CN102142179A - Vision-based intelligent security system - Google Patents
Vision-based intelligent security system Download PDFInfo
- Publication number
- CN102142179A CN102142179A CN 201010595879 CN201010595879A CN102142179A CN 102142179 A CN102142179 A CN 102142179A CN 201010595879 CN201010595879 CN 201010595879 CN 201010595879 A CN201010595879 A CN 201010595879A CN 102142179 A CN102142179 A CN 102142179A
- Authority
- CN
- China
- Prior art keywords
- people
- pet
- serial port
- card
- image
- 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
Landscapes
- Alarm Systems (AREA)
Abstract
The invention discloses a vision-based intelligent security system, and relates to an intelligent security system. The system comprises a plurality of cameras (1), an image acquisition card (2), a serial port expansion card (3), an upper computer (4), an alarm device (5) and a remote monitoring terminal (6), wherein the plurality of cameras (1) are connected with the image acquisition card (2) and the serial port expansion card (3) respectively; the image acquisition card (2) and the serial port expansion card (3) are connected with the upper computer (4); and the upper computer (4) is connected with the alarm device (5) and the remote monitoring terminal (6). The system can identify persons or pets, is convenient to use and more humanized, and has high intelligence degree and low cost.
Description
Technical field:
The present invention relates to a kind of intelligent safety and defence system, relate in particular to a kind of intelligent safety and defence system based on vision.
Background technology:
Existing intelligent safety and defence system based on vision is a kind of supervision mostly only, needs people constantly to monitor the situation of family by remote terminal, uses very inconvenient.The safety-protection system that has adopts and the infrared mode that combines, and when the infrared people of having detected or pet occurred, reminds people was passed through terminal monitors.If but pet frequently appears at monitor area, that frequently reminds people watch monitor terminal, the people is pestered beyond endurance.
Summary of the invention:
The purpose of this invention is to provide a kind of intelligent safety and defence system based on vision, it can solve the existing problem of traditional safety-protection system, and cost is low.
In order to solve the existing problem of traditional safety-protection system, the present invention is by the following technical solutions: it comprises many cameras, image pick-up card, serial port expanding module, host computer, warning device and remote monitoring terminal, multiple cameras links to each other with image pick-up card, serial port expanding module respectively, image pick-up card, serial port expanding module link to each other with host computer, and host computer links to each other with warning device and remote monitoring terminal.These many cameras are installed in respectively in a plurality of rooms that need security protection, so that can monitor a plurality of rooms.Warning device is settled respectively at home, on the remote monitoring terminal, in the room security.Remote monitoring terminal can be provided with a plurality of as the case may be, can monitor situation in the family by a plurality of people.Remote monitoring adopts multiple mode to realize, the one, the PC that can surf the Net, the 2nd, mobile phone with, the 3rd, special-purpose remote monitoring terminal.
Camera of the present invention is installed and is adopted ceiling to install, and makes the camera broad view, can monitor the whole zone that needs monitoring.
Software algorithm of the present invention is: the first step, adopt background image method of difference identification motion thing.At first, when the head of a family goes out, take a photo of home environment when safety-protection system starts and scheme as a setting.When people or pet enter in the camera visual field, photographic images again.It is poor that two images are done, and obtains the target area of motion thing.
In second step, identification is also differentiated people and pet.Based on the identification of depth-width ratio, background difference descendant's target area is farsighted much larger than wide, and pet is wider than height, and the depth-width ratio of the target area that can obtain according to the image difference point-score is discerned people and pet.
There was the people in the 3rd step if identify in the environment, warning device is sent instruction make its warning.If only with pet nobody, then do not report to the police in the environment.Simultaneously, people can be by the situation in each room of remote monitoring terminal real time monitoring family.
Background subtraction point-score of the present invention: a photo will taking home environment when safety-protection system starts is schemed as a setting.When pet or people enter in the environment photographic images.It is poor that two images are done, and obtains the target area of motion thing.
The method of background difference can be used following formulate:
(t) expression t two field picture is at coordinate (x, the gray-scale value of y) locating for x, y for Im in the formula; BkGd (x, y) the expression background picture is in that (x y) locates the gray-scale value of pixel; Threshold is a threshold value; Diff (x, y, t) value of the binary map that obtains of expression difference.If with respect to the grey scale change of background image during less than threshold value, this point shows as black in binary map in certain some place gray scale of current frame image; If this grey scale change greater than threshold value, then shows as bright spot in binary map.Like this, if when image changes for the gray scale of background image, just this change-detection can be come out by the background difference.Thereby form the binary map that has motion thing shape.
Difference image very easily is subjected to the pollution of noise.Ideally, static background parts should complete obiteration in difference image.But because the noise of sensor, difference image always has a lot of noises.And the edge of difference image is very rough often.In this case, can remove noise with the method for medium filtering.
Background difference descendant's target area is farsighted much larger than wide, and pet is wider than height, and the depth-width ratio of the target area that can obtain according to the image difference point-score is discerned people and pet.People's depth-width ratio is to be far longer than 2, and the depth-width ratio of pet is far smaller than 2.Can adopt 2 these threshold values to discern pet and people.
Advantage of the present invention is to identify people or pet, and is easy to use and more humane, intelligent degree height, and cost is low.
Description of drawings:
Fig. 1 is a structural representation of the present invention.
Embodiment:
With reference to Fig. 1, this embodiment is by the following technical solutions: it comprises multiple cameras 1, image pick-up card 2, serial port expanding module 3, host computer 4, warning device 5 and remote monitoring terminal 6, multiple cameras 1 links to each other with image pick-up card 2, serial port expanding module 3 respectively, image pick-up card 2, serial port expanding module 3 link to each other with host computer 4, and host computer 4 links to each other with warning device 5 and remote monitoring terminal 6.
Camera of the present invention is installed and is adopted ceiling to install, and makes the camera broad view, can monitor the whole zone that needs monitoring.
Software algorithm of the present invention is: the first step, adopt background image method of difference identification motion thing.At first, when the head of a family goes out, take a photo of home environment when safety-protection system starts and scheme as a setting.When people or pet enter in the camera visual field, photographic images again.It is poor that two images are done, and obtains the target area of moving object.
In second step, discern and tell people and pet.Based on the identification of depth-width ratio, background difference descendant's target area is farsighted much larger than wide, and pet is wider than height, and the depth-width ratio of the target area that can obtain according to the image difference point-score is discerned people and pet.
There was the people in the 3rd step if identify in the environment, instruction is reported to the police to warning device.If only with pet nobody, then do not report to the police in the environment.Simultaneously, people can be by the situation in each room of remote monitoring terminal real time monitoring family.
Background subtraction point-score of the present invention: a photo will taking home environment when safety-protection system starts is schemed as a setting.When pet or people enter in the environment photographic images.It is poor that two images are done, and obtains the target area of moving object.
The method of background difference can be used following formulate:
(t) (x y) locates gray-scale value to expression t two field picture to Im in the formula at coordinate for x, y; BkGd (x, y) the expression background picture is in that (x y) locates the gray-scale value of pixel; Threshold is a threshold value; Diff (x, y, t) value of the binary map that obtains of expression difference.If with respect to the grey scale change of background image during less than threshold value, this point shows as black in binary map in certain some place gray scale of current frame image; If this grey scale change greater than threshold value, then shows as bright spot in binary map.Like this, if when image changes for the gray scale of background image, just this change-detection can be come out by the background difference.Thereby form the binary map that has the moving object shape.
Difference image very easily is subjected to the pollution of noise.Ideally, static background parts should complete obiteration in difference image.But because the noise of sensor, difference image always has a lot of noises.And the edge of difference image is very rough often.In this case, can remove noise with the method for medium filtering.
Background difference descendant's target area is farsighted much larger than wide, and pet is wider than height, and the depth-width ratio of the target area that can obtain according to the image difference point-score is discerned people and pet.People's depth-width ratio is to be far longer than 2, and the depth-width ratio of pet is far smaller than 2.Can adopt 2 these threshold values to discern pet and people.
The advantage of this embodiment is to identify people or pet, and is easy to use and more humane, intelligent degree height, and cost is low.
Claims (1)
1. based on the intelligent safety and defence system of vision, it is characterized in that it comprises many cameras (1), image pick-up card (2), serial port expanding module (3), host computer (4), warning device (5) and remote monitoring terminal (6), multiple cameras (1) links to each other with image pick-up card (2), serial port expanding module (3) respectively, image pick-up card (2), serial port expanding module (3) link to each other with host computer (4), and host computer (4) links to each other with warning device (5) and remote monitoring terminal (6).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010595879 CN102142179A (en) | 2010-12-20 | 2010-12-20 | Vision-based intelligent security system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010595879 CN102142179A (en) | 2010-12-20 | 2010-12-20 | Vision-based intelligent security system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN102142179A true CN102142179A (en) | 2011-08-03 |
Family
ID=44409657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010595879 Pending CN102142179A (en) | 2010-12-20 | 2010-12-20 | Vision-based intelligent security system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102142179A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105336074A (en) * | 2015-10-28 | 2016-02-17 | 小米科技有限责任公司 | Alarm method and device |
| CN106027972A (en) * | 2016-06-17 | 2016-10-12 | 成都易云知科技有限公司 | Real-time image processing system |
| CN109643480A (en) * | 2016-07-22 | 2019-04-16 | 路晟（上海）科技有限公司 | Security system and method |
| CN110927731A (en) * | 2019-11-15 | 2020-03-27 | 深圳市镭神智能系统有限公司 | Three-dimensional protection method, three-dimensional detection device and computer readable storage medium |
| US11412714B2 (en) | 2018-12-28 | 2022-08-16 | Acer Incorporated | Pet monitoring method and pet monitoring system |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040068743A1 (en) * | 2002-10-04 | 2004-04-08 | Parry Travis J. | Systems and methods for providing local broadcast of an event to event attendees |
| CN101246627A (en) * | 2007-12-12 | 2008-08-20 | 广州市聚晖电子科技有限公司 | Security signal alarm system |
-
2010
- 2010-12-20 CN CN 201010595879 patent/CN102142179A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040068743A1 (en) * | 2002-10-04 | 2004-04-08 | Parry Travis J. | Systems and methods for providing local broadcast of an event to event attendees |
| CN101246627A (en) * | 2007-12-12 | 2008-08-20 | 广州市聚晖电子科技有限公司 | Security signal alarm system |
Non-Patent Citations (2)
| Title |
|---|
| 《全国先进制造技术高层论坛暨制造业自动化、信息化技术研讨会论文集》 20051231 陈建造,陈伟 智能安防系统设计 426-428 1 , * |
| 《第十六届电工理论学术年会论文集》 20041231 王长清,金波,吴慎山 基于计算机的数字监控报警系统的开发 139-142 1 , * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105336074A (en) * | 2015-10-28 | 2016-02-17 | 小米科技有限责任公司 | Alarm method and device |
| US10147288B2 (en) | 2015-10-28 | 2018-12-04 | Xiaomi Inc. | Alarm method and device |
| CN106027972A (en) * | 2016-06-17 | 2016-10-12 | 成都易云知科技有限公司 | Real-time image processing system |
| CN109643480A (en) * | 2016-07-22 | 2019-04-16 | 路晟（上海）科技有限公司 | Security system and method |
| US11412714B2 (en) | 2018-12-28 | 2022-08-16 | Acer Incorporated | Pet monitoring method and pet monitoring system |
| CN110927731A (en) * | 2019-11-15 | 2020-03-27 | 深圳市镭神智能系统有限公司 | Three-dimensional protection method, three-dimensional detection device and computer readable storage medium |
| CN110927731B (en) * | 2019-11-15 | 2021-12-17 | 深圳市镭神智能系统有限公司 | Three-dimensional protection method, three-dimensional detection device and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110188724B (en) | Method and system for helmet positioning and color recognition based on deep learning | |
| EP3744084B1 (en) | Method and apparatus of processing a signal from an event-based sensor | |
| CN104680555A (en) | Border-crossing detection method and border-crossing monitoring system based on video monitoring | |
| US10032283B2 (en) | Modification of at least one parameter used by a video processing algorithm for monitoring of a scene | |
| CN107437318B (en) | Visible light intelligent recognition algorithm | |
| CN102142179A (en) | Vision-based intelligent security system | |
| US8411947B2 (en) | Video processing to detect movement of an object in the scene | |
| CA2795896A1 (en) | Method and system for security system tampering detection | |
| WO2005041579A3 (en) | Method and system for processing captured image information in an interactive video display system | |
| CN109886205B (en) | Real-time safety belt monitoring method and system | |
| CN103366483A (en) | Monitoring alarm system | |
| CN103152558B (en) | Based on the intrusion detection method of scene Recognition | |
| CN110703760A (en) | Newly-increased suspicious object detection method for security inspection robot | |
| CN105469054A (en) | Model construction method of normal behaviors and detection method of abnormal behaviors | |
| CN112084813A (en) | Abnormal target detection method and device and storage medium | |
| CN108055449B (en) | Bucket tooth monitoring system based on thermal imaging | |
| KR101454644B1 (en) | Loitering Detection Using a Pedestrian Tracker | |
| CN101930540A (en) | Video-based multi-feature fusion flame detecting device and method | |
| CN115410324A (en) | Car as a house night security system and method based on artificial intelligence | |
| KR101428913B1 (en) | Surveillance system for overheating state utilizing micro-size image array sensor | |
| CN112633157B (en) | Real-time detection method and system for safety of AGV working area | |
| CN103049919A (en) | Embedded target detection algorithm | |
| CN104392201A (en) | Human fall identification method based on omnidirectional visual sense | |
| JP7384158B2 (en) | Image processing device, moving device, method, and program | |
| CN115690914A (en) | Abnormal behavior reminding method and device, electronic equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C12 | Rejection of a patent application after its publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20110803 |