CN102457657A - Camera device and method for carrying out human type detection by using the same - Google Patents
Camera device and method for carrying out human type detection by using the same Download PDFInfo
- Publication number
- CN102457657A CN102457657A CN201010511755XA CN201010511755A CN102457657A CN 102457657 A CN102457657 A CN 102457657A CN 201010511755X A CN201010511755X A CN 201010511755XA CN 201010511755 A CN201010511755 A CN 201010511755A CN 102457657 A CN102457657 A CN 102457657A
- Authority
- CN
- China
- Prior art keywords
- image
- human
- camera device
- detecting
- dynamic area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
A camera device and a method for carrying out human type detection by using the same are disclosed. The method comprises the following steps: acquiring monitored object images continuously shot by a lens of the camera device; detecting a dynamic area in the acquired images; if the dynamic area is detected, using a human type detecting technology to detect a human type image in the dynamic area. By using the device and the method of the invention, a speed of detecting the human type image can be accelerated.
Description
Technical field
The present invention relates to a kind of combination dynamic detection technology, camera device and the method thereof that can under the complex background environment, carry out human-like detecting fast.
Background technology
The monitored object image that traditional human-like method for detecting is all directly taken the photograph with camera device, the human-like image of detecting in entire image.Because most watch-dogs improve the resolution of camera device photographic images one after another for strengthening the definition of image quality.So if directly in the high-definition picture of picked-up, carry out human-like image detecting, because the size of high-definition picture is bigger, required detecting analysis time can be longer.
Even utilize other software that high-definition picture is carried out carrying out human-like image detecting again behind contract drawing, the reduction screen resolution, can not effectively reduce the time that detecting is analyzed.Because the computing operation of high-definition picture being carried out contract drawing, reduction resolution also need expend to a certain degree time and machine arithmetic amount.
Summary of the invention
In view of above content, be necessary to provide a kind of method of utilizing camera device to carry out human-like detecting, it can detect the dynamic area with the dynamic detection technology earlier in the image of picked-up, in this dynamic area, carry out human-like detecting then.
In one embodiment, this method comprises the steps:
Obtain the monitored object image of the continuous picked-up of camera lens of camera device;
In the image that obtains, detect the dynamic area; And
If detect the dynamic area, then utilize human-like detection techniques, the human-like image of detecting in this dynamic area.
Preceding method can be carried out by camera device, and wherein this camera device has one or more processors, memory and is kept at one or more modules, program or the instruction set that is used to carry out these methods in the memory.
The instruction that is used for carrying out preceding method can be included in and be configured to the program product carried out by one or more processors.
Compared to prior art, described camera device and utilize its method of carrying out human-like detecting is detected the dynamic area with the dynamic detection technology earlier in the image of picked-up, in this dynamic area, carry out human-like detecting then.Because this method only need be carried out human-like detecting to the dynamic area of pickup image, has significantly dwindled the image capturing range of the required processing of human-like detecting, has effectively saved the activity duration of human-like detecting, has promoted the operational paradigm of human-like detecting.Use the present invention under the complex environment condition, to utilize dynamic detection technology and human-like detection techniques to come to detect accurately and apace human-like position.The present invention can be used on environment such as building monitoring and security monitoring, and combines with existing watch-dog, does not need to buy more again the watch-dog of better quality, to save the cost of applying unit again.
Description of drawings
Fig. 1 is the block diagram of camera device preferred embodiment of the present invention.
Fig. 2 is the module map of human-like detecting system.
Fig. 3 is that the present invention utilizes camera device to carry out the flow chart of preferred embodiment of the method for human-like detecting.
Fig. 4 is the sketch map of a certain dynamic area.
Fig. 5 is the sketch map that human-like image is detected in the dynamic area in Fig. 4.
The main element symbol description
| Camera device | 2 |
| Human- |
20 |
| |
21 |
| |
22 |
| |
23 |
| Processor | ?24 |
| Image document | ?40 |
| The dynamic area | ?41 |
| Human-like image | ?42 |
| Image collection module | ?201 |
| The dynamic area detecting module | ?202 |
| Human-like image detecting module | ?203 |
| Lens regulating module | ?204 |
Embodiment
As shown in Figure 1, be the block diagram of camera device preferred embodiment of the present invention.In the present embodiment, this camera device 2 comprises camera lens 21, memory 22, driver 23 and processor 24.Said camera device 2 comprises; But be not limited to, can by software or hardware circuit mode drive patrol turntable video camera, fire ball video camera (SpeedDome) but and translation (Pan), tilt (Tilt), PTZ (Pan/Tilt/Zoom) video camera of convergent-divergent (Zoom) etc.
Wherein, said camera lens 21 is used for obtaining continuously the image of monitored object.In the present embodiment, said camera lens 21 be charge coupled device (Charge Coupled Device, CCD).Said driver 23 can be a CD-ROM drive motor, and the camera lens 21 that is used to drive camera device 2 moves.
Store human-like detecting system 20 in the said memory 22, this human-like detecting system 20 is used for detecting the dynamic area with the dynamic detection technology at the image of picked-up earlier, in this dynamic area, carries out human-like detecting then, and detailed process is referring to the description of Fig. 3.
In the present embodiment, said human-like detecting system 20 can be divided into one or more modules, and said one or more modules are configured to carried out by one or more processors (present embodiment is a processor 24), to accomplish the present invention.For example, consult shown in Figure 2ly, said human-like detecting system 20 is divided into image collection module 201, dynamic area detecting module 202, human-like image detecting module 203 and lens regulating module 204.The alleged module of the present invention is to accomplish the block of a specific function, is more suitable in describing the implementation of software in camera device 2 than program, and the function of each module will combine the flow chart of Fig. 3 to describe in detail.
As shown in Figure 3, be that the present invention utilizes camera device to carry out the flow chart of preferred embodiment of the method for human-like detecting.
Step S1, image collection module 201 is obtained the monitored object image of the camera lens 21 continuous picked-ups of camera device 2.For example, suppose that 21 each seconds taken 10 images by the camera lens of camera device 2, then be 0.1 second the blanking time of camera device 2 shootings, promptly every at a distance from 0.1 second, and the camera lens 21 of camera device 2 is taken an image.
Step S2, dynamic area detecting module 202 is detected the dynamic area in the image that obtains.Particularly, dynamic area detecting module 202 obtains earlier the present image and a last image of monitored object from continuous some images, and calculates the characteristic value (like the blue gray value of image) of a present image and a last image.Then, dynamic area detecting module 202 carries out fuzzy matching with the characteristic value of a present image and a last image, to find out the zone that conforms to of a present image and a last image.At last; Dynamic area detecting module 202 accurately matees the characteristic value in the zone that conforms to of a present image and a last image; And the characteristic value different portions is denoted as the dynamic area of said monitored object, this dynamic area that indicates is as the mobile object in the monitored object.
Consult shown in Figure 4ly, image document 40 is the image that the camera lens 21 of camera device 2 at a time absorbs, the dynamic area that label 41 representative dynamic area detecting modules 202 detect in image document 40.In Fig. 4, the character positions in the dynamic area 41 has taken place to move, and the dotted line representative figure is in the position in a last moment, and the solid line representative figure is in the position of current time.
Step S3, dynamic area detecting module 202 judges whether to detect the dynamic area.If detect the dynamic area, execution in step S4; If do not detect the dynamic area, return step S2.
Step S4, human-like image detecting module 203 is utilized human-like detection techniques, the human-like image of detecting in this dynamic area.In the present embodiment, said human-like detection techniques includes, but not limited to human-like characteristic information statistic law and feature samples comparison classification (TemplateMatching Method).
Particularly, human-like characteristic information statistic law comprises the steps:
(1) with image processing mode with digitized video background simplification to be measured;
(2) the human-like characteristic point data that surpasses each posture more than 100,000 in image data to be measured and the database is compared;
(3) estimate by the characteristic point data in the image to be measured with statistical and whether have human-like information to exist in the image to be measured.
Feature samples comparison classification comprises the steps:
(1) collect the human-like feature samples of each posture of some and the non-human-like feature samples of some earlier, for example, human-like pictures such as the stance front of collection some and stance side, sitting posture;
(2) after the non-human-like feature samples of waiting to accomplish human-like feature samples and the some of some different gestures is collected; Beginning is further carried out continuous training (Training) with neural network (Artificial NeuralNetwork) training method, continues to correct mistakes.If do not use the neural network training method, also can change and classify with the Ada-Boost classification, the human-like characteristic masterplate of accomplishing through training (Template) just or the Ada-Boost grader can supply follow-up human-like detecting (Testing) to use.
In other embodiments, this method also further comprises step: lens regulating module 204 is according to the mobile data of human-like image, and Control Driver 23 is adjusted the position of camera lenses 21.In the present embodiment, the mobile data of said human-like image comprises: the moving direction of human-like image, move angle and displacement.
For example, lens regulating module 204 is given driver 23 according to moving direction, move angle and the displacement sending controling instruction of human-like image.Driver 23 makes the camera lens 21 of camera device 2 aim at this human-like image according to the camera lens 21 of this control command travelling shot machine 2.Simultaneously, lens regulating module 204 amplifies the focal length of (Zoom in) camera lenses 21, make the human-like image that obtains more greatly, more clear.
The present invention is that example describes with the security monitoring field, and is same, and this method also can be applied to other association area, like home care, save supervision etc. from damage.
What should explain at last is; Above embodiment is only unrestricted in order to technical scheme of the present invention to be described; Although the present invention is specified with reference to preferred embodiment; Those of ordinary skill in the art should be appreciated that and can make amendment or be equal to replacement technical scheme of the present invention, and do not break away from the spirit and the scope of technical scheme of the present invention.
Claims (10)
1. a method of utilizing camera device to carry out human-like detecting is characterized in that, this method comprises the steps:
Obtain the monitored object image of the continuous picked-up of camera lens of camera device;
In the image that obtains, detect the dynamic area; And
If detect the dynamic area, then utilize human-like detection techniques, the human-like image of detecting in this dynamic area.
2. the method for utilizing camera device to carry out human-like detecting as claimed in claim 1 is characterized in that, the said step of in the image that obtains, detecting the dynamic area comprises:
From continuous images, obtain the present image and a last image of monitored object, and calculate the characteristic value of a present image and a last image;
The characteristic value of a present image and a last image is carried out fuzzy matching, to find out the zone that conforms to of a present image and a last image; And
The regional characteristic value that conforms to of a present image and a last image is accurately mated, and the characteristic value different portions is denoted as the dynamic area of said monitored object.
3. the method for utilizing camera device to carry out human-like detecting as claimed in claim 1 is characterized in that, said human-like detection techniques comprises human-like characteristic information statistic law and feature samples comparison classification.
4. the method for utilizing camera device to carry out human-like detecting as claimed in claim 1; It is characterized in that; This method also comprises step: according to the mobile data of human-like image, the position of the driver adjustment camera lens of control camera device makes this human-like image of alignment lens of camera device.
5. the method for utilizing camera device to carry out human-like detecting as claimed in claim 4 is characterized in that, this method also comprises step: the focal length of enlarging lens, make the human-like image that obtains more greatly, more clear.
6. a camera device is characterized in that, this camera device comprises:
Camera lens;
Memory;
Driver;
One or more processors; And
One or more modules, said one or more modules are stored in the said memory and are configured to and carried out by said one or more processors, and said one or more modules comprise:
Obtain the module of the continuous monitored object image that absorbs of camera lens of camera device;
The module of detecting dynamic area in the image that obtains; And
If detect the dynamic area, then utilize human-like detection techniques, the module of the human-like image of detecting in this dynamic area.
7. camera device as claimed in claim 6 is characterized in that, the said module of in the image that obtains, detecting the dynamic area comprises:
From continuous images, obtain the present image and a last image of monitored object, and calculate the submodule of the characteristic value of a present image and a last image;
The characteristic value of a present image and a last image is carried out fuzzy matching, to find out the regional submodule that conforms to of a present image and a last image; And
The regional characteristic value that conforms to of a present image and a last image is accurately mated, and the characteristic value different portions is denoted as the submodule of the dynamic area of said monitored object.
8. camera device as claimed in claim 6 is characterized in that, said human-like detection techniques comprises human-like characteristic information statistic law and feature samples comparison classification.
9. camera device as claimed in claim 6; It is characterized in that; Said one or more module also comprises: according to the mobile data of human-like image, the position of the driver adjustment camera lens of control camera device makes the module of this human-like image of alignment lens of camera device.
10. camera device as claimed in claim 9 is characterized in that, said one or more modules also comprise: the focal length of enlarging lens, make the human-like image that obtains more greatly, module more clearly.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010511755XA CN102457657A (en) | 2010-10-19 | 2010-10-19 | Camera device and method for carrying out human type detection by using the same |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201010511755XA CN102457657A (en) | 2010-10-19 | 2010-10-19 | Camera device and method for carrying out human type detection by using the same |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102457657A true CN102457657A (en) | 2012-05-16 |
Family
ID=46040274
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201010511755XA Pending CN102457657A (en) | 2010-10-19 | 2010-10-19 | Camera device and method for carrying out human type detection by using the same |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102457657A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106073167A (en) * | 2016-05-31 | 2016-11-09 | 常州米乔科技有限公司 | A kind of alternately sitting and standing office system and based reminding method thereof |
| TWI715376B (en) * | 2019-12-26 | 2021-01-01 | 台灣新光保全股份有限公司 | Security method and security system based on image analysis |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101017573A (en) * | 2007-02-09 | 2007-08-15 | 南京大学 | Method for detecting and identifying moving target based on video monitoring |
| CN101112099A (en) * | 2005-02-01 | 2008-01-23 | 松下电器产业株式会社 | Monitor recording device |
| CN101308575A (en) * | 2008-07-08 | 2008-11-19 | 北京中星微电子有限公司 | Target positioning method and sight protective apparatus |
| CN101339688A (en) * | 2008-08-27 | 2009-01-07 | 北京中星微电子有限公司 | Intrusion checking method and system |
| CN101339605A (en) * | 2008-08-14 | 2009-01-07 | 北京中星微电子有限公司 | Detection system and method for number of people based on video frequency monitoring |
-
2010
- 2010-10-19 CN CN201010511755XA patent/CN102457657A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101112099A (en) * | 2005-02-01 | 2008-01-23 | 松下电器产业株式会社 | Monitor recording device |
| CN101017573A (en) * | 2007-02-09 | 2007-08-15 | 南京大学 | Method for detecting and identifying moving target based on video monitoring |
| CN101308575A (en) * | 2008-07-08 | 2008-11-19 | 北京中星微电子有限公司 | Target positioning method and sight protective apparatus |
| CN101339605A (en) * | 2008-08-14 | 2009-01-07 | 北京中星微电子有限公司 | Detection system and method for number of people based on video frequency monitoring |
| CN101339688A (en) * | 2008-08-27 | 2009-01-07 | 北京中星微电子有限公司 | Intrusion checking method and system |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106073167A (en) * | 2016-05-31 | 2016-11-09 | 常州米乔科技有限公司 | A kind of alternately sitting and standing office system and based reminding method thereof |
| TWI715376B (en) * | 2019-12-26 | 2021-01-01 | 台灣新光保全股份有限公司 | Security method and security system based on image analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2018228410A1 (en) | Target object capturing method and device, and video monitoring device | |
| CN109040709B (en) | Video monitoring method and device, monitoring server and video monitoring system | |
| US20110115917A1 (en) | Surveillance system and surveilling method | |
| EP3641298B1 (en) | Method and device for capturing target object and video monitoring device | |
| CN109376601B (en) | Object tracking method based on high-speed ball, monitoring server and video monitoring system | |
| CN112954315B (en) | Image focusing measurement method and system for security camera | |
| CN110460769B (en) | Image correction method, image correction device, computer equipment and storage medium | |
| CN101640788B (en) | Method and device for controlling monitoring and monitoring system | |
| CN110072078B (en) | Monitoring camera, control method of monitoring camera, and storage medium | |
| CN101877764A (en) | Camera system and method for carrying out assisted drawing by utilizing same | |
| CN109905641B (en) | Target monitoring method, device, equipment and system | |
| CN1980384A (en) | Space mobile-object locking aim-searching device and method | |
| EP3432575A1 (en) | Method for performing multi-camera automatic patrol control with aid of statistics data in a surveillance system, and associated apparatus | |
| CN109389367B (en) | Personnel attendance checking method, device and storage medium | |
| CN102438122A (en) | Camera device and method for dynamically detecting monitoring object by using same | |
| CN102457657A (en) | Camera device and method for carrying out human type detection by using the same | |
| TWM560035U (en) | Image tracking device | |
| CN102469247A (en) | Photographic device and dynamic focusing method thereof | |
| CN112702513B (en) | Double-optical-pan-tilt cooperative control method, device, equipment and storage medium | |
| JP6301202B2 (en) | Shooting condition setting device and shooting condition setting method | |
| Srivastava et al. | Design and implementation of a real-time autofocus algorithm for thermal imagers | |
| CN101726963B (en) | Method for identifying dimension form of shot subject | |
| EP2570962A1 (en) | Video analysis | |
| CN116489317B (en) | Object detection method, system and storage medium based on image pickup device | |
| CN102447881A (en) | Camera device and method for monitoring image by camera device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20120516 |
|
| RJ01 | Rejection of invention patent application after publication |