connecting specific external equipment such as scanner by using computer,
Multifunctional equipment, can be very good to complete the task of image capturing and process, and technology is also highly developed.However, daily
Daily routines in, due to its unusual convenience, people increasingly tend to use mobile device, and this has significantly promoted and has passed through
Mobile device carries out the demand of image variants.
It is faced with obvious challenge using traditional image acquisition and treatment technology means on the mobile apparatus, such as, move
The restriction of the disposal ability of dynamic equipment and image resolution ratio is so that the Processing Algorithm based on scanner of routine is using movement
The Digital Image Processing of equipment capture is performed poor.
Additionally, suitable system resource can be expended using normal image Processing Algorithm, limited process on the mobile apparatus
Ability and storage also result in the performance bottleneck using normal image Processing Algorithm.
Another challenge is that the image being shot by mobile phone is different from traditional scanner, and the latter can verily reflect reality
The length-width ratio on border, size and dimension, the image of mobile device capture generally can not ensure such effect, and has the inclined of deformation
Difference.
For above-mentioned challenge it is necessary to provide algorithm and the application of a kind of image capturing and process, so that using moving
Equipment gathers, and processes and identification object, simultaneously using effective processing method to maintain relatively low calculating cost.
Content of the invention
For solving the deficiencies in the prior art, it is an object of the invention to provide a kind of use mobile device identification object is
System and its recognition methodss, the present invention, by being reduced to suitable resolution, decreases mobile device processor to greatest extent
Burden, and optimize computational efficiency, there is very strong robustness.
In order to realize above-mentioned target, the present invention adopts the following technical scheme that：
A kind of system of use mobile device identification object, including：The mobile device of capture digital picture, receives digital picture
Information simultaneously reduces the pretreatment module of image resolution ratio, receive the information of pretreatment module and compare, characteristic of division vector field homoemorphism
Plate features matrix module.
A kind of system of aforesaid use mobile device identification object, Digital Image Segmentation is become to have dimension by pretreatment module
Level and the pretreatment image reducing resolution of dimension vertical pixel.
A kind of recognition methodss of aforesaid use mobile device identification object, comprise the steps：
Step one, captures digital picture using mobile device；
Step 2, carries out reducing the pretreatment of image resolution ratio using processor to the digital picture receiving；
Step 3, is compared to pretreated information template characteristic matrix using processor and characteristic of division vector.
A kind of recognition methodss of aforesaid use mobile device identification object, step 2, using processor to receiving
Digital picture carries out reducing the pretreatment of image resolution ratio；The digital picture 300 being captured by mobile device can be divided into S
Part 304, each part 304 can comprise multiple pixels 306, forms the S x S comprising to tie up horizontal pixel and dimension vertical pixel
Grid.
A kind of recognition methodss of aforesaid use mobile device identification object, step 2, using processor to receiving
Digital picture carries out reducing the pretreatment of image resolution ratio；By Binarization methods, the digital service unit of masstone is become black
White image.
A kind of recognition methodss of aforesaid use mobile device identification object, step 2, using processor to receiving
Digital picture carries out reducing the pretreatment of image resolution ratio；By Binarization methods, the digital service unit of masstone is become black
White image；Binarization methods include：1）Determine color profile and the scheme of the class of an object, 2）The class of regulating object
Color profile or the detailed binaryzation parameter of color scheme, 3）Binaryzation parameter using adjusting thresholds image.
A kind of recognition methodss of aforesaid use mobile device identification object, step 2, using processor to receiving
Digital picture carries out reducing the pretreatment of image resolution ratio；By Binarization methods, the digital service unit of masstone is become black
White image；Binarization methods include：1）The background area of object is determined by the class mask of object, 2）The background area of regulating object
The color profile in domain or the detailed binaryzation parameter of color scheme, 3）Binaryzation parameter using adjusting thresholds image.
A kind of recognition methodss of aforesaid use mobile device identification object, it is one or more right that the class mask of object determines
The background area of elephant.
The invention has benefit that：The present invention provides a kind of system of use mobile device identification object and its identification
Method, the present invention captures digital picture by mobile device, is reduced the resolution of digital picture by pretreatment, then with polynary
The comparison of template characteristic matrix and classification, based on characteristics of objects, identify one or more objects belonging to special object apoplexy due to endogenous wind；Logical
Cross and be reduced to suitable resolution, decrease the burden of mobile device processor to greatest extent, and optimize computational efficiency, tool
There is very strong robustness.
Brief description
Fig. 1 is a kind of digital picture of embodiment of object of the present invention；
Fig. 2 for Fig. 1 be divided into multiple for producing pretreatment image；
Fig. 3 reduces the pretreatment image after characterization after resolution for Fig. 1；
The schematic diagram of the digital picture for many sub-regions of the present invention for the Fig. 4；
Fig. 5 is that a kind of mask of embodiment of Fig. 4 is shown；
Fig. 6 is that the mask of another embodiment of Fig. 4 is shown；
Fig. 7 is that the mask of another embodiment of Fig. 4 is shown；
Specific embodiment
Below in conjunction with the drawings and specific embodiments, the present invention is made with specific introduction.
A kind of system of use mobile device identification object, including：The mobile device of capture digital picture, receives digitized map
The information of picture simultaneously reduces the pretreatment module of image resolution ratio, receive the information of pretreatment module and compare, characteristic of division vector
Template characteristic matrix module.
Pretreatment module can be by digital picture R（There is y.sub.R dimension vertically divide with x.sub.R dimension horizontal pixel
Resolution）It is divided into the pretreatment image reducing resolution with S.sub.x dimension level and S.sub.y dimension vertical pixel.
Mobile device is the smart mobile phone commonly using on market；Processor is the processor of mobile phone, such as ARIXTRA
Mobile phone is valiant dragon, and i Phone is A4, A5.
A kind of recognition methodss of use mobile device identification object, comprise the steps：
Step one, captures digital picture using mobile device；
Step 2, carries out reducing the pretreatment of image resolution ratio using processor to the digital picture receiving；
Step 3, is compared to pretreated information template characteristic matrix using processor and characteristic of division vector.
The pretreatment concrete grammar reducing image resolution ratio is as follows：
Method one；As Fig. 1-2, the digital picture 300 being captured by mobile device can be divided into some 304, each portion
Divide 304 can comprise multiple pixels 306, formation comprises p.sub.S (x) dimension horizontal pixel and p.sub.S (y) ties up vertical pixel
Rectangular grid.p.sub.S(x)= p.sub.S(y)=4.In this example, digital picture 300 is divided into S part, wherein
Each part 304 corresponds to a part of the 302 of S x S grid.Pixel 312 in the pretreatment image upper left corner corresponds to
The digital picture of left 304.It should be noted that in order to reduce resolution, it would however also be possible to employ other methods generate pre-
Process image, the eigenvalue of image（Resolution, illumination, fuzziness）Object and the contrast of background, word makes with other symbol
With, etc..
Method two；The digital service unit of masstone is become black white image by Binarization methods by reference picture 4-7；Two
Value algorithm includes：1）Determine color profile and the scheme of the class of an object, 2）The color configuration of the class of regulating object
File or the detailed binaryzation parameter of color scheme, 3）Binaryzation parameter using adjusting thresholds image.
As a kind of embodiment, Fig. 4-5, improve the color profile that binaryzation can include determining that the class of an object
And scheme, the detailed binaryzation parameter of adjustment one or the color profile based on described object class or color scheme；With make
Binaryzation parameter with adjusting thresholds image.
As one kind preferably, Binarization methods determine the background area of object using class mask；Binarization methods include：1）
The background area of object is determined by the class mask of object, 2）The color profile of the background area of regulating object or color side
The detailed binaryzation parameter of case, 3）Binaryzation parameter using adjusting thresholds image.
It should be noted that the class mask of object determines the background area of one or more objects.Improved binaryzation side
Method can include determining that the class mask of an object.The class mask of object can be any kind of mask, there is provided interested
The specific region of feature class.
As shown in figure 5, improved binaryzation includes the region determined by class mask 420 determining an object, to exclude
The all of region of other in addition to single region interested of digital picture 400.Can be by both having obtained object class mask
With color profile and/or the Color scheme of object, adjust the colouring information of background area based on described object and/or face
Color scheme, and using the binaryzation parameter threshold after adjustment.
As shown in figure 5, object class mask 420 can also be used for multiple semi-cylindrical hills, optionally to specify area
Domain and/or parameter value, produce the layered approach of the bitonal image of high-quality.For example, literal fields 404,406 as multiple in Fig. 6
Binaryzation can be left（Can be using the parameter of adjustment）By mask after object class 420, with all from binarization exclusion
Non-textual region, etc..
The present invention provides a kind of system of use mobile device identification object and its recognition methodss, and the present invention passes through movement and sets
The resolution of digital picture is reduced by standby capture digital picture by pretreatment, then the comparison with polynary template characteristic matrix and
Classification, based on characteristics of objects, identifies one or more objects belonging to special object apoplexy due to endogenous wind；By being reduced to suitable resolution
Rate, decreases the burden of mobile device processor to greatest extent, and optimizes computational efficiency, has very strong robustness.
The basic principles, principal features and advantages of the present invention have been shown and described above.The technical staff of the industry should
Understand, the invention is not limited in any way for above-described embodiment, all obtained by the way of equivalent or equivalent transformation
Technical scheme, all falls within protection scope of the present invention.
Claims (8)
1. a kind of system of use mobile device identification object is it is characterised in that include：The mobile device of capture digital picture,
Receive the information of above-mentioned digital picture and reduce the pretreatment module of image resolution ratio, receive the information of above-mentioned pretreatment module simultaneously
Comparison, characteristic of division vector field homoemorphism plate features matrix module.
2. a kind of system of use mobile device identification object according to claim 1 is it is characterised in that above-mentioned pretreatment
Digital Image Segmentation is become to have dimension level and the pretreatment image reducing resolution of dimension vertical pixel by module.
3. a kind of use mobile device identification object according to claim 1 recognition methodss it is characterised in that include as
Lower step：
Step one, captures digital picture using mobile device；
Step 2, carries out reducing the pretreatment of image resolution ratio using processor to the digital picture receiving；
Step 3, is compared to pretreated information template characteristic matrix using processor and characteristic of division vector.
4. a kind of recognition methodss of use mobile device identification object according to claim 3 are it is characterised in that step
Two, the digital picture receiving is carried out reduce with the pretreatment of image resolution ratio using processor；The number being captured by mobile device
Word image 300 can be divided into S part 304, and each part 304 can comprise multiple pixels 306, formed and comprise dimension level
Pixel and the S x S grid of dimension vertical pixel.
5. a kind of recognition methodss of use mobile device identification object according to claim 3 are it is characterised in that step
Two, the digital picture receiving is carried out reduce with the pretreatment of image resolution ratio using processor；To be many by Binarization methods
The digital service unit of tone becomes black white image.
6. a kind of recognition methodss of use mobile device identification object according to claim 3 are it is characterised in that step
Two, the digital picture receiving is carried out reduce with the pretreatment of image resolution ratio using processor；To be many by Binarization methods
The digital service unit of tone becomes black white image；Binarization methods include：1）Determine the color configuration literary composition of the class of an object
Part and scheme, 2）Adjust the color profile of the class of above-mentioned object or the detailed binaryzation parameter of color scheme, 3）Using threshold
The binaryzation parameter of value adjustment image.
7. a kind of recognition methodss of use mobile device identification object according to claim 3 are it is characterised in that step
Two, the digital picture receiving is carried out reduce with the pretreatment of image resolution ratio using processor；To be many by Binarization methods
The digital service unit of tone becomes black white image；Binarization methods include：1）The back of the body of object is determined by the class mask of object
Scene area, 2）The color profile of the background area of regulating object or the detailed binaryzation parameter of color scheme, 3）Using threshold
The binaryzation parameter of value adjustment image.
8. a kind of recognition methodss of use mobile device identification object according to claim 7 are it is characterised in that above-mentioned right
The class mask of elephant determines the background area of one or more objects.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610551918.4A CN106407997A (en) | 2016-07-14 | 2016-07-14 | System for identifying object through mobile device and identification method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610551918.4A CN106407997A (en) | 2016-07-14 | 2016-07-14 | System for identifying object through mobile device and identification method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106407997A true CN106407997A (en) | 2017-02-15 |
Family
ID=58006927
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610551918.4A Pending CN106407997A (en) | 2016-07-14 | 2016-07-14 | System for identifying object through mobile device and identification method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106407997A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014160426A1 (en) * | 2013-03-13 | 2014-10-02 | Kofax, Inc. | Classifying objects in digital images captured using mobile devices |
| CN104537339A (en) * | 2014-12-19 | 2015-04-22 | 汉王科技股份有限公司 | Information identification method and information identification system |
| CN105308944A (en) * | 2013-03-13 | 2016-02-03 | 柯法克斯公司 | Classifying objects in images using mobile devices |
| CN105684058A (en) * | 2016-01-19 | 2016-06-15 | 冯旋宇 | License plate identifying method and license plate identifying system for intelligent traffic |
-
2016
- 2016-07-14 CN CN201610551918.4A patent/CN106407997A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2014160426A1 (en) * | 2013-03-13 | 2014-10-02 | Kofax, Inc. | Classifying objects in digital images captured using mobile devices |
| CN105283884A (en) * | 2013-03-13 | 2016-01-27 | 柯法克斯公司 | Classifying objects in digital images captured using mobile devices |
| CN105308944A (en) * | 2013-03-13 | 2016-02-03 | 柯法克斯公司 | Classifying objects in images using mobile devices |
| CN104537339A (en) * | 2014-12-19 | 2015-04-22 | 汉王科技股份有限公司 | Information identification method and information identification system |
| CN105684058A (en) * | 2016-01-19 | 2016-06-15 | 冯旋宇 | License plate identifying method and license plate identifying system for intelligent traffic |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107067389B (en) | A kind of blind evidence collecting method of distorted image | |
| Fadl et al. | Robust copy–move forgery revealing in digital images using polar coordinate system | |
| CN105426829B (en) | Video classification methods and device based on facial image | |
| WO2020147398A1 (en) | Reproduced image detection method and device, computer device and storage medium | |
| EP2366170B1 (en) | Media fingerprints that reliably correspond to media content with projection of moment invariants | |
| CN108960088A (en) | The detection of facial living body characteristics, the recognition methods of specific environment | |
| CN104573674A (en) | 1D (one-dimensional) barcode recognition for real-time embedded system | |
| CN112232204B (en) | Living body detection method based on infrared image | |
| CN103605993B (en) | Image-to-video face identification method based on distinguish analysis oriented to scenes | |
| CN102136065A (en) | Face super-resolution method based on convex optimization | |
| Yang et al. | Efficient reversible data hiding algorithm based on gradient-based edge direction prediction | |
| Dorai et al. | Detecting dynamic behavior in compressed fingerprint videos: Distortion | |
| CN110222647B (en) | Face in-vivo detection method based on convolutional neural network | |
| CN106713739B (en) | Two-dimensional code identification method based on electronic screen display | |
| CN1241389C (en) | Object-oriented self-synchronizing water mark detecting method for multimedia product | |
| CN102842133A (en) | Partial characteristic description method | |
| CN114612980A (en) | Deformed face detection based on multi-azimuth fusion attention | |
| CN112232205B (en) | Mobile terminal CPU real-time multifunctional face detection method | |
| CN115731591A (en) | Method, device and equipment for detecting makeup progress and storage medium | |
| CN111079689B (en) | Fingerprint image enhancement method | |
| CN106407997A (en) | System for identifying object through mobile device and identification method thereof | |
| Ahmad et al. | A joint biometrics and watermarking based framework for fingerprinting, copyright protection, proof of ownership, and security applications | |
| CN106446904A (en) | Image recognition method based on global binarization | |
| CN104182931A (en) | Super resolution method and device | |
| Kim et al. | Vehicle model recognition using SRGAN for low-resolution vehicle images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170215 |