CN105574519B - A kind of identification dynamic character features carry out the method and system of intelligent door unlatching - Google Patents
A kind of identification dynamic character features carry out the method and system of intelligent door unlatching Download PDFInfo
- Publication number
- CN105574519B CN105574519B CN201610074542.2A CN201610074542A CN105574519B CN 105574519 B CN105574519 B CN 105574519B CN 201610074542 A CN201610074542 A CN 201610074542A CN 105574519 B CN105574519 B CN 105574519B
- Authority
- CN
- China
- Prior art keywords
- matching
- image
- video image
- dynamic video
- dynamic
- 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
Classifications
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C9/00—Individual registration on entry or exit
- G07C9/00174—Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys
- G07C9/00563—Electronically operated locks; Circuits therefor; Nonmechanical keys therefor, e.g. passive or active electrical keys or other data carriers without mechanical keys using personal physical data of the operator, e.g. finger prints, retinal images, voicepatterns
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/48—Matching video sequences
Abstract
The invention discloses the method and system that a kind of identification dynamic character features carry out intelligent door unlatching, and method includes: acquisition dynamic video image；Extract several key frame images in dynamic video image；Scanning binary conversion treatment after frame image and extract dynamic video image feature stream；Judge dynamic video image feature stream and matching image feature stream with the presence or absence of matching relationship；It triggers user and fingerprint recognition is carried out based on the fingerprint identification device on intelligent door；When the user fingerprints information is consistent with the Fingerprint Identity information, the unlatching of intelligent door is triggered.The present invention can hand over and accurately be matched to associated video image content, and the corresponding fingerprint identification process of starting, achieve the purpose that accurately matching and authentication, to ensure the safety of entire identification verifying, the safety that intelligent door is opened.
Description
Technical field
The present invention relates to video image technical fields, and in particular to a kind of identification dynamic character features progress intelligent door unlatching
Method and system.
Background technique
It can be related to acquisition, transmission, processing, display and the playback etc. to video image data during video image processing
Process, these processes together form the integral cycle of a system, successional can operate.In video image processing technology
In range it is most important be exactly include compress technique and processing technique of video image of image etc..Currently, mainstream in the market
Video image processing technology include: intellectual analysis processing, technology that video Penetrating Fog is anti-reflection, the processing of wide dynamic, super-resolution processing,
Above four kinds of processing techniques are introduced separately below.
Intelligent video analysis technology is to solve the problems, such as the important means of the screening of field of video monitoring big data, retrieval technique.
Country's Intellectual Analysis Technology can be divided into two major classes at present: one kind is the shifting by the methods of foreground extraction to the object in picture
It is dynamic to be detected, different behaviors is distinguished by setting rule, such as mix line, article is left, circumference；Another kind of is to utilize mould
Formula identification technology targetedly models the object of monitoring required in picture, to reach to the certain objects in video
Carry out detection and related application, such as vehicle detection, stream of people's statistics, Face datection application.
Existing video acquisition is higher to dynamic requirements, and the dynamic of entire video acquisition requires high, existing intelligent door
System is not accurate enough for Dynamic Data Acquiring, the security mechanisms also not responded to, and lacks a kind of efficient image identification, identity
The mechanism of identification is suitable for high-grade community, the development of intelligence or intelligence community.
Summary of the invention
The object of the present invention is to provide the method and system that a kind of identification dynamic character features carry out intelligent door unlatching, reduce
The search difficulty of matching of video data is acquired, realizes the security control of intelligent door system.
For this purpose, the present invention provides a kind of methods that identification dynamic character features carry out intelligent door unlatching, including walk as follows
It is rapid:
Dynamic video acquisition is carried out to the personage entered within the scope of video acquisition based on camera, obtains dynamic video figure
Picture；
Video image pretreatment is carried out to the dynamic video image, and movement inspection is carried out to the dynamic video image
It surveys, extracts several key frame images in dynamic video image；
Binary conversion treatment is carried out to each frame image in several key frame images of extraction, after scanning binary conversion treatment
Frame image simultaneously extracts dynamic video image feature stream；
By the matching image feature stream progress of all matching images in dynamic video image feature stream and intelligent door system
Match, judges dynamic video image feature stream and matching image feature stream with the presence or absence of matching relationship；
In the feature stream to be matched for judging a certain matching image in dynamic video image feature stream and all matching images
It deposits after the matching, triggering user carries out fingerprint recognition based on the fingerprint identification device on intelligent door；
Based on the user fingerprints information of acquisition and Fingerprint Identity information associated by a certain matching image whether phase one
It causes；
When the user fingerprints information is consistent with the Fingerprint Identity information, the unlatching of intelligent door is triggered.
It is described that dynamic video acquisition is carried out to the personage entered within the scope of video acquisition based on camera, obtain dynamic vision
Frequency image includes:
Statistical learning method based on Adaboost algorithm detects character features, judges whether personage enters view
Frequency acquisition range；
When judging that personage enters video acquisition range, dynamic video acquisition is carried out to the personage, obtains dynamic vision
Frequency image.
It is described to include: to dynamic video image progress video image pretreatment
Noise reduction and image enhancement are carried out to dynamic video image.
It is described that motion detection is carried out to the dynamic video image, extract several key frame images in dynamic video image
Include:
The extraction of several key frames is carried out to the dynamic video image based on the motion detection of three-frame difference.
Frame image after the scanning binary conversion treatment simultaneously extracts dynamic video image feature stream and includes:
Frame image after binary conversion treatment is divided into 4,9,16 parts of same size；
To every sub-fraction in 4,9,16 parts, then it is equally divided into 4 partial regions；
From top to bottom, from left to right, each Minimum Area for successively scanning entire binary map checks each area pixel point
Distribution situation, the different character symbol in the region is obtained according to different distribution situations；
Completely the frame image after a binary conversion treatment, the frame image after obtaining binary conversion treatment are based on 4,9,16 to scan process
3 partial feature streams.
The distribution situation for checking each area pixel point, obtains the different spy in the region according to different distribution situations
In the step of sign symbol, pixel and character symbol corresponding relationship are as follows: a=1000；B=0100；C=0010；D=0001；E=
1100；F=0110；G=0011；H=1001；I=1010；J=0101；K=0111；L=1011；M=1101；N=1110；
O=1111；P=0000, wherein 1 represents the region and has a pixel distribution, 0 represents the region is distributed without pixel, and a to p is indicated
Character symbol.
It is described to flow into dynamic video image feature stream and the matching image feature of all matching images in intelligent door system
Row matching judges dynamic video image feature stream and matching image feature stream with the presence or absence of matching relationship:
It is special to take out matching image feature stream corresponding to each matching image and dynamic video image in all matching images
Sign stream；
3 feature streams for comparing matching image feature stream and dynamic video image feature stream respectively, count corresponding feature
The similarity of stream, and obtain each matching image and dynamic video similarity value in all matching images；
When judging that similarity value corresponding to a certain matching image in intelligent door system is greater than threshold value, then dynamic is determined
There are matching relationships for matching image in video image and intelligent door system, otherwise judge the dynamic video image and intelligent door
Matching relationship is not present in matching image in system.
Correspondingly, the present invention also provides a kind of systems that identification dynamic character features carry out intelligent door unlatching, comprising:
Video acquisition module is adopted for carrying out dynamic video to the personage entered within the scope of video acquisition based on camera
Collection obtains dynamic video image；
Video pre-filtering module, for carrying out video image pretreatment to the dynamic video image, and to the dynamic
Video image carries out motion detection, extracts several key frame images in dynamic video image；
Video image processing module carries out at binaryzation for each frame image in several key frame images to extraction
It manages, the frame image after scanning binary conversion treatment simultaneously extracts dynamic video image feature stream；
Video image matching module, for by all matching images in dynamic video image feature stream and intelligent door system
Matching image feature stream is matched, and is judged that dynamic video image feature stream whether there is with matching image feature stream and is matched pass
System；
Finger print information obtains module, for judging a certain in dynamic video image feature stream and all matching images
Feature stream to be matched with image is deposited after the matching, and triggering user carries out fingerprint knowledge based on the fingerprint identification device on intelligent door
Not；
Finger print information identification module, for associated by user fingerprints information and a certain matching image based on acquisition
Whether Fingerprint Identity information is consistent；
Intelligent door opening module, for triggering when the user fingerprints information is consistent with the Fingerprint Identity information
The unlatching of intelligent door.
The video acquisition module detects character features for the statistical learning method based on Adaboost algorithm,
Judge whether personage enters video acquisition range；When judging that personage enters video acquisition range, the personage is carried out
Dynamic video acquisition, obtains dynamic video image.
The video pre-filtering module is used to carry out dynamic video image noise reduction and image enhancement, and poor based on three frames
The motion detection divided carries out the extraction of several key frames to the dynamic video image.
It compared with prior art, is a kind of statistical learning algorithm the present invention is based on the Face datection algorithm of Adaboost, it
Differentiated by the statistics to Haar feature whether face, can identify whether personage enters video acquisition range, from
And start entire video acquisition process.For the dynamic of acquisition video, takes and motion detection is carried out to video image, closed
Key frame image forms the key frame video flowing of key frame images, to carry out to the key frame images in key frame video flowing special
Sign stream extracts, so that the precision that video image needs matched target image is had submitted, according to target image in video area point
Target image and image to be matched are carried out characteristic matching by cloth information extraction feature, thus identification or search associated video region,
It can also be tracked in video simultaneously.For the capacity in matching database, matching database can be obtained ahead of time
In each matching image feature stream, for feature stream in each image, the specific feelings being distributed according to each fritter target point
Condition, which obtains character symbol and forms feature stream, carries out images match, and only needs the entire target area of single pass, avoids
It computes repeatedly, matches, greatly accelerate images match speed and efficiency.For entire matching process, it can hand over and accurately match
To associated video image content, and the corresponding fingerprint identification process of starting, reach the mesh of accurate matching and authentication
, to ensure the safety of entire identification verifying, the safety that intelligent door is opened.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
Other attached drawings are obtained according to these attached drawings.
Fig. 1 is that the identification dynamic character features of the embodiment of the present invention carry out the method flow diagram of intelligent door unlatching；
Fig. 2 is that the identification dynamic character features of the embodiment of the present invention carry out the system construction drawing of intelligent door unlatching.
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 all other
Embodiment shall fall within the protection scope of the present invention.
As described above, passing through the invention proposes a kind of method that identification dynamic character features carry out intelligent door unlatching
Dynamic video acquisition is carried out to the personage entered within the scope of video acquisition based on camera, obtains dynamic video image；To dynamic
State video image carries out video image pretreatment, and carries out motion detection to dynamic video image, extracts in dynamic video image
Several key frame images；Binary conversion treatment is carried out to each frame image in several key frame images of extraction, scans two-value
Change treated frame image and extracts dynamic video image feature stream；By institute in dynamic video image feature stream and intelligent door system
There is the matching image feature stream of matching image to be matched, judges whether are dynamic video image feature stream and matching image feature stream
There are matching relationships；In the spy to be matched for judging a certain matching image in dynamic video image feature stream and all matching images
Sign stream is deposited after the matching, and triggering user carries out fingerprint recognition based on the fingerprint identification device on intelligent door；Use based on acquisition
Whether family finger print information and Fingerprint Identity information associated by a certain matching image are consistent；In user fingerprints information and institute
State Fingerprint Identity information it is consistent when, trigger the unlatching of intelligent door.
The method that the identification dynamic character features in the embodiment of the present invention carry out intelligent door unlatching is shown with reference to Fig. 1, Fig. 1
Flow chart, this method include the following:
S101, dynamic video acquisition is carried out to the personage entered within the scope of video acquisition based on camera, obtains dynamic
Video image；
It should be noted that being primarily based on the statistical learning method of Adaboost algorithm to personage in specific implementation process
Feature is detected, and judges whether personage enters video acquisition range；It is right when judging that personage enters video acquisition range
The personage carries out dynamic video acquisition, obtains dynamic video image.Character features are not full during entire video acquisition
When the property of face, do not start video acquisition generally, these character features include face, entire human body contour outline etc..
S102, video image pretreatment is carried out to dynamic video image；
It should be noted that needing to be related to carry out noise reduction to dynamic video image in whole image preprocessing process
And image enhancement, guarantee key-frame extraction in video data.
S103, motion detection is carried out to dynamic video image, extracts several key frame images in dynamic video image；
In the specific implementation process, several key frames are carried out to dynamic video image based on the motion detection of three-frame difference
It extracts.
S104, binary conversion treatment is carried out to each frame image in several key frame images of extraction, scanned at binaryzation
Frame image after reason simultaneously extracts dynamic video image feature stream；
Whole image feature stream process is as follows:
Frame image after binary conversion treatment is divided into 4,9,16 parts of same size；
To every sub-fraction in 4,9,16 parts, then it is equally divided into 4 partial regions；
From top to bottom, from left to right, each Minimum Area for successively scanning entire binary map checks each area pixel point
Distribution situation, the different character symbol in the region is obtained according to different distribution situations；
Completely the frame image after a binary conversion treatment, the frame image after obtaining binary conversion treatment are based on 4,9,16 to scan process
3 partial feature streams.
It should be noted that checking the distribution situation of each area pixel point, which is obtained according to different distribution situations
In the step of different character symbol in domain, pixel and character symbol corresponding relationship are as follows: a=1000；B=0100；C=0010；D=
0001；E=1100；F=0110；G=0011；H=1001；I=1010；J=0101；K=0111；L=1011；M=1101；
N=1110；O=1111；P=0000, as shown in following table table 1, wherein 1 represents the region and has pixel distribution, 0 represents the region
No pixel distribution, a to p indicate character symbol.
1 pixel of table is distributed table corresponding with character symbol
| a | b | c | d | ||||||||
| 1 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | ||||
| 0 | 0 | 0 | 0 | 0 | 1 | 1 | 0 | ||||
| e | f | g | h | ||||||||
| 1 | 1 | 0 | 1 | 0 | 0 | 1 | 0 | ||||
| 0 | 0 | 0 | 1 | 1 | 1 | 1 | 0 | ||||
| i | j | k | l | ||||||||
| 1 | 0 | 0 | 1 | 0 | 1 | 1 | 0 | ||||
| 0 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | ||||
| m | n | o | p | ||||||||
| 1 | 1 | 1 | 1 | 1 | 1 | 0 | 0 | ||||
| 1 | 0 | 0 | 1 | 1 | 1 | 0 | 0 |
The method for defining the matching degree of character an x, y, table is matched in correspondence, defines the similarity of two characters
For after dis (x, y): x and y exclusive or 1 number, dis (x, y) numerical value indicates more greatly character x, and the Regional Similarity that y is represented is more not
High (for example, after x is matched with x, dis=0 indicates that the two regions are infinitely similar).
S105, dynamic video image feature stream and the matching image feature of all matching images in intelligent door system are flowed into
Row matching；
S106, judge that dynamic video image feature stream and matching image feature stream whether there is matching relationship, if there is
Matching relationship then enters S107, otherwise terminates process；
In specific implementation process, the process of intelligent recognition piece identity judges personage's letter in dynamic video image feature
Whether breath is a sub-picture in matching image, and specific implementation step includes: to take out each matching in all matching images first
Matching image feature stream and dynamic video image feature stream corresponding to image；Matching image feature stream and dynamic vision are compared respectively
3 feature streams of frequency characteristics of image stream, count the similarity of corresponding feature stream, and obtain each matching in all matching images
Image and dynamic video similarity value；Judging similarity value corresponding to a certain matching image in intelligent door system greater than threshold
When value, then determining dynamic video image, there are matching relationships with the matching image in intelligent door system, otherwise judge the dynamic
Matching relationship is not present in matching image in video image and intelligent door system.
S107, triggering user carry out fingerprint recognition based on the fingerprint identification device on intelligent door；
After meeting identification, it is found that the user is a member in images match library, then enter fingerprint identification process, protects
Barrier is the legal input process under legitimate user's information.There is corresponding decoded information involved by each matching image, decoding letter
Breath ensures that user inputs the safety of information, to ensure the safety that intelligent door is opened.
S108, based on the user fingerprints information of acquisition and Fingerprint Identity information associated by a certain matching image whether phase one
It causes, if consistent, enter S109, otherwise terminate process；
S109, when user fingerprints information is consistent with the Fingerprint Identity information, trigger the unlatching of intelligent door.
Terminate.
In specific implementation process, the identification of dynamic character features is realized first, to restart the unlatching of gate inhibition, is ensured
The safety that entire intelligent door is opened ensures entire community or house safety, realizes intelligent residence.
The system structure that Fig. 2 also shows the progress intelligent door unlatching of the identification dynamic character features in the embodiment of the present invention is shown
It is intended to, which includes:
Video acquisition module is adopted for carrying out dynamic video to the personage entered within the scope of video acquisition based on camera
Collection obtains dynamic video image；
Video pre-filtering module, for carrying out video image pretreatment to the dynamic video image, and to the dynamic
Video image carries out motion detection, extracts several key frame images in dynamic video image；
Video image processing module carries out at binaryzation for each frame image in several key frame images to extraction
It manages, the frame image after scanning binary conversion treatment simultaneously extracts dynamic video image feature stream；
Video image matching module, for by all matching images in dynamic video image feature stream and intelligent door system
Matching image feature stream is matched, and is judged that dynamic video image feature stream whether there is with matching image feature stream and is matched pass
System；
Finger print information obtains module, for judging a certain in dynamic video image feature stream and all matching images
Feature stream to be matched with image is deposited after the matching, and triggering user carries out fingerprint knowledge based on the fingerprint identification device on intelligent door
Not；
Finger print information identification module, for associated by user fingerprints information and a certain matching image based on acquisition
Whether Fingerprint Identity information is consistent；
Intelligent door opening module, for triggering when the user fingerprints information is consistent with the Fingerprint Identity information
The unlatching of intelligent door.
In specific implementation process, the video acquisition module is for the statistical learning method based on Adaboost algorithm to personage
Feature is detected, and judges whether personage enters video acquisition range；It is right when judging that personage enters video acquisition range
The personage carries out dynamic video acquisition, obtains dynamic video image.
In specific implementation process, which is used to carry out noise reduction and image enhancement to dynamic video image,
And the extraction of several key frames is carried out to dynamic video image based on the motion detection of three-frame difference.
In specific implementation process, characteristics of image stream process is as follows in the video image processing module: after binary conversion treatment
Frame image be divided into 4,9,16 parts of same size；To every sub-fraction in 4,9,16 parts, then it is equally divided into 4 part areas
Domain；From top to bottom, from left to right, each Minimum Area for successively scanning entire binary map checks minute of each area pixel point
Cloth situation obtains the different character symbol in the region according to different distribution situations；Scan process is completely after a binary conversion treatment
Frame image, 3 feature streams of the frame image based on 4,9,16 parts after obtaining binary conversion treatment.
In specific implementation process, the video image matching module is for taking out each matching image institute in all matching images
Corresponding matching image feature stream and dynamic video image feature stream；Matching image feature stream and dynamic video image are compared respectively
3 feature streams of feature stream, count the similarity of corresponding feature stream, and obtain in all matching images each matching image with
Dynamic video similarity value；When judging that similarity value corresponding to a certain matching image in intelligent door system is greater than threshold value,
Then determining dynamic video image, there are matching relationships with the matching image in intelligent door system, otherwise judge the dynamic video figure
As matching relationship is not present with the matching image in intelligent door system.
Face datection algorithm the present invention is based on Adaboost is a kind of statistical learning algorithm, it passes through to Haar feature
Statistics come differentiate whether face, can identify whether personage enters video acquisition range, be adopted to start entire video
Collection process.It for the dynamic of acquisition video, takes and motion detection is carried out to video image, obtain key frame images, formed and closed
The key frame video flowing of key frame image, so that feature stream extraction is carried out to the key frame images in key frame video flowing, to mention
It has handed over video image to need the precision of matched target image, feature is extracted in video area distributed intelligence according to target image,
Target image and image to be matched are subjected to characteristic matching, thus identification or search associated video region, while can also carry out
It is tracked in video.For the capacity in matching database, each matching image in matching database can be obtained ahead of time
Feature stream obtain character symbol simultaneously according to the concrete condition that each fritter target point is distributed for feature stream in each image
It forms feature stream and carries out images match, and only need the entire target area of single pass, avoid and compute repeatedly, match,
Greatly accelerate images match speed and efficiency.For entire matching process, it can hand over and accurately be matched to associated video
Picture material, and the corresponding fingerprint identification process of starting, achieve the purpose that accurately matching and authentication, to ensure entire
The safety of identification verifying, the safety that intelligent door is opened.
It is provided for the embodiments of the invention the method and system that identification dynamic character features carry out intelligent door unlatching above
It is described in detail, used herein a specific example illustrates the principle and implementation of the invention, the above reality
The explanation for applying example is merely used to help understand method and its core concept of the invention；Meanwhile for the general technology of this field
Personnel, according to the thought of the present invention, there will be changes in the specific implementation manner and application range, in conclusion this theory
Bright book content should not be construed as limiting the invention.
Claims (9)
1. a kind of method that identification dynamic character features carry out intelligent door unlatching, which comprises the steps of:
Dynamic video acquisition is carried out to the personage entered within the scope of video acquisition based on camera, obtains dynamic video image；
Video image pretreatment is carried out to the dynamic video image, and motion detection is carried out to the dynamic video image, is mentioned
Take several key frame images in dynamic video image；
Binary conversion treatment is carried out to each frame image in several key frame images of extraction, the frame figure after scanning binary conversion treatment
Picture simultaneously extracts dynamic video image feature stream, and the frame image after the scanning binary conversion treatment simultaneously extracts dynamic video image feature
Stream includes: 4,9,16 parts that the frame image after binary conversion treatment is divided into same size；To each small in 4,9,16 parts
Part, then it is equally divided into 4 partial regions；From top to bottom, from left to right, each Minimum Area of entire binary map is successively scanned,
The distribution situation for checking each area pixel point obtains the different character symbol in the region according to different distribution situations；At scanning
Frame image after complete binary conversion treatment of reason, 3 features of the frame image based on 4,9,16 parts after obtaining binary conversion treatment
Stream；
Dynamic video image feature stream is matched with the matching image feature stream of all matching images in intelligent door system, is sentenced
Disconnected dynamic video image feature stream and matching image feature stream whether there is matching relationship；
Exist in the feature stream to be matched for judging a certain matching image in dynamic video image feature stream and all matching images
After matching, triggering user carries out fingerprint recognition based on the fingerprint identification device on intelligent door；
It is whether consistent based on the user fingerprints information of acquisition and Fingerprint Identity information associated by a certain matching image；
When the user fingerprints information is consistent with the Fingerprint Identity information, the unlatching of intelligent door is triggered.
2. the method that identification dynamic character features carry out intelligent door unlatching as described in claim 1, which is characterized in that the base
Dynamic video acquisition is carried out to the personage entered within the scope of video acquisition in camera, obtaining dynamic video image includes:
Statistical learning method based on Adaboost algorithm detects character features, judges whether personage enters video and adopt
Collect range；
When judging that personage enters video acquisition range, dynamic video acquisition is carried out to the personage, obtains dynamic video figure
Picture.
3. the method that identification dynamic character features carry out intelligent door unlatching as described in claim 1, which is characterized in that described right
The dynamic video image carries out video image pretreatment
Noise reduction and image enhancement are carried out to dynamic video image.
4. the method that identification dynamic character features carry out intelligent door unlatching as described in claim 1, which is characterized in that described right
The dynamic video image carries out motion detection, and several key frame images extracted in dynamic video image include:
The extraction of several key frames is carried out to the dynamic video image based on the motion detection of three-frame difference.
5. the method that identification dynamic character features carry out intelligent door unlatching as described in claim 1, which is characterized in that described to look into
The distribution situation for seeing each area pixel point, in the step of obtaining the different character symbol in the region according to different distribution situations,
Pixel and character symbol corresponding relationship are as follows: a=1000；B=0100；C=0010；D=0001；E=1100；F=0110；G=
0011；H=1001；I=1010；J=0101；K=0111；L=1011；M=1101；N=1110；O=1111；P=0000,
It wherein 1 represents the region and has a pixel distribution, 0 represents the region is distributed without pixel, and a to p indicates character symbol.
6. the method that identification dynamic character features carry out intelligent door unlatching as claimed in claim 5, which is characterized in that described to incite somebody to action
Dynamic video image feature stream is matched with the matching image feature stream of all matching images in intelligent door system, judges dynamic
Video image characteristic stream includes: with the presence or absence of matching relationship with matching image feature stream
Take out matching image feature stream and dynamic video image feature stream corresponding to each matching image in all matching images；
3 feature streams for comparing matching image feature stream and dynamic video image feature stream respectively, count corresponding feature stream
Similarity, and obtain each matching image and dynamic video similarity value in all matching images；
When judging that similarity value corresponding to a certain matching image in intelligent door system is greater than threshold value, then dynamic video is determined
There are matching relationships for matching image in image and intelligent door system, otherwise judge the dynamic video image and intelligent door system
In matching image be not present matching relationship.
7. a kind of system that identification dynamic character features carry out intelligent door unlatching characterized by comprising
Video acquisition module, for carrying out dynamic video acquisition to the personage entered within the scope of video acquisition based on camera,
Obtain dynamic video image；
Video pre-filtering module, for carrying out video image pretreatment to the dynamic video image, and to the dynamic video
Image carries out motion detection, extracts several key frame images in dynamic video image；
Video image processing module carries out binary conversion treatment for each frame image in several key frame images to extraction,
Scanning binary conversion treatment after frame image and extract dynamic video image feature stream, it is described scanning binary conversion treatment after frame image
And extracting dynamic video image feature stream includes: 4,9,16 parts that the frame image after binary conversion treatment is divided into same size；
To every sub-fraction in 4,9,16 parts, then it is equally divided into 4 partial regions；From top to bottom, from left to right, successively scanning is entire
Each Minimum Area of binary map checks the distribution situation of each area pixel point, obtains the area according to different distribution situations
The different character symbol in domain；The complete frame image after a binary conversion treatment of scan process, the frame image base after obtaining binary conversion treatment
3 feature streams in 4,9,16 parts；
Video image matching module, for by the matching of all matching images in dynamic video image feature stream and intelligent door system
Characteristics of image stream is matched, and judges dynamic video image feature stream and matching image feature stream with the presence or absence of matching relationship；
Finger print information obtains module, for judging a certain matching figure in dynamic video image feature stream and all matching images
The feature stream to be matched of picture is deposited after the matching, and triggering user carries out fingerprint recognition based on the fingerprint identification device on intelligent door；
Finger print information identification module, for fingerprint associated by user fingerprints information and a certain matching image based on acquisition
Whether identity information is consistent；
Intelligent door opening module, for when the user fingerprints information is consistent with the Fingerprint Identity information, triggering to be intelligent
The unlatching of door.
8. the system that identification dynamic character features carry out intelligent door unlatching as claimed in claim 7, which is characterized in that the view
Frequency acquisition module detects character features for the statistical learning method based on Adaboost algorithm, judge personage whether into
Enter to video acquisition range；When judging that personage enters video acquisition range, dynamic video acquisition is carried out to the personage, is obtained
Obtain dynamic video image.
9. the system that identification dynamic character features carry out intelligent door unlatching as claimed in claim 7, which is characterized in that the view
Frequency preprocessing module is used to carry out noise reduction and image enhancement, and the motion detection pair based on three-frame difference to dynamic video image
The dynamic video image carries out the extraction of several key frames.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610074542.2A CN105574519B (en) | 2016-02-02 | 2016-02-02 | A kind of identification dynamic character features carry out the method and system of intelligent door unlatching |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610074542.2A CN105574519B (en) | 2016-02-02 | 2016-02-02 | A kind of identification dynamic character features carry out the method and system of intelligent door unlatching |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105574519A CN105574519A (en) | 2016-05-11 |
| CN105574519B true CN105574519B (en) | 2018-12-11 |
Family
ID=55884627
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610074542.2A Active CN105574519B (en) | 2016-02-02 | 2016-02-02 | A kind of identification dynamic character features carry out the method and system of intelligent door unlatching |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105574519B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107168379A (en) * | 2017-06-07 | 2017-09-15 | 深圳市鑫益嘉科技股份有限公司 | A kind of dynamic tracing device and method for tracing |
| CN107967743A (en) * | 2017-12-21 | 2018-04-27 | 江苏国泰新点软件有限公司 | A kind of personal identification method being applied in e-bidding and system |
| CN110415398A (en) * | 2019-07-03 | 2019-11-05 | 华迪计算机集团有限公司 | A kind of gate inhibition's recognition methods and system based on double factor bio-identification |
| CN111539298A (en) * | 2020-04-20 | 2020-08-14 | 深知智能科技(金华)有限公司 | Identity information fusion system and method based on dynamic data |
| CN116453245B (en) * | 2023-04-20 | 2023-11-14 | 东莞市伟创动力科技有限公司 | Unlocking management method and system for electronic lock |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101140620A (en) * | 2007-10-16 | 2008-03-12 | 上海博航信息科技有限公司 | Human face recognition system |
| CN102799821A (en) * | 2012-07-11 | 2012-11-28 | 深圳市飞瑞斯科技有限公司 | Method for checking intelligent card and identity of card holder, and face identification identity checking device |
| CN103258191A (en) * | 2013-05-15 | 2013-08-21 | 苏州福丰科技有限公司 | Community access control system based on face recognition |
| KR20130104682A (en) * | 2012-03-15 | 2013-09-25 | 최상길 | Apparatus and method for automatically locking display and touch in mobile phone |
| CN103825871A (en) * | 2013-07-31 | 2014-05-28 | 深圳光启创新技术有限公司 | Authentication system and emission terminal, reception terminal and authority authentication method thereof |
-
2016
- 2016-02-02 CN CN201610074542.2A patent/CN105574519B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101140620A (en) * | 2007-10-16 | 2008-03-12 | 上海博航信息科技有限公司 | Human face recognition system |
| KR20130104682A (en) * | 2012-03-15 | 2013-09-25 | 최상길 | Apparatus and method for automatically locking display and touch in mobile phone |
| CN102799821A (en) * | 2012-07-11 | 2012-11-28 | 深圳市飞瑞斯科技有限公司 | Method for checking intelligent card and identity of card holder, and face identification identity checking device |
| CN103258191A (en) * | 2013-05-15 | 2013-08-21 | 苏州福丰科技有限公司 | Community access control system based on face recognition |
| CN103825871A (en) * | 2013-07-31 | 2014-05-28 | 深圳光启创新技术有限公司 | Authentication system and emission terminal, reception terminal and authority authentication method thereof |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105574519A (en) | 2016-05-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105574519B (en) | A kind of identification dynamic character features carry out the method and system of intelligent door unlatching | |
| Ma et al. | Robust precise eye location under probabilistic framework | |
| Davis et al. | A two-stage template approach to person detection in thermal imagery | |
| Kukharev et al. | Visitor identification-elaborating real time face recognition system | |
| CN108416336A (en) | A kind of method and system of intelligence community recognition of face | |
| CN105740675B (en) | A kind of method and system triggering empowerment management based on dynamic person recognition | |
| Ammouri et al. | Face and hands detection and tracking applied to the monitoring of medication intake | |
| Li et al. | Deep people counting with faster R-CNN and correlation tracking | |
| Bouchrika | Evidence evaluation of gait biometrics for forensic investigation | |
| Manikandan et al. | A neural network aided attuned scheme for gun detection in video surveillance images | |
| Hsiao et al. | EfficientNet based iris biometric recognition methods with pupil positioning by U-net | |
| Yang et al. | Detection and segmentation of latent fingerprints | |
| Anibou et al. | Classification of textured images based on discrete wavelet transform and information fusion | |
| Zhang et al. | A hybrid convolutional architecture for accurate image manipulation localization at the pixel-level | |
| CN112183504A (en) | Video registration method and device based on non-contact palm vein image | |
| Jaiswal et al. | Survey paper on various techniques of recognition and tracking | |
| Leo et al. | Highly usable and accurate iris segmentation | |
| Sukkar et al. | A Real-time Face Recognition Based on MobileNetV2 Model | |
| Li et al. | Action recognition based on multiple key motion history images | |
| Utami et al. | Face spoof detection by motion analysis on the whole video frames | |
| Ngambeki et al. | Real time face recognition using region-based segmentation algorithm | |
| Viriri et al. | Improving iris-based personal identification using maximum rectangular region detection | |
| Hadjkacem et al. | Multi-shot human re-identification using a fast multi-scale video covariance descriptor | |
| Hazar et al. | Real time face detection based on motion and skin color information | |
| Demirel et al. | Iris recognition system using combined colour statistics |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |