CN201965633U - Human face detection and recognition system - Google Patents
Human face detection and recognition system Download PDFInfo
- Publication number
- CN201965633U CN201965633U CN2011200084672U CN201120008467U CN201965633U CN 201965633 U CN201965633 U CN 201965633U CN 2011200084672 U CN2011200084672 U CN 2011200084672U CN 201120008467 U CN201120008467 U CN 201120008467U CN 201965633 U CN201965633 U CN 201965633U
- Authority
- CN
- China
- Prior art keywords
- image
- human face
- face detection
- image processing
- chip
- 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
Images
Landscapes
- Image Processing (AREA)
Abstract
The utility model relates to a human face detection and recognition system, which comprises an image processing chip, an information memory device, an image camcorder, a random access memory and an output device, wherein the image processing chip comprises a computing element which is used for executing relevant computation for image processing, an image detecting element which is used for capturing photographed images and screening proper facial features as computation units, and a writing area which is used for storing an executable image recognition module; the information memory device is used for storing all human image information and relevant feature information which are used by the image recognition module for comparison; the image camcorder is used for capturing human images of users within a certain distance for many times and transmitting the human images to the image processing chip for required processing; the random access memory is used for the random access of the information during computation; and the output device is used for outputting detection results. The human face detection and recognition system has the beneficial effects that a chip with a detection function only is enabled to have a recognition function in application, the use cost can be saved and the same goal can be achieved at a low cost.
Description
Technical field
The utility model relates to a kind of image identification system, especially a kind of human face detection identification system.
Background technology
As shown in Figure 2, the human face detection identification system mainly comprises traditionally: a data storage device, and this device stores everyone as image data, and is used as the comparison of image identification device; One image is shot with video-corder device, repeatedly captures the portrait image of the user in certain segment distance at place; One image detecting device, this detector filter out after the suitable portrait image, simulate a portrait by computing; One image identification device is compared the effigy and the data bank that calculate, and is judged both similarities with the portrait image data that stores.
When a user is positioned at image and shoots with video-corder the device segment distance, image is shot with video-corder device and people's face image of this user can be captured by the mode of shooting with video-corder, in the process of acquisition, not only capture an image, can capture a series of people's face image, then again via the image detecting device in the utility model, this image detecting device can be earlier shot with video-corder in people's face image of device under capturing at image, filter out angle, the suitable people's face image in position, the face feature that will be fit to is as arithmetic unit, as unique points such as the bridge of the nose, canthus again.By specific compute mode, form an effigy, then, to utilize via the effigy that computing forms the feature identification unit directly and possessor's face feature data of being read out of feature identification data bank do a series of identification analysis, if when the similarity of both face features of comparison exceeds a threshold value, think that promptly the portrait of this user and data bank institute access is same people; When if the similarity of both face features of comparison is lower than threshold value, then not think to be same people.
When carrying out image identification, must carry out the action of image detecting earlier, detecting the position of people's face in the picture, and then with data bank in the data that stores compare.Often these actions must be finished immediately in the application of reality, but suitable really huge of needed operand.So system can't finish these huge computings immediately.Cause to purchase just can the load needs of computing of expensive system.
The purpose of this utility model is to use effective configuration of existing apparatus, saving cost, and can reach the demand in the computing.
At first, must set up user's data earlier, its flow process be for, when allowing a user be positioned at image to shoot with video-corder the device segment distance, the camera lens (for CMOS pick-up lens or CCD pick-up lens) that CPU shoots with video-corder device via image captures this user's image, and be converted to the yuv format view data be stored to memory body (the synchronous dynamic random-access memory body, SDRAM) in the middle of.Yuv format is the kind for compiling true-color color space (color space), " Y " represents lightness (Luminance, Luma), " U " and " V " then is colourity, concentration (Chrominance, Chroma), is commonly used to describe simulating signal.Then, the transmission steering order starts portrait identification and analysis process to the human face recognition chip.And portrait detecting chip again sending controling instruction to people's face detection chip, the portrait area of searching and locating this user in the image of people's face detection chip yuv format from be stored in memory body.And this user's portrait extracted the portrait feature, again this user's portrait feature extracted and this user's status data are stored in the permanent memory body in the lump.Above-mentioned promptly is that people's face is registered, and face characteristic is extracted in the zone of promptly automatically extracting people's face from image then, is stored in the permanent memory body, in SD card, magnetic sheet etc.
Yet, during identification people face, the user must be positioned at this image earlier and shoot with video-corder the device segment distance, to allow pick-up lens be captured this user's image, CPU can be by after this pick-up lens acquisition image, with the format conversion of this image to yuv format and be stored in memory body (synchronous dynamic random-access memory body, SDRAM) in the middle of, simultaneously, CPU can send the transmission steering order to the human face recognition chip, and to start the flow process of human face detection and analysis, the human face recognition chip receives after the transmission steering order, sending controling instruction is to the human face detection chip, the human face detection chip is located the zone of this people's face and is extracted its feature from image, the feature extracted and the feature in the permanent memory body are compared again, and seeks similar feature, if find similar features then to compare success, otherwise i.e. failure.And its result can be sent to identification chip, and sends CPU to again, and CPU exports via output system.
The utility model content
Traditionally, with the algorithm of detecting and the algorithm of identification simultaneously framework in a chip.Because suitable big of this two algorithm calculated amount when actual operation is in order to reach the purpose of instant (Real-Time), so must use the needs that chip at a high speed can be realistic.But, the costliness that the high-speed chip of this type is suitable.
The purpose of this utility model is to use effective configuration of existing apparatus, saving cost, and can reach the demand in the computing.
For reaching above-mentioned purpose, the utility model provides a kind of human face detection identification system, comprises: an image processing chip, and this image processing chip comprises: the arithmetic element that can carry out the related operation of image processing; An image detecting element is the interior element that is built in the image processing chip, and it is the element that forms simultaneously in processing procedure with the image detecting element; A storing memory has the write area of executable module, and this write area has the function of storing memory, stores with the image identification module that will be write.Behind chip enable, the program of write area being written into arithmetic element, the image identification computing of being correlated with by this arithmetic element,
Wherein:
Also comprise one store everyone as image data and correlated characteristic data with as the used data storage device of image identification module comparison, it links this image processing chip.
Also comprise an image and shoot with video-corder device, user's the portrait image in the segment distance at acquisition place, and the image of being taken in is sent to the image processing chip to carry out needed processing.
Also comprise the random access memory of the usefulness that a data during as computing stores at random, it links this image processing chip.
Also comprise an output unit that is used to export the result of detecting.
This write area comprises the module that is write by the outside, and this module is compared for effigy that will calculate and the portrait image data that has stored, and judges the image identification module of both similarities.
During the image detecting element application, this image detecting element grasps the image of taking in and filters out after the suitable portrait image again, simulates a portrait by computing; By filtering out angle, the suitable people's face image in position in this portrait, the face feature that will be fit to is as arithmetic unit more again.
It is CMOS or CCD that image is shot with video-corder device usability optical element.
Useful technique effect of the present utility model is that upward only need detect by some application can be used, the chip that has only detecting function that has on the market, and it is because only have detecting function, and price is also comparatively cheap.In the utility model, the chip that is applied to the type carries it is write identification algorithm module, and make this kind have only the chip of detecting function also to have the function of identification simultaneously, but its overall price is still also cheap than the chip that originally has detecting and discriminating function simultaneously, thereby reached the saving cost, and can reach the demand in the computing.
Description of drawings
Fig. 1 shows inner bay composition of the present utility model.
Fig. 2 shows the process structure figure of prior art.
[main element symbol description]
10 image processing chips
11 arithmetic elements
12 image detecting elements
13 write areas
20 data storage devices
30 images are shot with video-corder device
40 random access memories
50 output units.
Embodiment
Propose a kind of human face detection identification system in the utility model, please refer to the Organization Chart of Fig. 1, system wherein of the present utility model comprises following elements:
One image processing chip 10 comprises:
One arithmetic element 11 can be carried out the related operation of image processing;
One image detecting element 12 is the interior element of building in the image processing chip 10, can be the element that forms simultaneously in processing procedure with image detecting element 12.During application, this image detecting element 10 grasps the image of taking in and filters out after the suitable portrait image again, simulates a portrait by computing.By filtering out angle, the suitable people's face image in position in this portrait, the face feature that will be fit to is as arithmetic unit more again;
One write area 13, can be used for writing executable software module, behind chip enable, the module of write area 13 can being written into arithmetic element 11, the computing of being correlated with by this arithmetic element 11, this write area 13 has the function of storing memory, the software module that is write can be stored;
Wherein this write area 13 comprises the module that is write by the outside, and this module is the image identification program module.This identification software is compared effigy that calculates and the portrait image data that has stored, and judges both similarities.
The disposal system that the utility model is relevant is;
One data storage device 20 links this image processing chip 10, and this device stores everyone as image data and correlated characteristic data, and is used as the comparison of image identification program module.
One image is shot with video-corder device 30, repeatedly captures the portrait image of the user in certain segment distance at place, and the image of being taken in is sent to image processing chip 10 has carried out needed processing.
One random access memory 40 (as RAM, DRAM, SRAM, SDRAM or the like) links this image processing chip 10, the usefulness that data stores at random with as computing the time.
One output unit 50 is used to export the result of detecting.
Its operating process is, when a user is positioned at image and shoots with video-corder device 30 certain segment distance, image is shot with video-corder device 30 and people's face image of this user can be captured by the mode of shooting with video-corder, in the process of acquisition, not only capture an image, can capture a series of people's face image, then again via the image detecting element 12 in the utility model, this image detecting element 12 can be earlier shot with video-corder in people's face image of device 30 under capturing at image, filter out angle, people's face image that the position is fit to, the face feature that will be fit to is as arithmetic unit, as the bridge of the nose again, unique points such as canthus.By arithmetic element of the present utility model, form an effigy, then, the effigy that will form via computing directly and possessor's face feature data of being read out of data storage device 20 do a series of identification analysis, if when the similarity of both face features of comparison exceeds a threshold value, think that promptly the portrait of this user and data storage device institute access is same people; When if the similarity of both face features of comparison is lower than threshold value, then not think to be same people, again its comparison result is exported via output unit 50.
Traditionally, with the algorithm of detecting and the algorithm of identification simultaneously framework in a chip.Because suitable big of this two algorithm calculated amount when actual operation is in order to reach the purpose of instant (Real-Time), so must use the needs that chip at a high speed can be realistic.But, the costliness that the high-speed chip of this type is suitable.
Some is used to go up only need detect and gets final product, and the chip that has only detecting function is also arranged on the market, because detecting function is only arranged, price is also comparatively cheap.The chip of using the type in the utility model carries and writes identification algorithm module, and makes this kind have only the chip of detecting function also to have the function of identification simultaneously, but overall price is still also cheap than the chip that originally has detecting and discriminating function simultaneously.
In sum, the design of consideration of the utility model hommization, quite realistic demand.Its concrete improvement has disappearance now, obviously has breakthrough progressive advantage compared to prior art, the enhancement that has effect really, and non-being easy to reached.The utility model disclosed or was exposed on domestic and the external document and market, had met patent statute.
Above-listed detailed description is specifying at a possible embodiments of the present utility model, only this embodiment is not in order to limit claim of the present utility model, allly do not break away from the equivalence that the utility model skill spirit does and implement or change, all should be contained in the claim of the present utility model.
Claims (7)
1. a human face detection identification system is characterized in that, comprises:
An image processing chip, this image processing chip comprises:
An arithmetic element of carrying out the related operation of image processing;
An image detecting element is the interior element that is built in the image processing chip, and it is the element that forms simultaneously in processing procedure with the image detecting element;
A storing memory has the write area of executable image identification module, the computing that this module is correlated with by this arithmetic element behind chip enable.
2. human face detection identification system according to claim 1 is characterized in that, also comprise one store everyone as image data and correlated characteristic data with as the used data storage device of image identification module comparison, it links this image processing chip.
3. human face detection identification system according to claim 1, it is characterized in that, also comprise an image and shoot with video-corder device, user's the portrait image in the segment distance at acquisition place, and the image of being taken in is sent to the image processing chip to carry out needed processing.
4. human face detection identification system according to claim 1 is characterized in that, also comprises the random access memory of the usefulness that a data during as computing stores at random, and it links this image processing chip.
5. human face detection identification system according to claim 1 is characterized in that, also comprises an output unit that is used to export the result of detecting.
6. human face detection identification system according to claim 1, it is characterized in that, this write area comprises the image identification module that is write by the outside, and this module is compared for effigy that will calculate and the portrait image data that has stored, and judges the image identification module of both similarities.
7. human face detection identification system according to claim 3 is characterized in that, it is CMOS or CCD that image is shot with video-corder device usability optical element.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011200084672U CN201965633U (en) | 2011-01-13 | 2011-01-13 | Human face detection and recognition system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2011200084672U CN201965633U (en) | 2011-01-13 | 2011-01-13 | Human face detection and recognition system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN201965633U true CN201965633U (en) | 2011-09-07 |
Family
ID=44528180
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2011200084672U Expired - Fee Related CN201965633U (en) | 2011-01-13 | 2011-01-13 | Human face detection and recognition system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN201965633U (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104361327A (en) * | 2014-11-20 | 2015-02-18 | 苏州科达科技股份有限公司 | Pedestrian detection method and system |
| CN110276234A (en) * | 2018-03-14 | 2019-09-24 | 群光电能科技股份有限公司 | Image identification module extending box and its expansion application method |
-
2011
- 2011-01-13 CN CN2011200084672U patent/CN201965633U/en not_active Expired - Fee Related
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104361327A (en) * | 2014-11-20 | 2015-02-18 | 苏州科达科技股份有限公司 | Pedestrian detection method and system |
| CN110276234A (en) * | 2018-03-14 | 2019-09-24 | 群光电能科技股份有限公司 | Image identification module extending box and its expansion application method |
| CN110276234B (en) * | 2018-03-14 | 2023-04-18 | 群光电能科技股份有限公司 | Image identification module expansion box and expansion use method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106446873B (en) | Face detection method and device | |
| CN104598897B (en) | Visual sensor, image processing method and device, visual interactive equipment | |
| CN108710847A (en) | Scene recognition method, device and electronic equipment | |
| CN108229369A (en) | Image capturing method, device, storage medium and electronic equipment | |
| CN108805047A (en) | A kind of biopsy method, device, electronic equipment and computer-readable medium | |
| CN102708383A (en) | System and method for detecting living face with multi-mode contrast function | |
| CN102663758A (en) | Image acquiring and processing method based on FPGA (field programmable gate array) serving as control core | |
| CN103412954A (en) | Virtual dynamic magazine using augmented reality technique | |
| CN111626163A (en) | Human face living body detection method and device and computer equipment | |
| CN109756673A (en) | A kind of method and terminal controlling display state | |
| WO2023045626A1 (en) | Image acquisition method and apparatus, terminal, computer-readable storage medium and computer program product | |
| CN108875556A (en) | Method, apparatus, system and the computer storage medium veritified for the testimony of a witness | |
| CN201965633U (en) | Human face detection and recognition system | |
| CN202534042U (en) | Digital watermark detection device for paper money | |
| CN105807888B (en) | A kind of electronic equipment and information processing method | |
| CN108259769B (en) | Image processing method, image processing device, storage medium and electronic equipment | |
| WO2021078145A1 (en) | Wireless sensing facial recognition device based on living body sensing and movement trend detection | |
| CN106973231A (en) | Picture synthetic method and system | |
| CN106134475B (en) | Many image capture devices remote control acquisition control device with memory function | |
| CN201860418U (en) | Facial colorful superposition photographic device | |
| CN103093175A (en) | Face identification system on chip based on field programmable gate array (FPGA) | |
| CN212256328U (en) | Face recognition system | |
| CN112712006A (en) | Target picture snapshot method, system, medium and device | |
| WO2021237736A1 (en) | Image processing method, apparatus and system, and computer-readable storage medium | |
| CN210721506U (en) | Dynamic face recognition terminal based on 3D camera |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 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: 20110907 Termination date: 20150113 |
|
| EXPY | Termination of patent right or utility model |