CN101557487B - Hard disk recorder with human face image capturing function and method for capturing a human face image - Google Patents
Hard disk recorder with human face image capturing function and method for capturing a human face image Download PDFInfo
- Publication number
- CN101557487B CN101557487B CN200910050916.7A CN200910050916A CN101557487B CN 101557487 B CN101557487 B CN 101557487B CN 200910050916 A CN200910050916 A CN 200910050916A CN 101557487 B CN101557487 B CN 101557487B
- Authority
- CN
- China
- Prior art keywords
- facial image
- image
- similarity
- unit
- real
- 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.)
- Expired - Fee Related
Links
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a hard disk recorder with human face image capturing function and a method for capturing a human face image. The hard disk recorder comprises a human face image capturing system. The human face image capturing system comprises an image acquisition unit, a first storage unit, a human face real-time comparison unit, an image real-time screening unit, a second storage unit anda human face afterward comparison unit, wherein the image real-time screening unit screens a human face image in the first storage unit according to the similarity calculated by the human face real-t ime comparison unit, and keeps only one human face posture image for the same person. The human face afterward comparison unit selects human face images which have similarities, with human face imagesset in the second storage unit, smaller than the set similarity, and stores the human face images into the second storage unit. The hard disk recorder and the method directly complete acquisition wor k of human face photographs, reduce workload of manually playing back, screening and capturing.
Description
Technical field
The present invention relates to a kind of DVR, relate in particular to a kind of DVR that facial image captures function that has, relate to a kind of facial image grasping means of above-mentioned DVR simultaneously.
Background technology
In reality, some crucial places need to be monitored and carry out image storage, conventionally can select the DVR storage of recording a video, yet traditional DVR exist following problem:
1, the generation per hour of conventional hard video tape recorder is about the video of 850MB, and the video of storage in generally 3 months is approximately 1793GB.It is large that the video record storing takies hard drive space; Often need hard disc data derivation or increase new hard disk;
2, the utilance of video recording is low, often needs artificial video playback, again intercepts photo and uses; In waste hard drive space, need a large amount of hand labors;
3, on the market the video compression ratio of each manufacturer's DVR each is different, the quality that causes recording a video is different, directly causes the key photo of intercepting to occur fuzzy, distortion phenomenon.
As Chinese patent CN02115324.8 has disclosed a kind of IP-based real-time multimedia intelligent monitoring method and system of ensureing; Described method and system comprise vision signal, the speech talkback signal that front end watch-dog is collected, or the digital signal that button and sensor device detect sends domain controller to; Domain controller by digital signal sample, digitlization and compression process and be encapsulated in IP message frame; Domain controller sends by an IP tunnel data that are encapsulated in IP message frame to Surveillance center under the control of network management center connected to the network.Still there are the problems referred to above in this scheme.
Summary of the invention
Technical problem to be solved by this invention is: a kind of DVR that facial image captures function that has is provided, it can be monitored in real time and captures monitored people's facial photo and store, its have the storage picture of collection take hard drive space little, directly utilize, picture is without the higher feature of compression quality repeatedly.
In addition, the present invention also provides the facial image grasping means of above-mentioned DVR.
For solving the problems of the technologies described above, the present invention adopts following technical scheme:
There is the DVR that facial image captures function, for storing the video information of obtaining from image unit; It is characterized in that, it comprises facial image grasping system, and in order to obtain the facial image obtaining from image unit, this facial image grasping system comprises:
Image acquisition units, is connected with described image unit, the image information of obtaining in order to the described image unit of real-time crawl, and intercepting facial image wherein.
Preferably, described facial image grasping system also comprises:
The first memory cell, in order to the facial image of memory image collecting unit Real-time Collection;
The real-time comparing unit of people's face, in order to the facial image of image acquisition units Real-time Collection is compared between two, calculates its similarity; Two images that are greater than setting similarity are considered to represent same person face;
Image screens unit in real time, in order to the similarity calculating according to the real-time comparing unit of people's face, screens the facial image in described the first memory cell, and the human face posture image of same person only retains one;
The second memory cell, in order to store satisfactory facial image;
People's face is comparing unit afterwards, in order to image is screened in real time to the setting facial image having in facial image after unit screening and described the second memory cell, compares, and calculates its similarity; The similarity of the facial image of setting in selection and described the second memory cell is all less than the facial image of setting similarity, is stored in described the second memory cell.
As a preferred embodiment of the present invention, described the first memory cell is buffer memory.
As a preferred embodiment of the present invention, described image acquisition units intercepting facial image is to buffer memory, and every setting-up time T1, the real-time comparing unit of people's face is compared above-mentioned facial image, calculates its similarity; If the similarity of two facial images is greater than setting similarity, delete a wherein facial image.
As a preferred embodiment of the present invention, described image acquisition units intercepting facial image is to buffer memory, and the facial image of number is set in every intercepting, and the real-time comparing unit of people's face is compared above-mentioned facial image, calculates its similarity; If the similarity of two facial images is greater than setting similarity, delete a wherein facial image.
As a preferred embodiment of the present invention, described image acquisition units intercepting facial image is to buffer memory, and every the facial image of setting-up time T1 or every intercepting setting number, the real-time comparing unit of people's face is compared above-mentioned facial image, calculates its similarity; If the similarity of two facial images is greater than setting similarity, delete a wherein facial image; Then, the real-time comparing unit of people's face compares the facial image in remaining facial image and memory cell in buffer memory.
As a preferred embodiment of the present invention, the face images in the second memory cell is compared in described face alignment unit, or compares the facial image in setting-up time section wherein.
As a preferred embodiment of the present invention, described system also comprises data processing unit, in order to process the facial image of described intercepting.
As a preferred embodiment of the present invention, described setting similarity is 75%, and the similarity of two images is more than or equal to 75% and is considered to represent same person face.
As a preferred embodiment of the present invention, described system also comprises time record cell, the time occurring in order to record each personnel of each facial image representative; When repeated detection occurs to same personnel, record respectively this people's time of occurrence.
A facial image grasping means for above-mentioned DVR, the method comprises the steps:
Image unit obtains image information;
Image acquisition units captures the image information that described image unit obtains in real time, and intercepting facial image wherein;
Face alignment unit compares the setting facial image having in the facial image of intercepting and memory cell, calculates its similarity; Two images that are greater than setting similarity are considered to represent same person face;
Optical sieving unit screens the facial image of described intercepting, and the facial image of intercepting and the setting facial image of cell stores are compared; The similarity of the facial image of setting in selection and memory cell is all less than facial image setting similarity, new intercepting and is stored in described memory cell.
As a preferred embodiment of the present invention, described the first memory cell is buffer memory.
As a preferred embodiment of the present invention, described image acquisition units intercepting facial image is to buffer memory, and every the facial image of setting-up time T1 or every intercepting setting number, above-mentioned facial image is compared in face alignment unit, calculates its similarity; If the similarity of two facial images is greater than setting similarity, delete a wherein facial image; Then, face alignment unit compares the facial image in remaining facial image and memory cell in buffer memory.
As a preferred embodiment of the present invention, the face images in the second memory cell is compared in described face alignment unit, or the facial image in comparison setting-up time section.
Beneficial effect of the present invention is: facial image grasping system and method that the present invention proposes, by capturing photo, guaranteed on the one hand the high-quality of photo, the misoperation that avoided that the monitored picture that in the past brought because of low video compression format is fuzzy, brings when distortion and manual operation and again compression bring quality to reduce; Directly complete on the other hand portrait photo's collecting work, reduced the workload of artificial playback, screening and crawl.
Accompanying drawing explanation
Fig. 1 is the composition schematic diagram of inventor's face image-capture system.
Embodiment
Below in conjunction with accompanying drawing, describe the preferred embodiments of the present invention in detail.
Embodiment mono-
Refer to Fig. 1, the present invention has disclosed a kind of DVR that facial image captures function that has, it comprises facial image grasping system, and this facial image grasping system comprises: image unit 11, image acquisition units 12, buffer unit 13, the real-time comparing unit 14 of people's face, image screen unit 15, data processing unit 16, the second memory cell 17, people's face comparing unit 18 afterwards in real time.Each unit is below described respectively.
[image unit]
Image unit 11 is in order to obtain image information; Can be the first-class device of monitoring camera, it is originated for image acquisition units 12 provides image.
[image acquisition units]
Image acquisition units 12 is connected with described image unit 11, the image information of obtaining in order to the described image unit 11 of real-time crawl, and intercepting facial image wherein.
Described image acquisition units 12 is kept in the facial image of intercepting to be stored in buffer unit 13.
In the present embodiment, image acquisition units 12 is a DSP microchip processor, can carry out monitoring in real time and capture the functions such as photo, telecommunication network transmission.Human image collecting equipment directly carries out timing to camera collection real-time video and intercepts human face photo.In image acquisition units 12 is per second, can intercept the portrait photo of 10 left and right, the mean size of intercepting photo is about 25KB (96*128), the photo of the raw about 250KB of common property per second.
[buffer unit]
Buffer unit 13 is in order to the facial image of memory image collecting unit 12 Real-time Collections; In the present embodiment, described image acquisition units 12 intercepting facial images are to buffer unit 13, every the facial image of setting-up time T1 or every intercepting setting number.
The real-time comparing unit 14 of people's face is compared the facial image of image acquisition units 12 Real-time Collections between two, calculates its similarity; Two images that are greater than setting similarity are considered to represent same person face.In the present embodiment, described setting similarity is 75%, and the similarity of two images is more than or equal to 75% and is considered to represent same person face.
Image screens in real time the similarity that unit 15 calculates according to the real-time comparing unit of people's face and screens the facial image in described buffer unit 13, and the human face posture image of same person only retains one.Then, people's face afterwards comparing unit 18 facial image in remaining facial image in buffer unit 13 and the second memory cell 17 is compared.
[the second memory cell]
The second memory cell 17 is in order to store satisfactory facial image, and the facial image of these the second memory cell 17 storages is as the comparison foundation of the facial image of image acquisition units 12 interceptings afterwards.And buffer unit 13 is as temporary storage part, the facial image in the time capable of setting storage.
[people's face is comparing unit afterwards]
People's face afterwards comparing unit 18 compares in order to image is screened in real time to the setting facial image having in facial image after unit 15 screening and described the second memory cell 17, calculates its similarity; The similarity of the facial image of setting in selection and described the second memory cell is all less than the facial image of setting similarity, is stored in described the second memory cell 17.In the present embodiment, described setting similarity is 75%, and the similarity of two images is more than or equal to 75% and is considered to represent same person face.
Described people's face face images that comparing unit 18 is compared in the second memory cell 17 afterwards, or the facial image of comparison setting regions.
[data processing unit]
Before the facial image of sectional drawing is stored in to described memory cell 17, facial image can also be processed facial image through data processing unit 16.As, 16 pairs of facial images of data processing unit compress processing; In the present embodiment, in the situation that not losing photographic quality, photo is unified to the standard photographs of boil down to 5KB size and stored.Photo storage adopts own document handling system, and comparison film data are carried out index stores, supports the inquiry afterwards of comparison film data.
More than introduced the formation of system of the present invention, below introduced the facial image grasping means of above-mentioned facial image grasping system, the method comprises the steps:
Steps A, image unit obtain image information;
Step B, image acquisition units capture the image information that described image unit obtains in real time, and intercepting facial image wherein;
Step C, the real-time comparing unit of people's face are compared the facial image of image acquisition units Real-time Collection between two, calculate its similarity; Two images that are greater than setting similarity are considered to represent same person face;
Step D, image screen in real time the similarity that unit calculates according to the real-time comparing unit of people's face and screen the facial image in described the first memory cell, and the human face posture image of same person only retains one;
Step e, optical sieving unit screen the facial image of described intercepting, and the facial image of intercepting and the setting facial image of cell stores are compared; The similarity of the facial image of setting in selection and memory cell is all less than facial image setting similarity, new intercepting and is stored in described memory cell.
Step F, people's face afterwards comparing unit screen in real time image the setting facial image having in facial image after unit screening and described the second memory cell and compare, and calculate its similarity; The similarity of the facial image of setting in selection and described the second memory cell is all less than the facial image of setting similarity, is stored in described the second memory cell.
In addition, described method also comprises the step of the video information that DVR storage is obtained from image unit.
Through above-mentioned improvement, one aspect of the present invention has guaranteed the high-quality of photo, and the misoperation that avoided that the monitored picture that in the past brought because of low video compression format is fuzzy, brings when distortion and manual operation and again compression bring quality to reduce; Directly complete on the other hand portrait photo's collecting work, reduced the workload of artificial playback, screening and crawl.
Embodiment bis-
The difference of the present embodiment and embodiment mono-is, in the present embodiment, described system also comprises time record cell, the time occurring in order to record each personnel of each facial image representative; When repeated detection occurs to same personnel, record respectively this people's time of occurrence.Recording method can realize by setting up a picture data storehouse, and memory data output is very little.
This scheme not only can be obtained people's face information, can obtain the time point that these personnel occur simultaneously.
Embodiment tri-
The difference of the present embodiment and embodiment mono-is, in the present embodiment, described system only comprises image acquisition units 12, buffer unit 13, face alignment unit 14, optical sieving unit 15 (also can comprise data processing unit 16), and do not comprise image unit 11, memory cell 17, this system and other system are used in conjunction with, and other system comprises image unit 11, memory cell 17.
Embodiment tetra-
The difference of the present embodiment and embodiment mono-is, in the present embodiment, the facial image grasping system of DVR only comprises image acquisition units, and other unit omit.
Here description of the invention and application is illustrative, not wants by scope restriction of the present invention in the above-described embodiments.Here the distortion of disclosed embodiment and change is possible, and for those those of ordinary skill in the art, the various parts of the replacement of embodiment and equivalence are known.Those skilled in the art are noted that in the situation that not departing from spirit of the present invention or substantive characteristics, and the present invention can be with other form, structure, layout, ratio, and realizes with other assembly, material and parts.In the situation that not departing from the scope of the invention and spirit, can carry out other distortion and change to disclosed embodiment here.
Claims (8)
1. there is the DVR that facial image captures function, for storing the video information of obtaining from image unit; It is characterized in that, it comprises facial image grasping system, and in order to obtain the facial image obtaining from image unit, this facial image grasping system comprises:
Image acquisition units, is connected with described image unit, the image information of obtaining in order to the described image unit of real-time crawl, and intercepting facial image wherein;
Described facial image grasping system also comprises: the first memory cell, in order to the facial image of memory image collecting unit Real-time Collection;
The real-time comparing unit of people's face, in order to the facial image of image acquisition units Real-time Collection is compared between two, calculates its similarity; Two images that are greater than setting similarity are considered to represent same person face;
Image screens unit in real time, in order to the similarity calculating according to the real-time comparing unit of people's face, screens the facial image in described the first memory cell, and the human face posture image of same person only retains one;
The second memory cell, in order to store satisfactory facial image;
People's face is comparing unit afterwards, in order to image is screened in real time to the setting facial image having in facial image after unit screening and described the second memory cell, compares, and calculates its similarity; The similarity of the facial image of setting in selection and described the second memory cell is all less than the facial image of setting similarity, is stored in described the second memory cell.
2. DVR according to claim 1, is characterized in that:
Described the first memory cell is buffer memory.
3. DVR according to claim 2, is characterized in that:
Described image acquisition units intercepting facial image is to buffer memory, and every the facial image of setting-up time T1 or every intercepting setting number, the real-time comparing unit of people's face is compared above-mentioned facial image, calculates its similarity;
If the similarity of two facial images is greater than setting similarity, delete a wherein facial image;
Then, the real-time comparing unit of people's face compares the facial image in remaining facial image in buffer memory and the second memory cell.
4. DVR according to claim 1, is characterized in that:
Described system also comprises data processing unit, in order to process the facial image of described intercepting.
5. DVR according to claim 1, is characterized in that:
Setting described similarity is 75%, and the similarity of two images is more than or equal to 75% and is considered to represent same person face.
6. DVR according to claim 1, is characterized in that:
Described system also comprises time record cell, the time occurring in order to record each personnel of each facial image representative; When repeated detection occurs to same personnel, record respectively this people's time of occurrence.
7. a facial image grasping means for DVR described in claim 1 to 6 any one, is characterized in that, the method comprises the steps:
Image unit obtains image information;
Image acquisition units captures the image information that described image unit obtains in real time, and intercepting facial image wherein;
The real-time comparing unit of the real-time comparing unit people of people's face face compares the setting facial image having in the facial image of intercepting and the first memory cell, calculates its similarity; Two images that are greater than setting similarity are considered to represent same person face;
Image screens in real time the similarity that unit calculates according to the real-time comparing unit of people's face and screens the facial image in described the first memory cell, and the human face posture image of same person only retains one;
Optical sieving unit screens the facial image of described intercepting, and the facial image of intercepting and the setting facial image of the first cell stores are compared; The similarity of the facial image of setting in selection and the first memory cell is all less than facial image setting similarity, new intercepting and is stored in the second memory cell.
8. facial image grasping means according to claim 7, is characterized in that:
Described the first memory cell is buffer memory;
Described image acquisition units intercepting facial image is to buffer memory, and every the facial image of setting-up time T1 or every intercepting setting number, the real-time comparing unit of the real-time comparing unit people of people's face face is compared above-mentioned facial image, calculates its similarity;
If the similarity of two facial images is greater than setting similarity, delete a wherein facial image;
Then, the real-time comparing unit of people's face compares the facial image in remaining facial image in buffer memory and the second memory cell.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910050916.7A CN101557487B (en) | 2009-05-08 | 2009-05-08 | Hard disk recorder with human face image capturing function and method for capturing a human face image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN200910050916.7A CN101557487B (en) | 2009-05-08 | 2009-05-08 | Hard disk recorder with human face image capturing function and method for capturing a human face image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101557487A CN101557487A (en) | 2009-10-14 |
| CN101557487B true CN101557487B (en) | 2014-08-13 |
Family
ID=41175410
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN200910050916.7A Expired - Fee Related CN101557487B (en) | 2009-05-08 | 2009-05-08 | Hard disk recorder with human face image capturing function and method for capturing a human face image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN101557487B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102880864A (en) * | 2012-04-28 | 2013-01-16 | 王浩 | Method for snap-shooting human face from streaming media file |
| CN105138985A (en) * | 2015-08-25 | 2015-12-09 | 北京拓明科技有限公司 | Real-name authentication method based on WeChat public number and system |
| TWI573093B (en) | 2016-06-14 | 2017-03-01 | Asustek Comp Inc | Method of establishing virtual makeup data, electronic device having method of establishing virtual makeup data and non-transitory computer readable storage medium thereof |
| CN109241942B (en) * | 2018-09-29 | 2022-05-03 | 佳都科技集团股份有限公司 | Image processing method and device, face recognition equipment and storage medium |
| CN109998496A (en) * | 2019-01-31 | 2019-07-12 | 中国人民解放军海军工程大学 | A kind of autonomous type body temperature automatic collection and respiratory monitoring system and method |
| CN110826390B (en) * | 2019-09-09 | 2023-09-08 | 博云视觉（北京）科技有限公司 | Video data processing method based on face vector characteristics |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1710590A (en) * | 2004-06-17 | 2005-12-21 | 上海银晨智能识别科技有限公司 | Human-face picture searching method |
| CN1975789A (en) * | 2006-12-22 | 2007-06-06 | 重庆交通大学 | Multifunction entrance guard, monitoring and alarming comprehensive system |
| CN101242490A (en) * | 2007-02-06 | 2008-08-13 | 上海银晨智能识别科技有限公司 | Portable DVR |
| CN101546377A (en) * | 2009-04-28 | 2009-09-30 | 上海银晨智能识别科技有限公司 | Face image capture system and method |
-
2009
- 2009-05-08 CN CN200910050916.7A patent/CN101557487B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1710590A (en) * | 2004-06-17 | 2005-12-21 | 上海银晨智能识别科技有限公司 | Human-face picture searching method |
| CN1975789A (en) * | 2006-12-22 | 2007-06-06 | 重庆交通大学 | Multifunction entrance guard, monitoring and alarming comprehensive system |
| CN101242490A (en) * | 2007-02-06 | 2008-08-13 | 上海银晨智能识别科技有限公司 | Portable DVR |
| CN101546377A (en) * | 2009-04-28 | 2009-09-30 | 上海银晨智能识别科技有限公司 | Face image capture system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN101557487A (en) | 2009-10-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101546377A (en) | Face image capture system and method | |
| CN101557487B (en) | Hard disk recorder with human face image capturing function and method for capturing a human face image | |
| US11197057B2 (en) | Storage management of data streamed from a video source device | |
| CN103971090B (en) | Image processing apparatus, image capture apparatus and image processing method | |
| TW544634B (en) | Thumbnail sequence generation system and method | |
| EP2573758B1 (en) | Method and apparatus for displaying summary video | |
| CN101217623B (en) | A quick manual focusing method | |
| KR20160118963A (en) | Real-time image stitching apparatus and real-time image stitching method | |
| CN108848354B (en) | VR content camera system and working method thereof | |
| KR20110132884A (en) | Apparatus for intelligent video information retrieval supporting multi channel video indexing and retrieval, and method thereof | |
| JP2004312264A (en) | Video image recording system, meta-data adding device, image pickup device, video signal recording device, video image recording method, meta-data format | |
| CN101646070A (en) | Novel intelligent security-protecting and monitoring method and special equipment thereof | |
| CN116916049B (en) | Video data online acquisition and storage system based on cloud computing technology | |
| KR102228457B1 (en) | Methed and system for synchronizing usage information between device and server | |
| JP2011101383A (en) | Image file generation apparatus | |
| CN101281649A (en) | Method for detecting surroundings variation movably | |
| CN105704387A (en) | Shooting method and device of intelligent terminal and intelligent terminal | |
| DE102012200417B4 (en) | Image data recording device | |
| CN103716590B (en) | A kind of high definition trial system based on electronic fingerprint stamp | |
| TW201136313A (en) | 3D image monitor system and method | |
| KR100490948B1 (en) | Method for transferring image signals and system using the method | |
| CN101971620A (en) | A method of recording quality images | |
| CN106303366A (en) | A kind of method and device of Video coding based on territorial classification coding | |
| TW201112017A (en) | Multithreading video image compilation and search method and system | |
| CN111510666A (en) | Distributed equipment intelligent monitoring system and method convenient to view |
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 | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140813 Termination date: 20200508 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |