CN108460357B - Windowing alarm detection system and method based on image recognition - Google Patents

Windowing alarm detection system and method based on image recognition Download PDF

Info

Publication number
CN108460357B
CN108460357B CN201810209075.9A CN201810209075A CN108460357B CN 108460357 B CN108460357 B CN 108460357B CN 201810209075 A CN201810209075 A CN 201810209075A CN 108460357 B CN108460357 B CN 108460357B
Authority
CN
China
Prior art keywords
windowing
image
gaussian distribution
time
alarm
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.)
Active
Application number
CN201810209075.9A
Other languages
Chinese (zh)
Other versions
CN108460357A (en
Inventor
王渤
孙秀斌
赵伟
杨金宝
杨晨
刘亚超
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Public Security Bureau Guard Bureau
Beijing Institute of Environmental Features
Original Assignee
Beijing Public Security Bureau Guard Bureau
Beijing Institute of Environmental Features
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Public Security Bureau Guard Bureau, Beijing Institute of Environmental Features filed Critical Beijing Public Security Bureau Guard Bureau
Priority to CN201810209075.9A priority Critical patent/CN108460357B/en
Publication of CN108460357A publication Critical patent/CN108460357A/en
Application granted granted Critical
Publication of CN108460357B publication Critical patent/CN108460357B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B5/00—Visible signalling systems, e.g. personal calling systems, remote indication of seats occupied
    • G08B5/22—Visible signalling systems, e.g. personal calling systems, remote indication of seats occupied using electric transmission; using electromagnetic transmission

Landscapes

  • Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Electromagnetism (AREA)
  • Alarm Systems (AREA)
  • Image Analysis (AREA)
  • Closed-Circuit Television Systems (AREA)

Abstract

The invention relates to a windowing alarm detection system and method based on image recognition, wherein the system comprises: the image acquisition module is used for acquiring a video image containing a window; the image preprocessing module is used for performing color space conversion on the video image and converting the video image from RGB into HSI image; the windowing detection module is used for determining a windowing alarm position according to brightness component image data in the HSI image; and the display control module is used for displaying the windowing alarm position. According to the invention, the collected video image is converted into the HSI image, and the brightness component image data in the HSI image is identified, so that the detection and alarm positioning of the window opening behavior of the building can be realized, and the effective monitoring and danger early warning of the buildings around the key area can be realized.

Description

Windowing alarm detection system and method based on image recognition
Technical Field
The invention relates to the field of computer image detection, in particular to a windowing alarm detection system and method based on image recognition.
Background
With the increasing maturity of video monitoring technology, the monitoring of major activity security site presents the characteristics of many monitoring points, wide distribution, and complex communication of a plurality of security units and mobile personnel on site. The monitoring of the outdoor surrounding building facades by the activity site is particularly important, and especially the monitoring and timely processing of dangerous behaviors such as window opening and object throwing become problems which need to be solved urgently in security work.
Therefore, it is desirable to develop a system and method capable of rapidly detecting window opening behavior and position.
Disclosure of Invention
The technical problem to be solved by the present invention is to provide a windowing alarm detection system and method based on image recognition, aiming at the above defects in the prior art.
In order to solve the technical problem, the invention provides a windowing alarm detection system based on image recognition, which comprises:
the image acquisition module is used for acquiring a video image containing a window;
the image preprocessing module is used for performing color space conversion on the video image and converting the video image from RGB into HSI image;
the windowing detection module is used for determining a windowing alarm position according to brightness component image data in the HSI image;
and the display control module is used for displaying the windowing alarm position.
In the windowing alarm detecting system based on image recognition according to the present invention, the windowing detecting module comprises:
the foreground extraction unit is used for judging whether the pixel belongs to the background or the foreground and extracting the length and the width of the foreground area;
and the size judging unit is used for judging whether the size of the foreground area meets a preset range or not, and if so, judging the foreground area to be a windowing behavior.
In the windowing alarm detection system based on image recognition according to the invention, the display control module is preferably further configured to input the pixel length L of the image to be detected0And width W0To the size determination unit.
In the windowing alarm detection system based on image recognition according to the present invention, preferably, the display control module is further configured to receive floor information of a video image labeled by a user, and generate parameter coordinate information of the floor; the windowing detection module also comprises a floor determination unit used for determining windowing floor information according to the windowing alarm position and the parameter coordinate information of the floor; the display control module is also used for displaying the windowing floor information.
In the windowing alarm detection system based on image recognition according to the present invention, preferably, the windowing alarm detection system further includes an alarm module, configured to alarm according to a windowing behavior detection result of the windowing detection module.
The invention also provides a windowing alarm detection system based on image recognition, which comprises:
the camera is used for acquiring a video image containing a window;
and the comprehensive display control device is used for performing color space conversion on the video image, converting RGB into an HSI image, determining a windowing alarm position according to brightness component image data in the HSI image, and displaying the windowing alarm position to a user.
The invention also provides a windowing alarm detection method based on image recognition, which comprises the following steps:
acquiring a video image containing a window;
performing color space conversion on the video image, and converting the video image from RGB to HSI image;
determining a windowing alarm position according to brightness component image data in the HSI image;
and displaying the windowing alarm position.
In the windowing alarm detection method based on image recognition according to the present invention, the determining a windowing alarm position according to luminance component image data in an HSI image comprises:
a foreground extraction substep, which is used for judging whether the pixel belongs to the background or the foreground and extracting the length and the width of a foreground area;
and a size judgment substep, judging whether the size of the foreground region meets a preset range, and judging the foreground region to be a windowing behavior if the size of the foreground region meets the preset range.
In the windowing alarm detecting method based on image recognition according to the present invention, preferably, the method further comprises: inputting pixel length L of image to be detected0And width W0A supply size determination substep.
In the windowing alarm detecting method based on image recognition according to the present invention, preferably, the method further comprises: receiving floor information of a video image marked by a user and generating parameter coordinate information of the floor; determining windowing floor information according to the windowing alarm position and the parameter coordinate information of the floor; and displaying the windowing floor information.
The windowing alarm detection system and method based on image recognition have the following beneficial effects: according to the invention, the collected video image is converted into the HSI image, and the brightness component image data in the HSI image is identified to determine the windowing alarm position, so that the detection and alarm positioning of the windowing behavior of the building can be realized, and the effective monitoring and danger early warning of the buildings around the key area can be realized.
Drawings
FIG. 1 is a block diagram of a windowing alarm detection system based on image recognition in accordance with a preferred embodiment of the present invention;
FIG. 2 is a block diagram of a windowing detection module in an image recognition-based windowing alarm detection system in accordance with a preferred embodiment of the present invention;
FIG. 3 is a diagram of a windowing alarm detection interface based on image recognition in accordance with a preferred embodiment of the present invention;
FIG. 4 is a diagram of a windowing alarm result display interface based on image recognition in accordance with a preferred embodiment of the present invention;
FIG. 5 is a hardware block diagram of a windowing alarm detection system based on image recognition according to the present invention;
fig. 6 is a flowchart of a windowing alarm detection method based on image recognition according to a preferred embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
Fig. 1 is a block diagram of a windowing alarm detecting system based on image recognition according to a preferred embodiment of the present invention. As shown in fig. 1, this embodiment provides a system including: the system comprises an image acquisition module 101, an image preprocessing module 102, a windowing detection module 103 and a display control module 104.
The image capture module 101 is used to capture video images containing windows.
The image preprocessing module 102 is configured to perform color space conversion on the video image from RGB to an H (hue) S (saturation) I (brightness) image. And passes the I-channel image component to the windowing detection module 103.
The windowing detection module 103 is configured to determine a windowing alert location according to luminance component image data in the HSI image.
The display control module 104 is used for displaying the windowing alarm position.
Fig. 2 is a block diagram of a windowing detection module in a windowing alarm detection system based on image recognition according to a preferred embodiment of the present invention. As shown in fig. 2, preferably, the windowing detection module 103 includes: a foreground extraction unit 201 and a size determination unit 202.
The detection of the windowing alarm is completed by the windowing detection module 103, which comprises the following specific steps: the foreground extraction unit 201 models each pixel point of the image by overlapping a plurality of gaussian distributions with different weights, each gaussian distribution corresponds to a state that may generate the brightness presented by the pixel point, and the weights and distribution parameters of each gaussian distribution are updated with time.
Observation data set { X for random variable X1,x2,…,xN}，xtT is more than or equal to 1 and less than or equal to N, wherein N is the total number of collected samples of the pixel; then a single sample point xtIts obeyed mixture gaussian distribution probability density function:
Figure BDA0001596693290000051
Figure BDA0001596693290000052
Figure BDA0001596693290000053
where k is the total number of Gaussian distribution patterns, η (x)t,μi,t,τi,t) Is the ith Gaussian distribution at time ti,tIs the mean value ofi,tIs its covariance matrix, δi,tIs the variance, I is the identity matrix, ωi,tThe weight of the ith gaussian distribution at time t.
The foreground extraction unit 201 is used to extract each new pixel value x by the following formula (4)tAnd comparing the current k models according to the following formula so as to judge whether the pixel belongs to the background or the foreground:
|xt-μi,t-1|≤2.5δi,t-1(4)
wherein xtIs the luminance sample of the pixel at time t, { x }1,x2,…,xN}，μi,t-1Is the mean value of the ith Gaussian distribution at time t-1, deltai,t-1Is the variance of the ith gaussian distribution at time t-1.
When the above formula (4) is satisfied, the pixel belongs to the foreground, otherwise the pixel belongs to the background. The length L and the width W of the foreground area are extracted through morphological processing such as expansion corrosion on the foreground.
The size determining unit 202 is configured to determine whether the size of the foreground region satisfies a preset range according to the following formula (5), and if so, determine that the foreground region is a windowing behavior:
Figure BDA0001596693290000054
wherein L and W are respectively the length and width of the extracted foreground region, LcAnd WcRespectively is a preset length lower limit value and a preset width lower limit value; l is0And W0Respectively the preset pixel length L of the window image to be detected0And width W0. Wherein L may be usedc10 and WcThe foreground below 10 is the noise interference rather than the windowing behavior, and the lower length limit and the lower width limit may be modified according to the detection situation. And when the windowing foreground meets the conditions, detecting the windowing behavior.
Preferably, the display control module 104 is used for displaying and controlling video images, including possible cradle head control and the like, automatically or manually aligning the camera with the facade of the building to be observed, selecting and confirming the actual observation range, operating the mouse to draw the actual observation area, marking floor information at the corresponding position of the image, and inputting the pixel length L of the window image to be detected0And width W0And the like. The windowing alarm detection interface based on image recognition of the present invention is shown in fig. 3. Preferably, the display control module 104 is used for inputting the pixel length L of the image to be detected by the user0And width W0To the size determination unit 202.
In a more preferred embodiment of the present invention, the display control module 104 is further configured to receive floor information of the video image labeled by the user, and generate parameter coordinate information of the floor. The windowing detection module 103 further comprises a floor determination unit for determining windowing floor information according to the windowing alarm position and the parameter coordinate information of the floor. The display control module 104 is further configured to display windowing floor information. That is, the windowing detection module 103 further matches the parameter coordinate information of the floor, which is initially generated by the display control module 104, according to the image coordinate of the windowing alarm position, thereby positioning the alarm floor. And finally, transmitting the image coordinate of the windowing alarm position and the windowing floor information to the display control module 104 for display. As shown in fig. 4, the windowing position is displayed in the captured video image by a wire frame highlighting the color, and the windowed floor information can also be displayed in the form of numbers or the like.
Preferably, the windowing alarm detection system of the present invention further comprises an alarm module, configured to alarm according to a detection result of the windowing behavior of the windowing detection module.
Please refer to fig. 5, which is a hardware structure diagram of the windowing alarm detecting system based on image recognition according to the present invention. As shown in fig. 5, the windowing alarm detecting system based on image recognition of the present invention mainly includes a camera 501 and a comprehensive display control device 503 communicating with the camera 501. Preferably, the system also includes a network switch 502.
The camera 501 is used to capture video images of windows, for example, a high-definition camera is used. The network switch 502 transmits and gathers the images collected by the camera 501 to the network switch 502 through the network, and finally the images are connected with the integrated display and control device 503. The image capturing module 101 can capture a video image containing a window through the camera 501.
The comprehensive display control device 503 is configured to perform color space conversion on the video image acquired by the camera 501, convert RGB into an HSI image, determine a windowing alarm position according to luminance component image data in the HSI image, and display the windowing alarm position to a user. Preferably, the upper computer display and control software deployed in the integrated display and control device 3 completes display and control of the video image, including possible pan-tilt control and the like, automatically or manually aligns the camera to the outer facade of the building to be observed, selects and confirms the actual observation range, operates the mouse to draw the actual observation area, marks floor information at the corresponding position of the image, and inputs the pixel length L of the window image to be detected0And width W0. The image preprocessing module 102, the windowing detection module 103 and the display control module 104 can be implemented by the integrated display control device 503.
Referring to fig. 6, a flowchart of a windowing alarm detection method based on image recognition according to a preferred embodiment of the invention is shown. As shown in fig. 6, the method provided by this embodiment includes the following steps:
first, in step S1, a video image containing a window is captured;
subsequently, in step S2, the video image is color space converted from RGB to HSI image;
subsequently, in step S3, a windowing alert position is determined from the luminance component image data in the HSI image;
finally, in step S4, the windowing alert location is displayed. Preferably, the windowed alert location is indicated in the originally captured video image by a highlighted colored wire frame.
Preferably, the determining the windowing alert location from the luminance component image data in the HSI image in step S3 includes:
1) a foreground extraction substep, which judges whether the pixel belongs to the background or the foreground through the following formula and extracts the length and the width of a foreground area;
|xt-μi,t-1|≤2.5i,t-1
wherein xtIs the luminance sample of the pixel at time t, { x }1,x2,…,xN}，xtT is more than or equal to 1 and less than or equal to N, wherein N is the total number of collected samples of the pixel; mu.si,t-1Is the mean value of the ith Gaussian distribution at time t-1, deltai,t-1Is the variance of the ith Gaussian distribution at the time t-1; wherein the observation data set { X for the random variable X1,x2,…,xN}，xtFor the luminance sample of the pixel at time t, then a single sample point xtThe obeyed mixed gaussian distribution probability density function is:
Figure BDA0001596693290000081
Figure BDA0001596693290000082
Figure BDA0001596693290000083
where k is the total number of Gaussian distribution patterns, η (x)t,μi,t,τi,t) Is the ith Gaussian distribution at time ti,tIs the mean value ofi,tIs its covariance matrix, δi,tIs the variance, I is the identity matrix, ωi,tThe weight of the ith Gaussian distribution at the time t;
when the formula is satisfied, the pixel belongs to the foreground, otherwise, the pixel belongs to the background;
2) and a size judgment substep, judging whether the size of the foreground region meets a preset range through the following formula, if so, judging the window-opening behavior:
Figure BDA0001596693290000084
wherein L and W are respectively the length and width of the extracted foreground region, LcAnd WcRespectively is a preset length lower limit value and a preset width lower limit value; l is0And W0Respectively the preset pixel length L of the window image to be detected0And width W0。
Preferably, in the windowing alarm detecting method based on image recognition, the method further includes: inputting pixel length L of image to be detected0And width W0A supply size determination substep.
More preferably, the method further comprises: receiving floor information of a video image marked by a user and generating parameter coordinate information of the floor; then, determining windowing floor information according to the windowing alarm position and the parameter coordinate information of the floor; and displaying the windowing floor information.
Preferably, the method further comprises: and alarming according to the window opening behavior detection result of the window opening detection module.
In conclusion, based on high-definition video image acquisition and computer image recognition technology, the window opening alarm detection system is constructed, so that window opening abnormal behaviors of the outer vertical surface of the building can be effectively and accurately detected, positioning and alarming are carried out, effective monitoring and danger early warning of buildings around a key area are realized, safety guard work of important activities is greatly guaranteed, and the system and the method particularly make a remarkable contribution in the aspects of saving police strength and the like. Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.

Claims (7)

1. An image recognition-based windowing alarm detection system, comprising:
the image acquisition module is used for acquiring a video image containing a window;
the image preprocessing module is used for performing color space conversion on the video image and converting the video image from RGB into HSI image;
the windowing detection module is used for determining a windowing alarm position according to brightness component image data in the HSI image;
the display control module is used for displaying the windowing alarm position;
the windowing detection module comprises:
the foreground extracting unit is used for judging whether the pixel belongs to the background or the foreground through the following formula and extracting the length and the width of the foreground area;
|xt-μi，t-1|≤2.5δi，t-1
wherein xtIs a luminance sample of the pixel at time t, mui，t-1Is the mean value of the ith Gaussian distribution at time t-1, deltai，t-1Is the variance of the ith Gaussian distribution at the time t-1; wherein the observation data set { X for the random variable X1,x2,…,xN}，xtFor the luminance sample of the pixel at time t, then a single sample point xtThe obeyed mixed gaussian distribution probability density function is:
Figure FDA0002399307770000011
Figure FDA0002399307770000012
Figure FDA0002399307770000013
where k is the total number of Gaussian distribution patterns, η (x)t,μi,t,τi,t) Is the ith Gaussian distribution at time ti,tIs the mean value of the ith Gaussian distribution at time t, τi,tCovariance matrix, delta, of ith Gaussian distribution at time ti,tIs the variance, I is the identity matrix, ωi,tThe weight of the ith Gaussian distribution at the time t;
when the formula is satisfied, the pixel belongs to the foreground, otherwise, the pixel belongs to the background;
the size judging unit is used for judging whether the size of the foreground area meets a preset range through the following formula, and if so, the window opening behavior is judged:
Figure FDA0002399307770000021
wherein L and W are respectively the length and width of the extracted foreground region, LcAnd WcRespectively is a preset length lower limit value and a preset width lower limit value; l is0And W0Respectively the preset pixel length L of the window image to be detected0And width W0。
2. The image recognition-based windowing alarm detection system according to claim 1, wherein said display control module is further configured to input a pixel length L of an image to be detected0And width W0To the size determination unit.
3. The image recognition based windowing alarm detection system according to claim 1 or 2, wherein:
the display control module is also used for receiving floor information of the video image marked by the user and generating parameter coordinate information of the floor;
the windowing detection module also comprises a floor determination unit used for determining windowing floor information according to the windowing alarm position and the parameter coordinate information of the floor;
the display control module is also used for displaying the windowing floor information.
4. The image recognition based windowing alarm detection system according to claim 1 or 2, wherein: the windowing alarm detection system further comprises an alarm module used for giving an alarm according to the windowing behavior detection result of the windowing detection module.
5. A windowing alarm detection method based on image recognition is characterized by comprising the following steps:
acquiring a video image containing a window;
performing color space conversion on the video image, and converting the video image from RGB to HSI image;
determining a windowing alarm position according to brightness component image data in the HSI image;
displaying a windowing alarm position;
the determining the windowing alarm position according to the brightness component image data in the HSI image comprises:
a foreground extraction substep, which judges whether the pixel belongs to the background or the foreground through the following formula and extracts the length and the width of a foreground area;
|xt-μi，t-1|≤2.5δi，t-1
wherein xtIs a luminance sample of the pixel at time t, mui，t-1Is the mean value of the ith Gaussian distribution at time t-1, deltai，t-1Is the variance of the ith Gaussian distribution at the time t-1; wherein the observation data set { X for the random variable X1,x2,…,xN}，xtFor the luminance sample of the pixel at time t, then a single sample point xtThe obeyed mixed gaussian distribution probability density function is:
Figure FDA0002399307770000031
Figure FDA0002399307770000032
Figure FDA0002399307770000033
where k is the total number of Gaussian distribution patterns, η (x)t,μi,t,τi,t) Is the ith Gaussian distribution at time ti,tIs the mean value of the ith Gaussian distribution at time t, τi,tCovariance matrix, delta, of ith Gaussian distribution at time ti,tIs the variance, I is the identity matrix, ωi,tThe weight of the ith Gaussian distribution at the time t;
when the formula is satisfied, the pixel belongs to the foreground, otherwise, the pixel belongs to the background;
and a size judgment substep, judging whether the size of the foreground region meets a preset range through the following formula, if so, judging the window-opening behavior:
Figure FDA0002399307770000034
wherein L and W are respectively the length and width of the extracted foreground region, LcAnd WcRespectively is a preset length lower limit value and a preset width lower limit value; l is0And W0Respectively the preset pixel length L of the window image to be detected0And width W0。
6. The image recognition-based windowing alarm detection method of claim 5, further comprising: inputting pixel length L of image to be detected0And width W0A supply size determination substep.
7. The image recognition-based windowing alarm detection method according to claim 5 or 6, further comprising:
receiving floor information of a video image marked by a user and generating parameter coordinate information of the floor;
determining windowing floor information according to the windowing alarm position and the parameter coordinate information of the floor;
and displaying the windowing floor information.
CN201810209075.9A 2018-03-14 2018-03-14 Windowing alarm detection system and method based on image recognition Active CN108460357B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810209075.9A CN108460357B (en) 2018-03-14 2018-03-14 Windowing alarm detection system and method based on image recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810209075.9A CN108460357B (en) 2018-03-14 2018-03-14 Windowing alarm detection system and method based on image recognition

Publications (2)

Publication Number Publication Date
CN108460357A CN108460357A (en) 2018-08-28
CN108460357B true CN108460357B (en) 2020-05-05

Family

ID=63219967

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810209075.9A Active CN108460357B (en) 2018-03-14 2018-03-14 Windowing alarm detection system and method based on image recognition

Country Status (1)

Country Link
CN (1) CN108460357B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111369761B (en) * 2019-03-05 2022-08-05 杭州海康威视系统技术有限公司 Early warning range determining method, device, equipment and system
CN110969604B (en) * 2019-11-26 2024-02-27 北京工业大学 Intelligent security real-time windowing detection alarm system and method based on deep learning
CN111726532B (en) * 2020-06-30 2021-08-27 北京环境特性研究所 Windowing alarm detection system and method

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102073854A (en) * 2011-01-14 2011-05-25 天津大学 Color license plate positioning method
CN105163067B (en) * 2015-08-12 2018-12-14 武汉大学 A kind of throwing object in high sky evidence-obtaining system based on digital image processing techniques

Also Published As

Publication number Publication date
CN108460357A (en) 2018-08-28

Similar Documents

Publication Publication Date Title
CN109484935B (en) Elevator car monitoring method, device and system
CN108460357B (en) Windowing alarm detection system and method based on image recognition
CN112396658B (en) Indoor personnel positioning method and system based on video
CN108389359B (en) Deep learning-based urban fire alarm method
CN110738135A (en) worker work step specification visual identification judgment and guidance method and system
CN103294994A (en) Apparatus for recognizing face based on environment adaptation
CN113903081A (en) Visual identification artificial intelligence alarm method and device for images of hydraulic power plant
CN108734913A (en) Forest fire patrol alarm system based on unmanned plane image recognition and method
CN112184773A (en) Helmet wearing detection method and system based on deep learning
CN110493574B (en) Security monitoring visualization system based on streaming media and AI technology
RU2713876C1 (en) Method and system for detecting alarm events when interacting with self-service device
CN113076859A (en) Safety monitoring method and system for face recognition, electronic equipment and storage medium
CN115620192A (en) Method and device for detecting wearing of safety rope in aerial work
CN112595730A (en) Cable breakage identification method and device and computer equipment
CN112614260A (en) Intelligent security system based on face recognition and positioning
CN113361420A (en) Mine fire monitoring method, device and equipment based on robot and storage medium
CN113808087A (en) Defect management and control method and device for surface of steel plate and computer readable storage medium
CN114943841A (en) Method and device for assisting operation safety control based on image recognition
CN109579934A (en) Water logging multiple monitoring method and system
CN113627321A (en) Image identification method and device based on artificial intelligence and computer equipment
KR20160003996A (en) Apparatus and method for video analytics
CN105979230A (en) Monitoring method and device realized through images by use of robot
CN113112427B (en) Monitoring method, device, equipment and storage medium for molten iron transportation
CN115346170A (en) Intelligent monitoring method and device for gas facility area
CN113225525A (en) Indoor monitoring method and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant