CN102036008A - Automatic focusing system and method for particular persons - Google Patents
Automatic focusing system and method for particular persons Download PDFInfo
- Publication number
- CN102036008A CN102036008A CN2009101778368A CN200910177836A CN102036008A CN 102036008 A CN102036008 A CN 102036008A CN 2009101778368 A CN2009101778368 A CN 2009101778368A CN 200910177836 A CN200910177836 A CN 200910177836A CN 102036008 A CN102036008 A CN 102036008A
- Authority
- CN
- China
- Prior art keywords
- characteristic vector
- focusing
- facial image
- human face
- database
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Automatic Focus Adjustment (AREA)
Abstract
The invention relates to an automatic focusing system for particular persons, comprising a data base, a face detection part and a face recognition part, wherein the data base is used for storing face images of most ethnic groups. As the face images of the same ethnic group have proximate feature vector, a preferential focusing object is set in the ethnic groups. The face detection part is used for detecting the face images, and the face recognition part is used for calculating the feature vector of the detected face images to compare with the feature vector of the preferential focusing object so as to find out a face image of which the feature vector is proximal to that of the preferential focusing object to focus preferentially. The invention also provides an automatic focusing method for the particular persons. Thus, the conveniences of automatic focusing are improved, and then the success rate of shooting is improved.
Description
Technical field
The present invention relates to a kind of autofocus system and method for image capturing device, refer to a kind of system and method that can do preferential focusing at the specific personage in the shooting picture especially.
Background technology
Existing capturing images device category is various, for example digital camera, video camera or mobile phone built-in camera etc., and each capturing images equipment all adopts different separately focusing technology, and in the present known focusing technology, comparatively Xian Jin focusing technology behaviour face focal modes.About the technology of people's face focal modes, briefly be to scan for the image in the picture comprehensively, when people's face appears in picture, then can focus to people's face.Early stage people's face focusing technology is to find out the people face do preferential focusing nearest with distance of camera lens, then comparatively progressive people's face focusing technology is to adopt the average distance position of maximum distance people's face and minimum distance people face to make preferential focusing position, so need only clear more that the approaching more people's face of distance priority focusing position can be taken.Yet when adopting present known people's face autofocus mode to do to take, the person of being taken that must work each time moves to the appropriate location, so all too inconvenience, if and the person of being taken has moved the position and because of carelessness away from preferential focusing position in shooting process, what just can't be taken is very clear, hence one can see that, the risk that present people's face focal modes still has the trouble part in the use and takes failure.
So the inventor improves thoughts on above-mentioned defective, be to concentrate on studies and cooperate the utilization of scientific principle, propose a kind of reasonable in design and effectively improve the present invention of above-mentioned defective finally.
Summary of the invention
In view of above problem, main purpose of the present invention makes that for a kind of specific personage's autofocus system and method are provided the focusing operation is convenient, has also promoted the success rate of taking.
To achieve the above object, the invention provides a kind of specific personage's autofocus system, comprise: a database, store most facial images of planting group in this database, the facial image that belongs to same group has approaching characteristic vector, the facial image that the belongs to different groups vector that takes on a different character, this database is selected wherein this group as a preferential focusing object in order to the user to be provided; One human face detection portion, this human face detection portion is in order to detect those facial images; One human face recognition portion, this human face recognition portion is electrically connected at this human face detection portion and this database, and this human face recognition portion is in order to the characteristic vector of calculating those facial images and find out and should preferential focusing object have at least one this facial image near characteristic vector; And a focusing execution portion, this focusing execution portion is electrically connected at this human face recognition portion, and this focusing execution portion has facial image near characteristic vector preferentially to focus in order to what this human face recognition portion was found out with this object of preferentially focusing.
The present invention provides a kind of specific personage the method for focusing automatically in addition, its step comprises: set up a database with facial image of multiple group, the facial image that belongs to same group has approaching characteristic vector, the facial image that the belongs to different groups vector that takes on a different character; Select one of them group in the process of taking pictures as a preferential focusing object; Facial image in the detecting picture; Calculate the characteristic vector of those facial images that detect; The facial image that detects of comparison is found out with the characteristic vector of preferential focusing object and the facial image that should preferential focusing object has near characteristic vector; And do preferential focusing at having near the facial image of characteristic vector of finding with this preferential focusing object.
The present invention has following useful effect: when the user need take the photo with theme feature, theme feature is the beauty for example, handsome boy or child, as long as the people's face that meets theme feature in the database is set at preferential focusing object, focusing execution portion just can do preferential focusing to the people's face that meets theme feature in the picture automatically, and the person of being taken that need work when not needing to take each time moves to the appropriate location, can finish the focusing action, so it is convenient in the operation, the shooting error that also can avoid subject person when taking, to cause because of the shift position, and then improve the success rate of taking.
Description of drawings
Fig. 1 is the calcspar of the specific personage's autofocus system of the present invention;
Fig. 2 is the calcspar of inventor's face Identification Division;
Fig. 3 is focus the automatically flow chart of method of the specific personage of the present invention;
Fig. 4 sets up the method flow diagram of database images data for the present invention;
Fig. 5 is focus the automatically schematic diagram () of shooting embodiment of method of the specific personage of the present invention;
Fig. 6 is focus the automatically schematic diagram (two) of shooting embodiment of method of the specific personage of the present invention.
[main element description of reference numerals]
Human face detection portion 1
Human face recognition portion 2
Characteristic vector calculating part 21
Group's division 22
Characteristic vector comparison portion 23
Focusing execution portion 4
The man 51
The woman 52
The old man 53
The fat person 55
Focusing frame 56
Embodiment
As shown in Figure 1, the invention provides a kind of specific personage's autofocus system, comprise a human face detection portion 1, a human face recognition portion 2, a database 3 and a focusing execution portion 4, wherein this human face detection portion 1 is electrically connected at this human face recognition portion 2, and this human face recognition portion 2 is electrically coupled with this database 3 and this focusing execution portion 4.
This human face detection portion 1 can adopt has learnt the most neural network of facial image and non-face image or vector machine (Vector Support Machine) that most facial images and non-face image have been learnt in employing in advance of opening in advance, and this human face detection portion 1 is responsible for detecting the image of people face position from the view data of many inputs.
As shown in Figure 2, this human face recognition portion 2 comprises a characteristic vector calculating part 21, heap sort portion of gang 22 and a characteristic vector comparison portion 23, and wherein this group's division 22 and characteristic vector comparison portion 23 are electrically connected at this characteristic vector calculating part 21 respectively.This characteristic vector calculating part 21 is electrically connected at this human face detection portion 1, the facial image that this characteristic vector calculating part 21 is detected at this human face detection portion 1 goes to calculate the characteristic vector (Eigen Vector) of each facial image, and the feature of characteristic vector representative face, for example popularly think the face that belongs to the beauty, there are some representational features in the capital, the for example ratio of face or the area of face etc., and these features adopt vector representations.The facial image that 22 of this group's divisions will have near characteristic vector goes to be categorized as same group, so-called group can distinguish with sex, for example schoolgirl and boy student, perhaps distinguish with the age, for example old man, youth and child, or distinguish with fat or thin and beauty and ugliness, also or with the similar face of household as differentiation.Because the face feature of different groups is difference to some extent, so belong to the facial image of different groups, its characteristic vector also can be inequality, and the facial image that belongs to same group then has approaching characteristic vector.Because this group's division 22 is electrically connected at this database 3, so can be stored in this database 3 through the good facial image of group division 22 classification, make to store in this database 3 and mostly plant the facial image of group and can provide the user from database 3, to select the preferential focusing object of a group as acquiescence.
This characteristic vector comparison portion 23 is responsible for the characteristic vector of comparison facial image, and this characteristic vector comparison portion 23 is electrically connected at this database 3 and this focusing execution portion 4, so human face detection portion 1 detects extraneous facial image each time, all can see through characteristic vector comparison portion 23 facial image and the human face detection portion 1 that has stored in the database 3 detected the comparison that new facial image carries out characteristic vector in the external world, and then find out and the preferential focusing object given tacit consent to has extremely facial image near characteristic vector.Find out when this characteristic vector comparison portion 23 many facial images of comparison have at least one facial image of approaching characteristic vector with the preferential focusing object of giving tacit consent to after, this 23 in characteristic vector comparison portion can send signal and inform this focusing execution portion 4, and this focusing execution portion 4 receives then to have extremely at these characteristic vector comparison portion 23 that find with default preferential focusing objects behind the signal and preferentially focuses near the facial image of characteristic vector.
As shown in Figure 3, the invention provides a kind of specific personage method of focusing automatically, its execution in step comprises S101~S106.
S101: the user sets up a database 3 with facial image of multiple group voluntarily, and the characteristic vector of the facial image of different groups can be inequality, and the facial image of same group has hand-to-hand characteristic vector.
S102: the user selects one of them group preferential focusing object as an acquiescence in the process of taking pictures in the database 3 that stores majority kind group, so-called preferential focusing object represents to meet in the picture personage of theme feature of this time taking pictures, and theme feature is beauty, handsome boy or child etc. for example.
S103: see through the facial image that human face detection portion 1 detects in the picture of taking pictures, the mode of its detecting can adopt detecting moving object and edge (edge) image to detect the profile of finding out people's face, can adopt colour of skin detecting algorithm to detect the position of people's face, can adopt profile to detect the organ of the face of classifying.
S104: see through the characteristic vector that characteristic vector calculating part 21 calculates those facial images that detect via human face detection portion 1, and characteristic vector is promptly represented face feature.
S105: see through facial image that 23 comparisons of characteristic vector comparison portion detect and find out with the characteristic vector of default preferential focusing object and the facial image that should preferential focusing object has extremely near characteristic vector.
S106: this focusing execution portion 4 at characteristic vector comparison portion 23 that find have near the facial image of characteristic vector with this preferential focusing object do preferential focusing.
As shown in Figure 4, mode how to set up this database 3 about the user comprises the following step.
S201: the user can capture many view data with people's face via various print medias or network, and those view data with people's face are inputed to human face detection portion 1;
S202: see through human face detection portion 1 detects people face position from those view data image;
S203: see through the characteristic vector that characteristic vector calculating part 21 calculates those facial images that detect; And
S204: see through group's division 22 and be categorized as same group, have most databases 3 of planting the facial image of group and set up out at last with having near characteristic vector person in those facial images that detect.
As shown in Figure 5, carry out the embodiment of actual photographed for adopting the specific personage of the present invention focusing method automatically, has the group that five faces have different characteristic in the shooting picture 5, be respectively man 51, woman 52, old man 53, child 54 and fat person 55, the preferential focusing object that this embodiment presets is woman 52, so even woman 52 face is apart from camera lens people's face farthest, focusing frame 56 still can be aimed at woman 52 and preferentially face's work of woman 52 be focused automatically.
As shown in Figure 6, carry out another embodiment of actual photographed for adopting the specific personage of the present invention focusing method automatically, also have five different groups in the shooting picture 5, be respectively man 51, woman 52, old man 53, child 54, fat person 55, the preferential focusing object of this embodiment acquiescence changes man 51 into, even man 51 face is apart from camera lens people's face farthest, focusing frame 56 is still aimed at man 51 and preferentially face's work of man 51 is focused automatically.
Specific personage's autofocus system of the present invention and method have following advantage:
1, when the user need take the photo with theme feature, as long as the facial image that meets theme feature in the database 3 is set at preferential focusing object, focusing execution portion 4 just can do preferential focusing to the people's face that meets theme feature in the picture automatically, and the person of being taken that need work when not needing to take each time moves the position, so that operation is gone up is convenient.
2, when taking,, improve the success rate of taking relatively even subject person shift position when taking can not change the priority that acquiescence is focused yet.
The above only is the present invention's preferred embodiment wherein, is not to be used for limiting claim protection range of the present invention, and promptly all equalizations of being made according to claim protection range of the present invention change and revise, and are all claim scope of the present invention and contain.
Claims (10)
1. specific personage's autofocus system is characterized in that, comprising:
One database, store most facial images of planting group in this database, the facial image that belongs to same group has approaching characteristic vector, the facial image that the belongs to different groups vector that takes on a different character, this database is selected wherein this group as a preferential focusing object in order to the user to be provided;
One human face detection portion, this human face detection portion is in order to detect those facial images;
One human face recognition portion, this human face recognition portion is electrically connected at this human face detection portion and this database, and this human face recognition portion is in order to the characteristic vector of calculating those facial images and find out and should preferential focusing object have at least one this facial image near characteristic vector; And
One focusing execution portion, this focusing execution portion is electrically connected at this human face recognition portion, and this focusing execution portion has facial image near characteristic vector preferentially to focus in order to this human face recognition portion is found out with the object of should preferentially focusing.
2. specific personage's autofocus system as claimed in claim 1 is characterized in that, this human face detection portion is for having learnt most neural networks of opening facial image and non-face image in advance.
3. specific personage's autofocus system as claimed in claim 1 is characterized in that, this human face detection portion is for having learnt most vector machine of opening facial image and non-face image in advance.
4. specific personage's autofocus system as claimed in claim 1, it is characterized in that, this human face recognition portion comprises a characteristic vector calculating part and heap sort portion of gang, this characteristic vector calculating part is in order to calculate the characteristic vector of facial image, this group's division and this characteristic vector calculating part and this database electrically connect, and this group's division is categorized as same group in order to the facial image that will have near characteristic vector.
5. specific personage's autofocus system as claimed in claim 4, it is characterized in that, this human face recognition portion more comprises a characteristic vector comparison portion, this characteristic vector comparison portion is electrically connected at this characteristic vector calculating part, this focusing execution portion and this database, and this characteristic vector comparison portion is in order to comparison and find out with the object of should preferentially focusing and have facial image near characteristic vector.
6. specific personage method of focusing automatically is characterized in that step comprises:
Set up a database with facial image of multiple group, the facial image that belongs to same group has approaching characteristic vector, the facial image that the belongs to different groups vector that takes on a different character;
Select one of them group in the process of taking pictures as a preferential focusing object;
Detect the facial image in the picture of taking pictures;
Calculate the characteristic vector of those facial images that detect;
The facial image that detects of comparison is found out with the characteristic vector of preferential focusing object and the facial image that should preferential focusing object has near characteristic vector; And
Do preferential focusing at having near the facial image of characteristic vector of finding with this preferential focusing object.
7. the specific personage as claimed in claim 6 method of focusing automatically is characterized in that the step of setting up this database comprises:
Import many view data;
Detect the image of people face position in the view data by those inputs;
Calculate the characteristic vector of those facial images that detect; And
Be categorized as same group with having near characteristic vector person in those facial images that detect.
8. the specific personage as claimed in claim 6 method of focusing automatically is characterized in that, the step of this detecting facial image adopts detecting moving object and edge image to detect the profile of finding out people's face.
9. the specific personage as claimed in claim 6 method of focusing automatically is characterized in that, the step of this detecting facial image adopts colour of skin detecting algorithm to detect the position of people's face.
10. the specific personage as claimed in claim 6 method of focusing automatically is characterized in that, the step of this detecting facial image adopts profile to detect the organ of the face of classifying.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009101778368A CN102036008A (en) | 2009-09-25 | 2009-09-25 | Automatic focusing system and method for particular persons |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2009101778368A CN102036008A (en) | 2009-09-25 | 2009-09-25 | Automatic focusing system and method for particular persons |
Publications (1)
Publication Number | Publication Date |
---|---|
CN102036008A true CN102036008A (en) | 2011-04-27 |
Family
ID=43888265
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2009101778368A Pending CN102036008A (en) | 2009-09-25 | 2009-09-25 | Automatic focusing system and method for particular persons |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN102036008A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103108124A (en) * | 2012-12-28 | 2013-05-15 | 上海鼎为软件技术有限公司 | Image acquisition method, device and mobile terminal |
CN103347117A (en) * | 2013-05-31 | 2013-10-09 | 东莞宇龙通信科技有限公司 | Communication terminal shooting method and communication terminal |
CN103888659A (en) * | 2012-12-19 | 2014-06-25 | 卡西欧计算机株式会社 | Image Capture Apparatus That Can Determine Appropriate Focus Position, Image Capture Method, And Storage Medium |
CN104125387A (en) * | 2013-04-25 | 2014-10-29 | 宏碁股份有限公司 | Photographing guidance method and electronic device |
CN106713734A (en) * | 2015-11-17 | 2017-05-24 | 华为技术有限公司 | Auto focusing method and apparatus |
CN108009521A (en) * | 2017-12-21 | 2018-05-08 | 广东欧珀移动通信有限公司 | Humanface image matching method, device, terminal and storage medium |
TWI660211B (en) * | 2014-12-26 | 2019-05-21 | 日商三美電機股份有限公司 | Lens driving device, camera module, and camera mounting device |
CN110418064A (en) * | 2019-09-03 | 2019-11-05 | 北京字节跳动网络技术有限公司 | Focusing method, device, electronic equipment and storage medium |
CN111083369A (en) * | 2019-12-26 | 2020-04-28 | 上海传英信息技术有限公司 | Face focusing method and device, mobile terminal and storage medium |
-
2009
- 2009-09-25 CN CN2009101778368A patent/CN102036008A/en active Pending
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103888659B (en) * | 2012-12-19 | 2018-03-20 | 卡西欧计算机株式会社 | Camera device, image capture method and computer-readable storage medium |
CN103888659A (en) * | 2012-12-19 | 2014-06-25 | 卡西欧计算机株式会社 | Image Capture Apparatus That Can Determine Appropriate Focus Position, Image Capture Method, And Storage Medium |
CN103108124B (en) * | 2012-12-28 | 2017-07-18 | 上海鼎为电子科技(集团)有限公司 | Image acquiring method, device and mobile terminal |
CN103108124A (en) * | 2012-12-28 | 2013-05-15 | 上海鼎为软件技术有限公司 | Image acquisition method, device and mobile terminal |
CN104125387A (en) * | 2013-04-25 | 2014-10-29 | 宏碁股份有限公司 | Photographing guidance method and electronic device |
CN103347117A (en) * | 2013-05-31 | 2013-10-09 | 东莞宇龙通信科技有限公司 | Communication terminal shooting method and communication terminal |
TWI660211B (en) * | 2014-12-26 | 2019-05-21 | 日商三美電機股份有限公司 | Lens driving device, camera module, and camera mounting device |
CN106713734A (en) * | 2015-11-17 | 2017-05-24 | 华为技术有限公司 | Auto focusing method and apparatus |
CN106713734B (en) * | 2015-11-17 | 2020-02-21 | 华为技术有限公司 | Automatic focusing method and device |
CN108009521A (en) * | 2017-12-21 | 2018-05-08 | 广东欧珀移动通信有限公司 | Humanface image matching method, device, terminal and storage medium |
CN108009521B (en) * | 2017-12-21 | 2021-05-04 | Oppo广东移动通信有限公司 | Face image matching method, device, terminal and storage medium |
CN110418064A (en) * | 2019-09-03 | 2019-11-05 | 北京字节跳动网络技术有限公司 | Focusing method, device, electronic equipment and storage medium |
CN110418064B (en) * | 2019-09-03 | 2022-03-04 | 北京字节跳动网络技术有限公司 | Focusing method and device, electronic equipment and storage medium |
CN111083369A (en) * | 2019-12-26 | 2020-04-28 | 上海传英信息技术有限公司 | Face focusing method and device, mobile terminal and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102036008A (en) | Automatic focusing system and method for particular persons | |
US8866931B2 (en) | Apparatus and method for image recognition of facial areas in photographic images from a digital camera | |
EP2634727B1 (en) | Method and portable terminal for correcting gaze direction of user in image | |
US20090059061A1 (en) | Digital photographing apparatus and method using face recognition function | |
CN103945121A (en) | Information processing method and electronic equipment | |
US20150207986A1 (en) | Information processing apparatus, information processing method, and program | |
US8644614B2 (en) | Image processing apparatus, image processing method, and storage medium | |
KR20090045870A (en) | Imaging apparatus, method for controlling the same, and program | |
US20170161553A1 (en) | Method and electronic device for capturing photo | |
CN105303161A (en) | Method and device for shooting multiple people | |
CN103780843A (en) | Image processing method and electronic device | |
JP6157165B2 (en) | Gaze detection device and imaging device | |
JP5640388B2 (en) | Image processing apparatus, imaging apparatus, and image processing program | |
KR20100055946A (en) | Method and apparatus for generating thumbnail of moving picture | |
CN107463903B (en) | Face key point positioning method and device | |
JP5641813B2 (en) | Imaging apparatus and imaging method, image processing apparatus and image processing method | |
JP2003331265A (en) | Eye image imaging device and iris authentication device | |
CN104867112B (en) | Photo processing method and device | |
CN105872352A (en) | Method and device for shooting picture | |
KR101077267B1 (en) | Stenography Input System And Method For Conference Using Face Recognition | |
US8199978B2 (en) | Method for image auto-selection and computer system thereof | |
JP2010198110A (en) | Image processor and electronic still camera | |
CN111726531B (en) | Image shooting method, processing method, device, electronic equipment and storage medium | |
CN114173061A (en) | Multi-mode camera shooting control method and device, computer equipment and storage medium | |
KR101398471B1 (en) | Image processing apparatus and controlling method thereof |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20110427 |