CN104601965B - Camera occlusion detection method - Google Patents

Camera occlusion detection method Download PDF

Info

Publication number
CN104601965B
CN104601965B CN201510065007.6A CN201510065007A CN104601965B CN 104601965 B CN104601965 B CN 104601965B CN 201510065007 A CN201510065007 A CN 201510065007A CN 104601965 B CN104601965 B CN 104601965B
Authority
CN
China
Prior art keywords
camera
blocked
region
msub
mover
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
CN201510065007.6A
Other languages
Chinese (zh)
Other versions
CN104601965A (en
Inventor
张宇
霍晓龙
张韬
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.)
Dalian Jiayun Electronics and Technology Co., Ltd.
Original Assignee
DALIAN JIAYUN ELECTRONICS AND TECHNOLOGY Co Ltd
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 DALIAN JIAYUN ELECTRONICS AND TECHNOLOGY Co Ltd filed Critical DALIAN JIAYUN ELECTRONICS AND TECHNOLOGY Co Ltd
Priority to CN201510065007.6A priority Critical patent/CN104601965B/en
Publication of CN104601965A publication Critical patent/CN104601965A/en
Application granted granted Critical
Publication of CN104601965B publication Critical patent/CN104601965B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Image Analysis (AREA)
  • Studio Devices (AREA)

Abstract

The present invention relates to digital image processing techniques field, and in particular to camera occlusion detection method, step 1：Establish camera resource configuration module, step 2：Surveillance center judges whether camera runs according to running state information, and Surveillance center obtains present frame digital picture, step 3 from the camera in running status：Image preprocessing, step 4：The luminance background of image is established, using mean filter method, by image smoothing into smooth intensity map, may be blocked region S using dynamic thresholding method extraction imagei, step 5：To the region S that may be blockediConnected domain analysis, setting regions threshold value ST, and select >=region threshold STPossibility be blocked region Si, it is defined as high probability and is blocked region S2, step 6：High probability is calculated to be blocked region S2The probability P that is blocked, step 7：Disconnected probability P and the default probability P that is blocked of being blockedTIf P ＞ PT, then judge that the high probability is blocked region as the final region that is blocked, be otherwise determined as de-occlusion region.This method detection efficiency is high.

Description

Camera occlusion detection method
Technical field
The present invention relates to digital image processing techniques field, and in particular to camera occlusion detection method.
Background technology
With widely using for monitoring system network, thousands of individual control points have been laid out largely in monitoring system network Camera, it is for the number of camera, then up to up to ten thousand.Because camera number is huge, the shooting of control point in addition Head is installed on eminence and shelter mostly, and power down can be caused if long-time unmanned maintenance, blurs, block, so as to influence to supervise Control the normal use of point.And intellectualized detection how is carried out to large number of camera, turning into industry pursues a goal, existing skill In art, have and image procossing is carried out to the data for imaging overhead pass by Surveillance center, and according to image procossing testing result, judge Whether camera loses Jiao, if slowly movement causes scene fade, if accumulates a large amount of dusts and causes to block, if flating Deng.
The detection technique for shelter is primarily present following a few point defects in the prior art：Due to shelter have it is a variety of, Someone deliberately carried out by chewing gum, the scraps of paper and cloth etc., also have naturally generate such as dust, mud and spot etc., greatly Small shape differs, and the size of shelter can influence Detection results, can not detect the occlusion area of some shapes and sizes.And ring The conversion of environmental light can be also disturbed testing result.Existing detection technique can also be judged by accident to some unshielding things.
The content of the invention
Solves above-mentioned technical problem, the invention provides a kind of camera occlusion detection method, can quickly judge to take the photograph As whether head is blocked, judgment accuracy is improved, while can also effectively avoid erroneous judgement from misjudging.
In order to achieve the above object, the technical solution adopted in the present invention is a kind of camera occlusion detection method, including Following steps：
Step 1：Camera resource configuration module is established, camera resource configuration module communicates with Surveillance center, described to take the photograph As head resource configuration module includes camera information registration unit and camera running status unit, the camera information registration Unit record camera title and running state information, the camera running status unit are used to detect whether the camera is located In use state, and generating run status information is sent to camera information registration unit,
Step 2：Surveillance center obtains camera title and running state information from camera resource configuration module, according to fortune Row status information judges whether camera runs, and sends a command to the camera in running status, and Surveillance center is from fortune Present frame digital picture is obtained in camera in row state,
Step 3：Image preprocessing, by digital picture binaryzation, gray level image is converted to, and remove using morphological image Tiny black region in gray level image,
Step 4：The luminance background of image is established, using mean filter method, by image smoothing into smooth intensity map, It may be blocked region S using dynamic thresholding method extraction imagei,
Step 5：To the region S that may be blockediConnected domain analysis, setting regions threshold value ST, and select >=region threshold ST's May be blocked region Si, it is defined as high probability and is blocked region S2,
Step 6：High probability is calculated to be blocked region S2The probability P that is blocked,
Step 7：Disconnected probability P and the default probability P that is blocked of being blockedTIf P ＞ PT, then judge that the high probability is blocked region For the final region that is blocked, otherwise it is determined as de-occlusion region.
Further, in step 3, following steps are specifically included：
Step 31：Color image pixel is i (x, y), and gray level image j (x, y) is obtained by binaryzation,
Step 32：Following operation is carried out to gray level image j (x, y) using structural element s (x, y),
J (x, y) is expanded by s (x, y),
J (x, y) is corroded by s (x, y),
J (x, y) is opened with s (x, y),
J (x, y) is closed with s (x, y),
Further, the step 4 specifically includes following steps：
Step 41：G (x, y) all pixels value is normalized between [0,1] first and obtained Wherein Ix,yFor the pixel value of (x, y), ImaxFor 255,
Step 42：CalculateMean flow rate M and variance S2, formula is
Step 43：Calculate dynamic threshold T=M-S2/ C, wherein, C is a constant, and extraction image may be blocked region
The present invention compared with prior art, has the following advantages that by using above-mentioned technical proposal：
The present invention detects whether the camera is in use state, and generate by establishing camera resource configuration module Running state information is sent to camera information registration unit, targetedly can carry out occlusion detection to camera, avoid Poll increases detection time and testing cost, while removes the tiny black region in gray level image by using morphological image Domain, probability of miscarriage of justice can be reduced, algorithm of the invention is easy and effective, judges that speed is fast, and judging efficiency is high.
Embodiment
As a specific embodiment, a kind of camera occlusion detection method of the invention, comprise the following steps：
Step 1：Camera resource configuration module is established, camera resource configuration module communicates with Surveillance center, described to take the photograph As head resource configuration module includes camera information registration unit and camera running status unit, the camera information registration Unit record camera title and running state information, the camera running status unit are used to detect whether the camera is located In use state, and generating run status information is sent to camera information registration unit.Pass through the execution of step 1 so that place Can obtain timely priority treatment in the camera of use state, avoid due to line up poll cause these cameras due to Block and influence use demand.
Step 2：Surveillance center obtains camera title and running state information from camera resource configuration module, according to fortune Row status information judges whether camera runs, and sends a command to the camera in running status, and Surveillance center is from fortune Present frame digital picture is obtained in camera in row state.
Step 3：Image preprocessing, by digital picture binaryzation, gray level image is converted to, and remove using morphological image Tiny black region in gray level image, the interference in these regions can be avoided by removing tiny black region, to testing result Influence.
Further, in step 3, following steps are specifically included：
Step 31：Color image pixel is i (x, y), and gray level image j (x, y) is obtained by binaryzation,
Step 32：Following operation is carried out to gray level image j (x, y) using structural element s (x, y),
J (x, y) is expanded by s (x, y),
J (x, y) is corroded by s (x, y),
J (x, y) is opened with s (x, y),
J (x, y) is closed with s (x, y),
Step 4：The luminance background of image is established, using mean filter method, by image smoothing into smooth intensity map, It may be blocked region S using dynamic thresholding method extraction imagei,
Step 41：G (x, y) all pixels value is normalized between [0,1] first and obtained Wherein Ix,yFor the pixel value of (x, y), ImaxFor 255,
Step 42：CalculateMean flow rate M and variance S2, formula is
Step 43：Calculate dynamic threshold T=M-S2/ C, wherein, C is a constant, and extraction image may be blocked region
Step 5：To the region S that may be blockediCarry out connected domain analysis, setting regions threshold value ST, and select area >=region Threshold value STPossibility be blocked region Si, it is defined as high probability and is blocked region S2, calculation formula is：ST=W*H/B, W are picture Height, H are image width, and B is constant, and B scope is 30-40,
Step 6：High probability is calculated to be blocked region S2The probability P that is blocked, calculation formula is P=Si*C (M*S2),
Step 7：Judge be blocked probability P and the default probability P that is blockedTIf P ＞ PT, then judge that the high probability is blocked area Domain is the final region that is blocked, and is otherwise determined as de-occlusion region.PTChoosing value scope be 85%-99.5%, sentenced by probability It is disconnected, it can effectively avoid erroneous judgement from misjudging, improve detection success rate.
The present invention detects whether the camera is in use state, and generate by establishing camera resource configuration module Running state information is sent to camera information registration unit, targetedly can carry out occlusion detection to camera, avoid Poll increases detection time and testing cost, while removes the tiny black region in gray level image by using morphological image Domain, probability of miscarriage of justice can be reduced, algorithm of the invention is easy and effective, judges that speed is fast, and judging efficiency is high.
Although specifically showing and describing the present invention with reference to preferred embodiment, those skilled in the art should be bright In vain, do not departing from the spirit and scope of the present invention that appended claims are limited, in the form and details can be right The present invention makes a variety of changes, and is protection scope of the present invention.

Claims (1)

  1. A kind of 1. camera occlusion detection method, it is characterised in that：Comprise the following steps：
    Step 1：Camera resource configuration module is established, camera resource configuration module communicates with Surveillance center, the camera Resource configuration module includes camera information registration unit and camera running status unit, the camera information registration unit Whether recording camera title and running state information, the camera running status unit are used to detect the camera and be in make With state, and generating run status information is sent to camera information registration unit,
    Step 2：Surveillance center obtains camera title and running state information from camera resource configuration module, according to operation shape State information judges whether camera runs, and sends a command to the camera in running status, Surveillance center from operation shape Present frame digital picture is obtained in camera in state,
    Step 3：Image preprocessing, by digital picture binaryzation, gray level image is converted to, and gray scale is removed using morphological image Tiny black region in image, is specifically included：
    Step 31：Color image pixel is i (x, y), and gray level image j (x, y) is obtained by binaryzation,
    Step 32：Following operation is carried out to gray level image j (x, y) using structural element s (x, y),
    J (x, y) is expanded by s (x, y),
    J (x, y) is corroded by s (x, y),
    J (x, y) is opened with s (x, y),J (x, y) closes Yong s (x, y) Close,
    Step 4：The luminance background of image is established, using mean filter method, by image smoothing into smooth intensity map, is used Dynamic thresholding method extraction image may be blocked region Si,
    Step 41：G (x, y) all pixels value is normalized between [0,1] first and obtainedWherein Ix,yFor the pixel value of (x, y), ImaxFor 255,
    Step 42：CalculateMean flow rate M and variance S2, formula is
    <mrow> <mi>M</mi> <mo>=</mo> <mfrac> <mrow> <msub> <mover> <mi>g</mi> <mo>&amp;OverBar;</mo> </mover> <mn>1</mn> </msub> <mo>+</mo> <msub> <mover> <mi>g</mi> <mo>&amp;OverBar;</mo> </mover> <mn>2</mn> </msub> <mo>+</mo> <msub> <mover> <mi>g</mi> <mo>&amp;OverBar;</mo> </mover> <mn>3</mn> </msub> <mo>+</mo> <msub> <mover> <mi>g</mi> <mo>&amp;OverBar;</mo> </mover> <mn>4</mn> </msub> <mo>+</mo> <mn>...</mn> <mo>+</mo> <mo>+</mo> <msub> <mover> <mi>g</mi> <mo>&amp;OverBar;</mo> </mover> <mi>n</mi> </msub> </mrow> <mi>n</mi> </mfrac> <mo>,</mo> <msup> <mi>S</mi> <mn>2</mn> </msup> <mo>=</mo> <mfrac> <mrow> <msub> <mover> <mrow> <mo>(</mo> <mi>g</mi> </mrow> <mo>&amp;OverBar;</mo> </mover> <mn>1</mn> </msub> <mo>-</mo> <mi>M</mi> <msup> <mo>)</mo> <mn>2</mn> </msup> <mo>+</mo> <msub> <mover> <mrow> <mo>(</mo> <mi>g</mi> </mrow> <mo>&amp;OverBar;</mo> </mover> <mn>2</mn> </msub> <mo>-</mo> <mi>M</mi> <msup> <mo>)</mo> <mn>2</mn> </msup> <mo>+</mo> <msub> <mover> <mrow> <mo>(</mo> <mi>g</mi> </mrow> <mo>&amp;OverBar;</mo> </mover> <mn>3</mn> </msub> <mo>-</mo> <mi>M</mi> <msup> <mo>)</mo> <mn>2</mn> </msup> <mo>+</mo> <mn>...</mn> <mo>+</mo> <msub> <mover> <mrow> <mo>(</mo> <mi>g</mi> </mrow> <mo>&amp;OverBar;</mo> </mover> <mi>n</mi> </msub> <mo>-</mo> <mi>M</mi> <msup> <mo>)</mo> <mn>2</mn> </msup> </mrow> <mi>n</mi> </mfrac> <mo>,</mo> </mrow>
    Step 43：Calculate dynamic threshold T=M-S2/ C, wherein, C is a constant, and extraction image may be blocked region
    Step 5：To the regional connectivity domain analysis that may be blocked, setting regions threshold value ST, and select >=region threshold STPossibility quilt Occlusion area, it is defined as high probability and is blocked region S2,
    Step 6：High probability is calculated to be blocked region S2The probability P that is blocked,
    Step 7：Judge be blocked probability P and the default probability P that is blockedTIf P ＞ PT, then judge that the high probability is blocked region S2 For the final region that is blocked, otherwise it is determined as de-occlusion region.
CN201510065007.6A 2015-02-06 2015-02-06 Camera occlusion detection method Active CN104601965B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510065007.6A CN104601965B (en) 2015-02-06 2015-02-06 Camera occlusion detection method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510065007.6A CN104601965B (en) 2015-02-06 2015-02-06 Camera occlusion detection method

Publications (2)

Publication Number Publication Date
CN104601965A CN104601965A (en) 2015-05-06
CN104601965B true CN104601965B (en) 2018-01-16

Family

ID=53127425

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510065007.6A Active CN104601965B (en) 2015-02-06 2015-02-06 Camera occlusion detection method

Country Status (1)

Country Link
CN (1) CN104601965B (en)

Families Citing this family (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105611188B (en) * 2015-12-23 2019-06-04 北京奇虎科技有限公司 Camera occlusion detection method and device based on automatic exposure
CN105744268A (en) * 2016-05-04 2016-07-06 深圳众思科技有限公司 Camera shielding detection method and device
CN109120916B (en) * 2017-06-22 2020-06-05 杭州海康威视数字技术股份有限公司 Camera fault detection method and device and computer equipment
CN110059592B (en) * 2019-03-29 2021-01-08 浙江中控太阳能技术有限公司 Multi-camera-based mirror field cloud blocking detection method and device
JP7362284B2 (en) * 2019-03-29 2023-10-17 キヤノン株式会社 Image processing method, image processing device, program, image processing system, and learned model manufacturing method
CN110321819B (en) * 2019-06-21 2021-09-14 浙江大华技术股份有限公司 Shielding detection method and device of camera equipment and storage device
CN110544211B (en) * 2019-07-26 2024-02-09 纵目科技（上海）股份有限公司 Method, system, terminal and storage medium for detecting lens attached object
CN112399220B (en) * 2019-08-18 2022-10-28 海信视像科技股份有限公司 Camera physical switch locking state display method and display equipment
CN110913209B (en) * 2019-12-05 2021-06-04 杭州飞步科技有限公司 Camera shielding detection method and device, electronic equipment and monitoring system
CN113096059B (en) * 2019-12-19 2023-10-31 合肥君正科技有限公司 Method for eliminating interference shielding detection of night light source by in-vehicle monitoring camera
CN113011216B (en) * 2019-12-19 2024-04-02 合肥君正科技有限公司 Multi-classification threshold self-adaptive shielding detection method
CN113177944B (en) * 2021-06-30 2021-09-17 深之蓝海洋科技股份有限公司 Underwater lens stain detection method and underwater robot

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1758752A (en) * 2005-06-01 2006-04-12 北京中星微电子有限公司 Method for monitoring state of network camera head
CN101101580A (en) * 2007-08-06 2008-01-09 北京中星微电子有限公司 Information prompt method during pick-up head utilization and computer device
CN102176244A (en) * 2011-02-17 2011-09-07 东方网力科技股份有限公司 Method and device for determining shielding condition of camera head
CN203708376U (en) * 2013-12-02 2014-07-09 长春大京科技有限公司 A monitor mechanism used for detecting working states of a video device

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2005048088A1 (en) * 2003-11-13 2005-05-26 Koninklijke Philips Electronics, N.V. Method and apparatus for theft protection for devices in a network
CN201750476U (en) * 2010-01-14 2011-02-16 高国钧 Mobile-phone real-time remote-monitoring device
JP5241782B2 (en) * 2010-07-30 2013-07-17 株式会社日立製作所 Surveillance camera system having a camera abnormality detection device
CN103152601A (en) * 2013-03-15 2013-06-12 广州市澳视光电子技术有限公司 Intelligent failure-reporting camera and network management client system thereof

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1758752A (en) * 2005-06-01 2006-04-12 北京中星微电子有限公司 Method for monitoring state of network camera head
CN101101580A (en) * 2007-08-06 2008-01-09 北京中星微电子有限公司 Information prompt method during pick-up head utilization and computer device
CN102176244A (en) * 2011-02-17 2011-09-07 东方网力科技股份有限公司 Method and device for determining shielding condition of camera head
CN203708376U (en) * 2013-12-02 2014-07-09 长春大京科技有限公司 A monitor mechanism used for detecting working states of a video device

Also Published As

Publication number Publication date
CN104601965A (en) 2015-05-06

Similar Documents

Publication Publication Date Title
CN104601965B (en) Camera occlusion detection method
CN105744232B (en) A kind of method of the transmission line of electricity video external force damage prevention of Behavior-based control analytical technology
CN101739550B (en) Method and system for detecting moving objects
CN101236606B (en) Shadow cancelling method and system in vision frequency monitoring
CN102348128B (en) Surveillance camera system having camera malfunction detection function
TWI475524B (en) System and method for inspection of cars that violate traffic regulations using images
CN103235938B (en) The method and system of car plate detection and indentification
CN102968625B (en) Ship distinguishing and tracking method based on trail
WO2022027931A1 (en) Video image-based foreground detection method for vehicle in motion
CN111723644A (en) Method and system for detecting occlusion of surveillance video
CN103226712B (en) A kind of remnant object detection method based on finite state machine
CN112528861B (en) Foreign matter detection method and device applied to ballast bed in railway tunnel
CN109871752A (en) A method of lane line is extracted based on monitor video detection wagon flow
CN109359593B (en) Rain and snow environment picture fuzzy monitoring and early warning method based on image local grid
KR102006763B1 (en) Method and System for Analyzing Road Congestion Using CCTV
CN103971135A (en) Human body target detection method based on head and shoulder depth information features
CN103400113A (en) Method for detecting pedestrian on expressway or in tunnel based on image processing
CN104662560A (en) Method and system for processing video image
CN110880184A (en) Method and device for carrying out automatic camera inspection based on optical flow field
Su et al. A new local-main-gradient-orientation HOG and contour differences based algorithm for object classification
CN103021179A (en) Real-time monitoring video based safety belt detection method
CN109325426B (en) Black smoke vehicle detection method based on three orthogonal planes time-space characteristics
CN103489202A (en) Intrusion detection method based on videos
CN114708532A (en) Monitoring video quality evaluation method, system and storage medium
CN104573657B (en) It is a kind of that detection method is driven based on the blind of feature of bowing

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
TA01 Transfer of patent application right

Effective date of registration: 20171122

Address after: 100086 Beijing city Haidian District North Sanhuan Road 43, Tsing Wun contemporary building 12A11

Applicant after: BEIJING ZHITAO SCIENCE & TECHNOLOGY CO., LTD.

Address before: 528000 Guangdong City, Chancheng Province, China Road, No. four quarter, No. 33, creative industrial park, building 5, floor 9, Foshan Mobile

Applicant before: Wu Libin

TA01 Transfer of patent application right
CB03 Change of inventor or designer information

Inventor after: Zhang Yu

Inventor after: Huo Xiaolong

Inventor after: Zhang Tao

Inventor before: Wu Libin

CB03 Change of inventor or designer information
TA01 Transfer of patent application right

Effective date of registration: 20171222

Address after: No. 2, North China Road, 427-5 two, Ganjingzi District, Dalian, Liaoning

Applicant after: Dalian Jiayun Electronics and Technology Co., Ltd.

Address before: 100086 Beijing city Haidian District North Sanhuan Road 43, Tsing Wun contemporary building 12A11

Applicant before: BEIJING ZHITAO SCIENCE & TECHNOLOGY CO., LTD.

TA01 Transfer of patent application right
GR01 Patent grant
GR01 Patent grant