CN103188427A - Image acquisition device capable of simplifying image characteristic value set and control method thereof - Google Patents
Image acquisition device capable of simplifying image characteristic value set and control method thereof Download PDFInfo
- Publication number
- CN103188427A CN103188427A CN2011104542903A CN201110454290A CN103188427A CN 103188427 A CN103188427 A CN 103188427A CN 2011104542903 A CN2011104542903 A CN 2011104542903A CN 201110454290 A CN201110454290 A CN 201110454290A CN 103188427 A CN103188427 A CN 103188427A
- Authority
- CN
- China
- Prior art keywords
- characteristic value
- value group
- image
- characteristic
- simplifying
- 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.)
- Granted
Links
Images
Abstract
The invention discloses an image acquisition device capable of simplifying an image characteristic value set and a control device of the image acquisition device. The image acquisition device comprises a characteristic converting module, a characteristic simplifying module, a template storage module, and an identification module, wherein the characteristic converting module enables an image captured by the image acquisition device to be converted into a characteristic image. The characteristic image comprises a characteristic value group. According to a lookup table, the characteristic simplifying module executes simplification to a program to generate the simplified characteristic valve group. At last, the identification module compares the simplified characteristic valve group with a plurality of templates to be tested and stored in the template storage module to identify an object to be tested in the image. The device and the method can resolve the problem that the image acquisition device with the object identification function in prior technology needs to consume a great number of memories and computing resources so that over high cost is caused.
Description
Technical field
The present invention relates to a kind of image capture unit and control method thereof, particularly picked image can be converted to characteristic image relevant for a kind of, and the characteristic value group in the simplification characteristic image, with the memory body demand that reduces image capture unit and image capture unit and the control method thereof of saving its calculation resources.
Background technology
In recent years, because keen competition on the market, electronic product is par more and more, and function is also more and more powerful.Wherein, the object detecting function has begun to be added in the image capture unit miscellaneous.For example, digital camera (Digital Still Camera), digital code camera (Digital Camera), monitor (Monitor) or even camera cell phone (Camera Phone) etc., and multiple application, for example identification of human face recognition, pedestrian, car plate, vehicle recognition or other object arranged.But owing to often needing to spend a large amount of memory bodys, the image capture unit that possesses the object detecting function stores face characteristic template data such as (Template), more need to consume huge calculation resources, so the cost of image capture unit also can significantly increase.
Above-mentioned situation is obvious especially for the small hand-held camera head, and the small hand-held camera head is owing to considering on the cost, and its resource is also quite limited, particularly is most that obviously the memory body cost of higher-order is high especially in the use of memory body.Therefore, if the object detecting function will be incorporated on the small hand-held camera head, the demand of its memory body is also inevitable significantly to be increased, and directly is reflected on its manufacturing cost, causes competitiveness of product to descend.
Therefore, how to develop a kind of image capture unit, possess the object detecting function, and can be under the prerequisite that does not influence the object detecting accuracy, the hardware cost of saving image capture unit effectively is the problem that institute of the present invention desire solves.
Summary of the invention
Because the problem of above-mentioned known technology, purpose of the present invention is exactly that a kind of image capture unit and control method thereof of simplifying image feature value group is being provided, image capture unit with the tool object identification function that solves known technology need expend a large amount of memory bodys and calculation resources, causes the too high problem of its cost.
For achieving the above object, the present invention takes following technical scheme:
The present invention proposes a kind of image capture unit of simplifying image feature value group, comprises: the Feature Conversion module is characteristic image with video conversion to be measured, and characteristic image comprises the characteristic value cohort; The feature reduction module, according to a threshold value, carry out and simplify procedures, from the characteristic value cohort, the characteristic value group deletion of look-up table will be present in, each deleted characteristic value group is then one pre-conditioned and exist the characteristic value group of this look-up table to replace to meet, and the constant but kind of characteristic value cohort sum is reduced, and produces the characteristic value cohort after simplifying; Template stores module, stores several determinand templates (Template); And the identification module, the characteristic value cohort after this simplification and several determinand templates are compared, with the determinand in the identification image to be measured.
According to purpose of the present invention, a kind of control method of simplifying image feature value group is proposed again, comprise the following step: utilize the Feature Conversion module is characteristic image with video conversion to be measured, characteristic image comprises the characteristic value cohort; By the feature reduction module according to a look-up table, carry out and simplify procedures, from the characteristic value cohort, the characteristic value group deletion of look-up table will be present in, each deleted characteristic value group then replaces to meet characteristic value group pre-conditioned and that be present in look-up table, constant but the kind of characteristic value cohort sum is reduced, produce the characteristic value cohort after simplifying; Store several determinand templates of module stores by template; And characteristic value cohort and several determinand templates after will simplifying by the identification module compare, with the determinand in the identification image to be measured.
Preferably, pre-conditionedly decided by otherness, each the deleted value of levying group needs to replace with the characteristic value group with its otherness minimum.
Preferably, look-up table is set up with the characteristic value group that the complexity of characteristic value group is lower than a complexity (Entropy) threshold value.
Preferably, after execution is simplified procedures, in the characteristic value cohort after the simplification each characteristic value group all can find with its transposition after corresponding characteristic value group.
Preferably, after execution is simplified procedures, in the characteristic value cohort after the simplification each characteristic value group all can find with its mirror after corresponding characteristic value group.
Preferably, several determinand templates bases are present in the characteristic value group of look-up table, and via advancing algorithm, SVMs algorithm or Principal Component Analysis Method+linearity to have the calculation of identification force analysis to set up.
The invention has the beneficial effects as follows: from the above, according to image capture unit and the control method thereof of simplifying image feature value group of the present invention, it can have one or more following advantage:
(1) this can simplify the image capture unit of image feature value group and the kind that control method can be simplified the characteristic value group in the characteristic image thereof, therefore can significantly reduce the memory body demand of image capture unit, and then reduce its manufacturing cost.
(2) this can simplify the image capture unit of image feature value group and the kind that control method can reduce the characteristic value group in the characteristic image thereof, and therefore the calculation resources that also can significantly save image capture unit improves its efficient.
(3) this can to simplify the image capture unit of image feature value group and control method thereof be that characteristic value group with the otherness minimum replaces each deleted characteristic value group, therefore rather than directly deletion can not reduce the accuracy of object detecting.
(4) though this can simplify the image capture unit of image feature value group and control method is simplified the kind of the characteristic value group in the characteristic image, but each characteristic value group in the characteristic value cohort after simplifying all can find with its transposition after or corresponding characteristic value group behind the mirror, the characteristic that therefore can keep former characteristic value cohort is to carry out relevant application.
Description of drawings
Fig. 1 is the calcspar of first embodiment of the image capture unit of simplifying image feature value group of the present invention.
Fig. 2 is the flow chart of first embodiment of the image capture unit of simplifying image feature value group of the present invention.
Fig. 3 is the calcspar of second embodiment of the image capture unit of simplifying image feature value group of the present invention.
Fig. 4 is one of the schematic diagram of an embodiment of characteristic value group.
Fig. 5 be the characteristic value group an embodiment schematic diagram two.
Fig. 6 is the flow chart of second embodiment of the image capture unit of simplifying image feature value group of the present invention.
Fig. 7 is the schematic diagram of the 3rd embodiment of the image capture unit of simplifying image feature value group of the present invention.
Fig. 8 is the schematic diagram of the 4th embodiment of the image capture unit of simplifying image feature value group of the present invention.
Fig. 9 is the flow chart of simplifying the control method of image feature value group of the present invention.
Drawing reference numeral: 1: image capture unit; 3,7,8: digital camera; 10,30,70,80: image; 11,31: the Feature Conversion module; 111,311: characteristic image; 1111,3111,81: the characteristic value cohort; 12,32: the feature reduction module; 121,321: the characteristic value cohort after the simplification; 13: template stores module; 131: several determinand templates; 14: the identification module; 141: identification result; 33: face template stores module; 331: several face templates; 34: the human face recognition module; 341,82: limitting casing; 41,51: validity feature value group; 42,52: invalid characteristic value group; 71: form; 710,711: square; S21 ~ S24, S61 ~ S65, S91 ~ S94: steps flow chart.
Embodiment
Hereinafter with reference to correlative type, illustrate that according to the image capture unit of image feature value group and the embodiment of control method thereof of simplifying of the present invention be convenient to understand for making, the similar elements among the following embodiment illustrates with identical symbology.
See also Fig. 1, be the calcspar of first embodiment of the image capture unit of simplifying image feature value group of the present invention.As shown in the figure, image capture unit 1 of the present invention comprises Feature Conversion module 11, feature reduction module 12, template storage module 13 and identification module 14.Feature Conversion module 11 can be converted to characteristic image 111 with image capture unit 1 picked image 10, and this characteristic image 111 comprises a characteristic value cohort 1111 of being made up of several characteristic value groups.
It is noted that this pre-conditioned can deciding with the otherness between the characteristic value group.Each deleted characteristic value group can utilize the characteristic value group with its otherness minimum to replace, rather than directly deletion is not present in the characteristic value group of look-up table.The benefit of this mode is that the expression ability of deleted characteristic value group can be replaced by the characteristic value group with its otherness minimum, therefore still can be retained, and can reduce the kind of characteristic value group in the characteristic value cohort 1111 simultaneously again.So, namely can under the situation of the object detecting accuracy that does not influence image capture unit, reduce the memory body demand of image capture unit 1 effectively.
At last, the characteristic value cohort 121 after 14 of identification modules are just simplified is compared with several determinand templates 131 that are stored in the template storage module 13, to produce an identification result 141, gets final product the determinand in the identification image 10.
What deserves to be mentioned is, though the present invention has deleted the characteristic value group that is not present in look-up table, each characteristic value group after simplifying still can find with its transposition after or corresponding characteristic value group behind the mirror.That is to say that the characteristic value cohort 121 after the simplification still can keep the characteristic of former characteristic value group 1111, therefore also can utilize this characteristic, carry out the application on the various object detectings.
Certainly, being stored in template, to store several determinand templates 131 in the module 13 also be that the characteristic value group that meets the condition that threshold value sets is set up, so could with simplification after characteristic value cohort 121 compare, with the identification object.
See also Fig. 2, be the flow chart of first embodiment of the image capture unit of simplifying image feature value group of the present invention.
In step S21, capture an image by image capture unit.
In step S22, be characteristic image by the Feature Conversion module with this video conversion, this characteristic image comprises a characteristic value cohort.
In step S23, simplify procedures via the execution of feature reduction module, delete in this characteristic value cohort, the characteristic value group that is not present in look-up table, each deleted characteristic value group is then to be present in this look-up table and to replace with the characteristic value group of its otherness minimum, to produce the characteristic value cohort after simplifying.
In step S24, compare with regard to the characteristic value cohort after simplifying and several determinand templates that are stored in the template storage module by the identification module, with the determinand in the identification image.
Image capture unit and the control method thereof of simplifying image feature value group of the present invention also can be applicable to any camera head with object identification function.For example, digital camera (Digital Still Camera), digital code camera (Digital Camera), monitor (Monitor) or even according to mobile phone (Camera Phone) etc., with the identification of executor's face, pedestrian, car plate, vehicle or other various items.Following embodiment is example with digital camera identification people face, but the present invention is not as limit.
See also Fig. 3, be the calcspar of second embodiment of the image capture unit of simplifying image feature value group of the present invention.As shown in the figure, digital camera 3 of the present invention comprises Feature Conversion module 31, feature reduction module 32, face template storage module 33 and human face recognition module 34.Feature Conversion module 31 can be converted to characteristic image 311 with digital camera 3 picked image 30, and this characteristic image 311 comprises a characteristic value cohort 3111 of being made up of several characteristic value groups.
Feature reduction module 32 can be simplified procedures according to being carried out by a look-up table, this look-up table is to decide according to the complexity of each characteristic value group (Entropy) threshold value, complexity (entropy) is higher than the characteristic value group of this threshold value and is then deleted, the characteristic value group that complexity is lower than this threshold value then keeps, each deleted characteristic value group then replaces with the characteristic value group with its otherness minimum, to produce the characteristic value cohort 321 after simplifying.
In addition, for keeping characteristics value cohort 3111 characteristic originally, feature reduction module 32 more can keep in the characteristic value cohort 321 after the simplification with each characteristic value group transposition after or corresponding characteristic value group behind the mirror, to carry out various human face recognition functions.
At last, characteristic value cohort 321 after 34 of human face recognition modules are just simplified is compared with several face templates 331 that are stored in the face template storage module 33, produce a limitting casing 341, people's face in can identification image 30 is finished the work that the user can utilize digital camera 3 to focus and take after the identification.
In addition, several face templates 331 are to be present in look-up table, it is the lower characteristic value group of entropy, and via advancing algorithm (Boosting), SVMs algorithm (Support Vector Mechine, SVM) or Principal Component Analysis Method+linearity have the identification force analysis (Principle Component Analysis+Linear Discriminant Analysis, PCA+LDA) calculation and get.
What deserves to be mentioned is, when the image capture unit of known technology is characteristic image with video conversion, the utilization rate of each characteristic value group that this characteristic image comprises is not quite similar, and wherein some often is used, and wherein has another partly then seldom to use.For example, resolution is 1920 * 1080 image, and wherein each point all comprises a characteristic value group, comprises 500 kinds of different characteristic value groups altogether, often is used and wherein only there are 250 kinds, then seldom uses for other 250 kinds.The frequency that the characteristic value group is used is then relevant with its complexity or entropy (Entropy).Generally speaking, the lower characteristic value group of entropy is than the tool systematicness, and expression power is stronger, and frequency of utilization is also higher.
And concept proposed by the invention, utilize to analyze the entropy of each characteristic value group just, setting up a look-up table, and will be wherein than the tool systematicness according to this look-up table, expression power is kept than strong and the higher characteristic value group of frequency of utilization, and the higher characteristic value group of deletion complexity.Each deleted characteristic value group is then with minimum with its otherness and be present in this look-up table and the lower characteristic value group of entropy is replaced.So, not only can keep the expression ability of each deleted characteristic value group, also can reduce the quantity of characteristic value group kind simultaneously.
See also Fig. 4, be one of schematic diagram of an embodiment of characteristic value group.By it is evident that among the figure, the arrangement of each characteristic value is than the tool systematicness in the validity feature value group 41, and the arrangement of each characteristic value is then comparatively at random in the invalid characteristic value group 42.This invalid characteristic value group 42 is because its arrangement is comparatively at random, so its expression power is relatively poor, and frequency of utilization is low.And this threshold value of feature reduction module of the present invention is analyzed the entropy of each characteristic value group, and the invalid characteristic value group 42 that entropy is not inconsistent unification complexity threshold value is deleted.
See also the 5th figure, be two of the schematic diagram of an embodiment of characteristic value group.As shown in FIG., invalid characteristic value group 52 its arrangements are comparatively at random, and entropy is higher.After invalid characteristic value group 52 was deleted, feature reduction module of the present invention then utilized with its otherness lower, and then the preferable validity feature value group 51 of systematicness replaces these invalid characteristic value groups 52.
By can significantly finding out among the figure, only there is the difference of a characteristic value between validity feature value group 51 and the invalid characteristic value group 52.Feature reduction module of the present invention then can utilize the entropy of analyzing each characteristic value, and the validity feature value group of finding out with deleted invalid characteristic value group 52 otherness minimums 51 replaces this invalid characteristic value group 52.And validity feature value group 51 is close with the expression power of invalid characteristic value 52, therefore can replace invalid characteristic value 52 fully.Therefore, the characteristic value cohort after feature reduction module of the present invention is simplified can not lost its expression power originally, and image capture unit still can be kept originally object identification ability accurately.In addition, the expression mode difference of the characteristic value group of different algorithms for example has different arrow number etc., but all can use method proposed by the invention to be simplified.
What deserves to be mentioned is, because the quantity of characteristic value group kind is breezy relevant with the memory body demand of image capture unit, so utilize characteristic value cohort after the method simplification of the present invention can significantly reduce the memory body demand of image capture unit.On the other hand, because the minimizing of characteristic value group kind, the calculation resources that also can save image capture unit effectively makes image capture unit more efficient.And via the test of reality, under the prerequisite that does not influence the detecting quality, the memory internal body of digital camera (Internal RAM) has reduced 34%, the size of ROM (ROM Size) has reduced 36%, so the present invention can reduce the size of the required memory body of image capture unit really.
See also Fig. 6, be the flow chart of second embodiment of the image capture unit of simplifying image feature value group of the present invention.
In step S61, capture an image by digital camera.
In step S62, be characteristic image by the Feature Conversion module with this video conversion, this characteristic image comprises a characteristic value cohort.
In step S63, by the feature reduction module according to a look-up table, characteristic value group with the high complexity threshold value of deletion entropy, keep the characteristic value group that entropy is lower than this complexity threshold value, each deleted characteristic value group then replaces with the characteristic value group with its otherness minimum, to produce the characteristic value cohort after simplifying.
In step S64, then compare with regard to the characteristic value cohort after simplifying and several face templates that are stored in the face template storage module by the human face recognition module, with the people's face in the identification image.
In step S65, the work that utilizes digital camera to focus and take.
See also Fig. 7, be the schematic diagram of the 3rd embodiment of the image capture unit of simplifying image feature value group of the present invention.Present embodiment is example with class Hull feature algorithm (Haar-Like Features).As shown in FIG., the Haar-Like algorithm is to utilize 7 picked image 70 of a form 71 scanning digital cameras, and the square that form 71 is divided into more than two or two is added and subtracted mutually again, is converted to the feature signal again.As shown in FIG., square 710 is positioned at the position of people's face forehead, so square 710 captured is the more shallow signal of color, and square 711 is positioned at the position of people's face eyes, so square 711 acquisitions are to the darker signal of color.And the signal that square 711 is captured and square 710 subtract each other and can learn that the signal that both capture has bigger difference, can judge that therefore this form 71 probably namely is the position at people's face place.
It is lower that above-mentioned mode belongs to entropy, than tool systematicness and the comparison method that more often is used, certainly also have in form 71, the comparison method that while adds and subtracts mutually with a plurality of squares that are positioned at diverse location, but this method is not more had a systematicness, also less being used can utilize the lower comparison method of entropy to replace.By method proposed by the invention, can pass through look-up table, execution is simplified procedures, and it is higher to delete some entropy, and the comparison method that more seldom is used is like this then can reduce the memory body demand of digital camera 7 effectively.
See also Fig. 8, be the schematic diagram of the 4th embodiment of the image capture unit of simplifying image feature value group of the present invention.(Histogram of Oriented Gradient HOG) illustrates concept of the present invention to present embodiment with direction gradient histogram algorithm.
As shown in the figure, the user uses digital camera 8 to take the personage, and the Feature Conversion module of digital camera 8 utilizes the HOG algorithm image 80 to be converted to the characteristic value cohort of being made up of several characteristic value groups 81.The feature reduction module then can be according to a look-up table, and execution is simplified procedures, and the characteristic value group that entropy wherein is higher than this entropy threshold value is deleted, and uses the characteristic value group of its otherness minimum to replace.The human face recognition module is then simplified this back result and is stored the module comparison with being stored in face template in advance, can produce limitting casing 82.After finishing the work of human face detection again, the user just can focus and work such as shooting.
Although it is aforementioned in the process of the explanation image capture unit of simplifying image feature value group of the present invention, the concept of simplifying the control method of image feature value group of the present invention also is described simultaneously, but for ask clear for the purpose of, below still illustrate flow chart in addition and describe in detail.
See also Fig. 9, be the flow chart of simplifying the control method of image feature value group of the present invention.
In step S91, utilize the Feature Conversion module that video conversion to be measured is characteristic image, this characteristic image comprises a characteristic value cohort.
In step S92, by the feature reduction module according to a threshold value, carry out and simplify procedures, from then in the characteristic value cohort, the characteristic value group deletion of look-up table will be present in, each deleted characteristic value group then replaces to meet characteristic value group pre-conditioned and that be present in this look-up table, and the constant but kind of characteristic value cohort sum is reduced, and produces the characteristic value cohort after simplifying.
In step S93, store several determinand templates of module stores by template.
In step S94, characteristic value cohort and several determinand templates after will simplifying by the identification module are compared, with the determinand in the identification image to be measured.
The detailed description of the control method of simplifying image feature value group of the present invention and execution mode were described when the front narration image capture unit of simplifying image feature value group of the present invention, at this for just repeated description no longer of schematic illustration.
In sum, therefore the image capture unit of image feature value group and the kind that control method can be simplified the characteristic value group in the characteristic image thereof simplified of the present invention can significantly reduce the memory body demand of image capture unit, and then reduce its manufacturing cost.The present invention can reduce the kind of the characteristic value group in the characteristic image, and therefore the calculation resources that also can significantly save image capture unit improves its efficient.The present invention is that the characteristic value group with the otherness minimum replaces each deleted characteristic value group, rather than directly deletion, therefore can not reduce the accuracy of object detecting.Each characteristic value group in the characteristic value cohort after method of the present invention is simplified all can find with its transposition after or corresponding characteristic value group behind the mirror, therefore can keep the characteristic of former characteristic value cohort, to carry out relevant application.Therefore, the present invention can improve the shortcoming of known technology really.
The above only is illustrative, but not is restricted person.Anyly do not break away from spirit of the present invention and category, and to its equivalent modifications of carrying out or change, all should be contained in the accompanying Claim book.
Claims (12)
1. the image capture unit that can simplify image feature value group is characterized in that it comprises:
One Feature Conversion module is a characteristic image with a video conversion to be measured, and described characteristic image comprises a characteristic value cohort;
One feature reduction module, according to a look-up table, simplify procedures, from described characteristic value cohort, the characteristic value group deletion of described look-up table will be present in, each deleted characteristic value group then replaces to meet a characteristic value group pre-conditioned and that be present in described look-up table, and the constant but kind of described characteristic value cohort sum is reduced, and produces a characteristic value cohort after simplifying;
One template stores module, stores several determinand templates; And
One identification module is compared the characteristic value cohort after the described simplification and described several determinand templates, with the determinand in the described image to be measured of identification.
2. the image capture unit of simplifying image feature value group according to claim 1 is characterized in that, described pre-conditionedly decided by otherness, and described each deleted characteristic value group needs to replace with the characteristic value group with its otherness minimum.
3. the image capture unit of simplifying image feature value group according to claim 2 is characterized in that, described look-up table is set up with the characteristic value group that the complexity of characteristic value group is lower than a complexity threshold value.
4. the image capture unit of simplifying image feature value group according to claim 3, it is characterized in that, after carrying out described simplifying procedures, in the characteristic value cohort after the described simplification each characteristic value group all can find with its transposition after corresponding characteristic value group.
5. the image capture unit of simplifying image feature value group according to claim 3, it is characterized in that, after carrying out described simplifying procedures, in the characteristic value cohort after the described simplification each characteristic value group all can find with its mirror after corresponding characteristic value group.
6. the image capture unit of simplifying image feature value group according to claim 1, it is characterized in that, described several determinand template utilizations are present in the characteristic value group of described look-up table, and via advancing algorithm, SVMs algorithm or Principal Component Analysis Method+linearity to have the calculation of identification force analysis to set up.
7. the control method that can simplify image feature value group is characterized in that it comprises the following step:
Utilizing a Feature Conversion module is a characteristic image with a video conversion to be measured, and described characteristic image comprises a characteristic value cohort;
By a feature reduction module according to a look-up table, simplify procedures, from described characteristic value cohort, the characteristic value group deletion of described look-up table will be present in, each deleted characteristic value group then replaces to meet a characteristic value group pre-conditioned and that be present in described look-up table, constant but the kind of described characteristic value cohort sum is reduced, produce a characteristic value cohort after simplifying;
Store several determinand templates of module stores by a template; And
Compare by identification module characteristic value cohort and described several determinand templates after with described simplification, with the determinand in the described image to be measured of identification.
8. the control method of simplifying image feature value group according to claim 7 is characterized in that, described pre-conditionedly decided by otherness, and each the deleted value of levying group needs to replace with the characteristic value group with its otherness minimum.
9. the control method of simplifying image feature value group according to claim 8 is characterized in that, described look-up table is set up with the characteristic value group that the complexity of characteristic value group is lower than a complexity threshold value.
10. the control method of simplifying image feature value group according to claim 9 is characterized in that, after carrying out described simplifying procedures, in the characteristic value cohort after the described simplification each characteristic value group all can find with its transposition after corresponding characteristic value group.
11. the control method of simplifying image feature value group according to claim 9 is characterized in that, after carrying out described simplifying procedures, in the characteristic value cohort after the described simplification each characteristic value group all can find with its mirror after corresponding characteristic value group.
12. the control method of simplifying image feature value group according to claim 7, it is characterized in that, described several determinand templates are present in the characteristic value group of described look-up table, and through and via advancing algorithm, SVMs algorithm or Principal Component Analysis Method+linearity to have the calculation of identification force analysis to set up.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110454290.3A CN103188427B (en) | 2011-12-30 | 2011-12-30 | Image capture unit and the control method thereof of image feature value group can be simplified |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110454290.3A CN103188427B (en) | 2011-12-30 | 2011-12-30 | Image capture unit and the control method thereof of image feature value group can be simplified |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103188427A true CN103188427A (en) | 2013-07-03 |
| CN103188427B CN103188427B (en) | 2016-08-10 |
Family
ID=48679379
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110454290.3A Active CN103188427B (en) | 2011-12-30 | 2011-12-30 | Image capture unit and the control method thereof of image feature value group can be simplified |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103188427B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015051697A1 (en) * | 2013-10-10 | 2015-04-16 | Harman International Industries, Incorporated | Image processing |
| CN106922192A (en) * | 2014-12-10 | 2017-07-04 | 英特尔公司 | Using the type of face detection method and device of look-up table |
| TWI681342B (en) * | 2018-12-11 | 2020-01-01 | 瑞昱半導體股份有限公司 | Object position determination circuit |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5982912A (en) * | 1996-03-18 | 1999-11-09 | Kabushiki Kaisha Toshiba | Person identification apparatus and method using concentric templates and feature point candidates |
| CN1963918A (en) * | 2005-11-11 | 2007-05-16 | 株式会社东芝 | Compress of speaker cyclostyle, combination apparatus and method and authentication of speaker |
| US7372981B2 (en) * | 2003-04-28 | 2008-05-13 | Industrial Technology Research Institute | Statistical facial feature extraction method |
| CN101615248A (en) * | 2009-04-21 | 2009-12-30 | 华为技术有限公司 | Age estimation method, equipment and face identification system |
| CN102110225A (en) * | 2009-12-28 | 2011-06-29 | 比亚迪股份有限公司 | Outdoor face identifying method and system |
-
2011
- 2011-12-30 CN CN201110454290.3A patent/CN103188427B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5982912A (en) * | 1996-03-18 | 1999-11-09 | Kabushiki Kaisha Toshiba | Person identification apparatus and method using concentric templates and feature point candidates |
| US7372981B2 (en) * | 2003-04-28 | 2008-05-13 | Industrial Technology Research Institute | Statistical facial feature extraction method |
| CN1963918A (en) * | 2005-11-11 | 2007-05-16 | 株式会社东芝 | Compress of speaker cyclostyle, combination apparatus and method and authentication of speaker |
| CN101615248A (en) * | 2009-04-21 | 2009-12-30 | 华为技术有限公司 | Age estimation method, equipment and face identification system |
| CN102110225A (en) * | 2009-12-28 | 2011-06-29 | 比亚迪股份有限公司 | Outdoor face identifying method and system |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015051697A1 (en) * | 2013-10-10 | 2015-04-16 | Harman International Industries, Incorporated | Image processing |
| CN106922192A (en) * | 2014-12-10 | 2017-07-04 | 英特尔公司 | Using the type of face detection method and device of look-up table |
| CN106922192B (en) * | 2014-12-10 | 2021-08-24 | 英特尔公司 | Face detection method and apparatus using lookup table |
| TWI681342B (en) * | 2018-12-11 | 2020-01-01 | 瑞昱半導體股份有限公司 | Object position determination circuit |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103188427B (en) | 2016-08-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106909888B (en) | Face key point tracking system and method applied to mobile equipment terminal | |
| CN105740775B (en) | Three-dimensional face living body identification method and device | |
| KR102255215B1 (en) | Method of detecting object in image and image processing device | |
| CN105426829B (en) | Video classification methods and device based on facial image | |
| CN111160269A (en) | Face key point detection method and device | |
| WO2021248733A1 (en) | Live face detection system applying two-branch three-dimensional convolutional model, terminal and storage medium | |
| EP2608108A1 (en) | Face feature vector construction | |
| US8421877B2 (en) | Feature data compression device, multi-directional face detection system and detection method thereof | |
| CN103810490A (en) | Method and device for confirming attribute of face image | |
| Vazquez-Fernandez et al. | Built-in face recognition for smart photo sharing in mobile devices | |
| US9323989B2 (en) | Tracking device | |
| CN104866806A (en) | Self-timer system and method with face positioning auxiliary function | |
| KR20140019950A (en) | Method for generating 3d coordinate using finger image from mono camera in terminal and mobile terminal for generating 3d coordinate using finger image from mono camera | |
| HN et al. | Human Facial Expression Recognition from static images using shape and appearance feature | |
| CN103188427A (en) | Image acquisition device capable of simplifying image characteristic value set and control method thereof | |
| CN108154130B (en) | Target image detection method and device, storage medium and robot | |
| CN105528616A (en) | Face recognition method and device | |
| CN110688950A (en) | Face living body detection method and device based on depth information | |
| CN103473807A (en) | 3D model transformation system and method | |
| CN109711287A (en) | Face acquisition method and Related product | |
| US8908974B2 (en) | Image capturing device capable of simplifying characteristic value sets of captured images and control method thereof | |
| Luo et al. | Dynamic face recognition system in recognizing facial expressions for service robotics | |
| Jin et al. | An FPGA-based parallel hardware architecture for real-time face detection using a face certainty map | |
| CN104050457A (en) | Human face gender identification method based on small sample training library | |
| US11893791B2 (en) | Pre-processing image frames based on camera 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 | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20161222 Address after: China Taiwan Hsinchu Science Park Hsinchu City Road No. 12 5 floor Patentee after: Glomerocryst semiconductor limited company Address before: Hsinchu City, Taiwan, China Patentee before: Huajing Science & Technology Co., Ltd. |
|
| TR01 | Transfer of patent right |