CA2860014C - Video based indoor leak detection - Google Patents
Video based indoor leak detection Download PDFInfo
- Publication number
- CA2860014C CA2860014C CA2860014A CA2860014A CA2860014C CA 2860014 C CA2860014 C CA 2860014C CA 2860014 A CA2860014 A CA 2860014A CA 2860014 A CA2860014 A CA 2860014A CA 2860014 C CA2860014 C CA 2860014C
- Authority
- CA
- Canada
- Prior art keywords
- video
- color
- leak
- image analyzer
- pixel values
- 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
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M3/00—Investigating fluid-tightness of structures
- G01M3/38—Investigating fluid-tightness of structures by using light
Abstract
Apparatus and corresponding method are invented for detecting indoor liquid leak in equipment, such as a pump in a pump station. The apparatus includes an image analyzer configured to receive video from one or more cameras and analyze the video to determine if a leak on the equipment is present by comparing the video with a reference and a target color and the image analyzer having an output port for delivering a notification upon the image analyzer determining that a leak is present on the equipment. If a leak alarm happens, the operator can check the remote live video and/or stored clips to confirm the leak and activate a response to stop it.
Description
VIDEO BASED INDOOR LEAK DETECTION
TECHNICAL FIELD
[0001] Indoor leak detection.
BACKGROUND
TECHNICAL FIELD
[0001] Indoor leak detection.
BACKGROUND
[0002] Indoor liquid leak detection is very important for many facilities, such as home water tank, gasoline pump station, etc. Detecting a leak and generating an alert can avoid the damage of the facility, save money and protect the environment.
SUMMARY
SUMMARY
[0003] An indoor liquid leak detection system for the detection of a leak from a gasoline pump station is disclosed. The system may be used for other liquid containers and carriers. The system makes use of color video camera(s) to monitor the pumps. An alarm is activated when a leak is detected over a predetermined time interval. The invention can be used for industrial applications where the detection of leaking liquids is desirable to prevent environmental damage or other unwanted results. The invented system in some embodiments may be easy to install, not disturb the equipment's operation, and not change the equipment's architecture.
[0004] In an embodiment, there is provided a video-based indoor leak detection system.
The detection system comprises a video capturing device (for example a video camera) having a field of view; an image analyzer configured to analyze the video to determine if a leak on the equipment is present; and an output port for delivering a notification upon the image analyzer determining that a leak is present. Further embodiments may also include one or more of the following features: a video recording unit for recording events, for example video corresponding to an alarm, which may be stored according to the time and the classification for further reference; a server may comprise the image analyzer and video recording unit;
the server may be connected to the video capturing device via a telecommunications network; and the video may comprise pixel values that are representative of color and the image analyzer may determine if a leak on the equipment is present from the pixel values that are representative of color.
The detection system comprises a video capturing device (for example a video camera) having a field of view; an image analyzer configured to analyze the video to determine if a leak on the equipment is present; and an output port for delivering a notification upon the image analyzer determining that a leak is present. Further embodiments may also include one or more of the following features: a video recording unit for recording events, for example video corresponding to an alarm, which may be stored according to the time and the classification for further reference; a server may comprise the image analyzer and video recording unit;
the server may be connected to the video capturing device via a telecommunications network; and the video may comprise pixel values that are representative of color and the image analyzer may determine if a leak on the equipment is present from the pixel values that are representative of color.
[0005] In an embodiment, there is provided a method of detecting a leak in equipment, the method comprising acquiring video of the target with video camera, in which the video comprises pixel values that are representative of color, sending the video to an image analyzer, the image analyzer analyzing video to determine if a leak is present on the equipment from the pixel values that are representative of color; and upon the image analyzer determining that a leak is present, generating a notification of the presence of a leak. Further embodiments may have one or more of the following features: at the time a leak is determined, events such as video data corresponding to an alarm may be stored to the local hard disk of the server or a removable disk for further reference; analyzing the video may comprise comparing the color of the video with color information from a reference frame; analyzing the video may comprise comparing the color of the video with a target color; analyzing the video may comprise comparing a difference between a frame being analyzed and a reference frame to a difference between two reference images; analyzing the video may comprise comparing the color of the video with a target color and comparing a difference between a frame being analyzed and a reference frame to a difference between two reference images; and in which the presence of a leak is determined upon determining that a portion of the frame being analyzed is similar to or darker than the target color and is also darker than the reference frame and the difference between the frame being analyzed and the reference frame is larger than the difference between the two reference images by an amount exceeding a threshold according to a metric; analyzing may be carried out by analysis of one or more regions of interest (ROIs) in the video; the one or more regions of interest may be user selected; and the equipment may comprise a pump.
[0006] These and other aspects of the device and method are set out in the claims, which are incorporated here by reference.
BRIEF DESCRIPTION OF THE FIGURES
BRIEF DESCRIPTION OF THE FIGURES
[0007] Fig. 1 is a block diagram of the video based leak detection system.
[0008] Fig. 2 illustrates method steps carried out by the image analyzer.
Date Recue/Date Received 2021-03-12 DETAILED DESCRIPTION
Date Recue/Date Received 2021-03-12 DETAILED DESCRIPTION
[0009] Immaterial modifications may be made to the embodiments described here without departing from what is covered by the claims.
[0010] A video based leak detection system is shown in Fig. 1 where one or more video cameras 102 are installed to view a scene 100, here a pump station with several pumps. The camera 102 acquires the video of the scene 100 including the pumps. Regions of interest (ROls) 101 may be defined within the scene as described below.
[0011] A video is a sequence of consecutive frames. A frame is a set of pixel values determined by light received in the one or more cameras 102 at locations on an imaging device in the camera corresponding to locations of the view. The video can be sent to a conventional input port of a server 105 directly, for example via a conventional cable, or indirectly via a suitable communications channel 103 such as a radio that communicates through a telecommunications network 104. A direct connection can also use a radio and an indirect connection through a network 104 can also use a cable. In some embodiments, the server 105 may be formed of a local processor (for example to carry out basic control of the camera) as well as a remote processor for carrying out other functions such as image analysis.
In some embodiments with a local processor and a remote processor, the local processor may be directly connected to the one or more video cameras 102 and the remote processor may be connected to the local processor via network 104.
In some embodiments with a local processor and a remote processor, the local processor may be directly connected to the one or more video cameras 102 and the remote processor may be connected to the local processor via network 104.
[0012] In the embodiment shown the server 105 analyzes the video and then determines if the scene 100 has a leak or not. The server comprises an image analyzer.
Here, "comprises"
may indicate that either the image analyzer is a separate physical component of the server or the image analyzer is software operating on a processor of the server. The server may also comprise a video recording unit, in the same sense of "comprise". The video recording unit may be connected to the image analyzer, with a connection that can be a physical connection or a software connection. A pump is an example of a target to which the method and apparatus may be applied. The method and apparatus may be applied to other targets. The server 105 may cause a notification 111 to be generated and output from an output port of the server 105, which may be any of a variety of notifications, and may be communicated directly to an operator or through a network 106, which may be the same network as network 104, via suitable methods such as through a fixed station 107, web interface 108, email 109 or smart phone 110.
The output port may be for example a USB port, network interface, display driver or other conventional output port of a computer. The notification 111 may comprise an electronic message indicative of a leak, or the generation of a sound, light signal, mechanical signal or other signal that may be observed by a human or monitoring equipment. The server 105 may communicate two-way with a further remote server (not shown) or other remote device such as smart phone 110. In on demand operation, the remote device may instruct the server 105 to acquire video from the camera 102 and the video is then processed by the server 105. In some embodiments, the server 105 may be a remote server and need not be situated with the camera 102.
Here, "comprises"
may indicate that either the image analyzer is a separate physical component of the server or the image analyzer is software operating on a processor of the server. The server may also comprise a video recording unit, in the same sense of "comprise". The video recording unit may be connected to the image analyzer, with a connection that can be a physical connection or a software connection. A pump is an example of a target to which the method and apparatus may be applied. The method and apparatus may be applied to other targets. The server 105 may cause a notification 111 to be generated and output from an output port of the server 105, which may be any of a variety of notifications, and may be communicated directly to an operator or through a network 106, which may be the same network as network 104, via suitable methods such as through a fixed station 107, web interface 108, email 109 or smart phone 110.
The output port may be for example a USB port, network interface, display driver or other conventional output port of a computer. The notification 111 may comprise an electronic message indicative of a leak, or the generation of a sound, light signal, mechanical signal or other signal that may be observed by a human or monitoring equipment. The server 105 may communicate two-way with a further remote server (not shown) or other remote device such as smart phone 110. In on demand operation, the remote device may instruct the server 105 to acquire video from the camera 102 and the video is then processed by the server 105. In some embodiments, the server 105 may be a remote server and need not be situated with the camera 102.
[0013] The server 105 performs the image analysis, leak detection and notification sending task. For image analysis, color information is proposed to be used. If there is a leak, some portions of the current frame, for example, the part of floor with accumulated leaking liquid, should be darker than the reference and similar to or darker than the target color. In some embodiments, the analyzer may intelligently compare the video with a reference, using more sophisticated statistical analysis than a greater than ¨ less than analysis.
For example, where a suite of training images is available, including some leak and non-leak images, perhaps being separated in time by hours or days, statistical analysis such as Bayesian or neural network analysis may be applied to determine a probability of a leak.
For example, where a suite of training images is available, including some leak and non-leak images, perhaps being separated in time by hours or days, statistical analysis such as Bayesian or neural network analysis may be applied to determine a probability of a leak.
[0014] Since the video is captured at various distance, direction and zoom levels, the users may be required in an initialization step 112 to defined one or more regions of interest (ROT). In an embodiment a software interface in the server 105 lets the users draw rectangle boxes around ROIs 101. In other embodiments the user may draw other shapes to define ROIs.
The use of regions within a scene allows the detection unit to individually check the leak in each region.
The use of regions within a scene allows the detection unit to individually check the leak in each region.
[0015] For each ROT 101 defined in step 112, the users can set one or more target color (TC) through the user interface (not shown) in step 113. The target color may be a color representing a color of a potentially leaking liquid. The target color representing a color of a potentially leaking liquid may represent an expected color of the liquid in an expected context, for example a mixed color of the floor and the liquid when the leaking liquid is poured on the floor.
Date Recue/Date Received 2021-04-19
Date Recue/Date Received 2021-04-19
[0016] After the ROT is defined in step 112 and TC is set in step 113, a reference frame (RF) and a reference frame difference (RFD) are saved in step 114 for further processing. The RFD is obtained by subtracting two frames.
[0017] In step 115, for each ROT 101, the current frame is compared with the TC set in step 113. The points that are similar to or darker than the TC are considered.
If the considered points are changed a lot compared with RF, for example if the differences from the reference frame are greater than a threshold, they are defined as potential points.
If the considered points are changed a lot compared with RF, for example if the differences from the reference frame are greater than a threshold, they are defined as potential points.
[0018] Because of the lighting change (or reflection) in the environment, some potential points may not be related to a leak. Therefore, in step 116 we will filter them out by comparing with the reference frame difference (RFD). Even though there is no leak at the time that the RFD
is obtained in step 114, we still can see some difference in the image. This is because of the background lighting change or some reflection by the floor. If there is a leak, the amount of frame difference should be much larger than the reference frame difference.
For each region of interest, if the amount of potential points is larger than the frame difference according to a metric by a threshold, and if the area occupies a certain percentage of the whole region, the server determines that there is a leak in that ROT in step 117. If any ROT is determined to have a leak in step 117 the server generates an alarm in step 118. The metric listed above can be the number of pixels that differ from the reference frame by more than a second threshold, and the threshold can be a fixed number of pixels or a multiple of the reference frame difference.
is obtained in step 114, we still can see some difference in the image. This is because of the background lighting change or some reflection by the floor. If there is a leak, the amount of frame difference should be much larger than the reference frame difference.
For each region of interest, if the amount of potential points is larger than the frame difference according to a metric by a threshold, and if the area occupies a certain percentage of the whole region, the server determines that there is a leak in that ROT in step 117. If any ROT is determined to have a leak in step 117 the server generates an alarm in step 118. The metric listed above can be the number of pixels that differ from the reference frame by more than a second threshold, and the threshold can be a fixed number of pixels or a multiple of the reference frame difference.
[0019] Events, for example video corresponding to an alarm, can be stored by the server 105 to a local hard disk of the server or a removable disk according to time or classification for further reference.
Date Recue/Date Received 2021-04-19
Date Recue/Date Received 2021-04-19
Claims (17)
PRIVILEGE IS CLAIMED ARE DEFINED AS FOLLOWS:
1. A video based leak detection system for detecting an indoor liquid leak from equipment, the system comprising:
a color video camera having a field of view for generating a video, in which each frame of the video comprises a set of pixel values that are representative of color of objects in the field of view;
a server comprising an image analyzer connected to receive the video and configured to analyze the video by comparing the set of pixel values that are representative of color of objects in the field of view to a preselected target color representing a color that a liquid potentially leaking from the equipment is expected to have, given an expected context when the liquid is outside the equipment in the field of view of the camera, and the image analyzer being configured to determine that a leak on the equipment is present based on the set of pixel values being similar in color and similar or darker in brightness than the preselected target color; and the image analyzer having an output port for delivering a notification upon the image analyzer determining that a leak is present.
a color video camera having a field of view for generating a video, in which each frame of the video comprises a set of pixel values that are representative of color of objects in the field of view;
a server comprising an image analyzer connected to receive the video and configured to analyze the video by comparing the set of pixel values that are representative of color of objects in the field of view to a preselected target color representing a color that a liquid potentially leaking from the equipment is expected to have, given an expected context when the liquid is outside the equipment in the field of view of the camera, and the image analyzer being configured to determine that a leak on the equipment is present based on the set of pixel values being similar in color and similar or darker in brightness than the preselected target color; and the image analyzer having an output port for delivering a notification upon the image analyzer determining that a leak is present.
2. The video based leak detection system of claim 1 further comprising a video recording unit connected to the image analyzer for recording at least a portion of the video, the portion of the video corresponding to the determination that a leak is present.
3. The video based leak detection system of claim 2 in which the server also comprises the video recording unit.
4. The video based leak detection system of claim 1 in which the server is connected to the color video camera via a communications channel of a telecommunications network.
5. A method of detecting indoor liquid leak from equipment, the method comprising:
acquiring video with a color video camera, in which each frame of the video comprises a set of pixel values that are representative of color of objects in a field of view;
sending the video to an image analyzer;
Date Recue/Date Received 2021-04-19 the image analyzer analyzing the video by comparing the set of pixel values that are representative of color of objects in the field of view to a preselected target color representing a color that a liquid potentially leaking from the equipment is expected to have, given an expected context when the liquid is outside the equipment in the field of view of the camera, the image analyzer determining that a leak is present on the equipment based on the set of pixel values being similar in color and similar or darker in brightness than the preselected target color; and upon the image analyzer determining that a leak is present, generating a notification of the presence of a leak.
acquiring video with a color video camera, in which each frame of the video comprises a set of pixel values that are representative of color of objects in a field of view;
sending the video to an image analyzer;
Date Recue/Date Received 2021-04-19 the image analyzer analyzing the video by comparing the set of pixel values that are representative of color of objects in the field of view to a preselected target color representing a color that a liquid potentially leaking from the equipment is expected to have, given an expected context when the liquid is outside the equipment in the field of view of the camera, the image analyzer determining that a leak is present on the equipment based on the set of pixel values being similar in color and similar or darker in brightness than the preselected target color; and upon the image analyzer determining that a leak is present, generating a notification of the presence of a leak.
6. The method of claim 5 in which analyzing the video comprises comparing the set of pixel values that are representative of color of objects in the field of view with color information from a reference frame having the field of view and being taken by the color video camera when no leak is present.
7. The method of claim 5 in which analyzing the video comprises comparing a difference between a frame being analyzed and a reference frame to a difference between two reference images, the reference frame and the two reference images having the field of view and being taken by the color video camera when no leak is present.
8. The method of claim 7 in which determining the presence of a leak comprises determining that the difference between the frame being analyzed and the reference frame is larger than the difference between the two reference images by an amount exceeding a threshold according to a metric.
9. The method of any one of claims 5-8 in which analyzing is carried out by analysis of one or more regions of interest in the video.
10. The method of claim 9 in which the one or more regions of interest is user selected.
11. The method of any one of claims 5-10 in which the equipment comprises a pump.
Date Recue/Date Received 2021-04-19
Date Recue/Date Received 2021-04-19
12. The video based leak detection system according to any one claims 1-4 in which the image analyzer is also configured to compare the set of pixel values that are representative of color of objects in the field of view to a reference frame, and the image analyzer being configured to determine that a leak is present based on the set of pixel values being similar to or darker than the preselected target color and having differences from the reference frame that are greater than a threshold.
13. The method according to claim 5 in which the image analyzer compares the set of pixel values that are representative of color of objects in the field of view to the target color and a reference frame, the image analyzer determining that a leak is present on the equipment based on the set of pixel values being similar to or darker than the target color and having differences from the reference frame that are greater than a threshold.
14. The video based leak detection system according to any one of claims 1-4, and 12 in which the target color is user selected.
15. The method according to any one of claims 5-11, and 13 in which the target color is user selected.
16. The video based leak detection system according to any one of claims 1-4, 12, and 14 in which the image analyzer is configured to compare the set of pixel values that are representative of color of objects in the field of view to multiple preselected target colors.
17. The method according to any one of claims 5-11, 13, and 15 in which the image analyzer compares the set of pixel values that are representative of color of objects in the field of view to multiple preselected target colors.
Date Recue/Date Received 2021-04-19
Date Recue/Date Received 2021-04-19
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA2860014A CA2860014C (en) | 2014-08-19 | 2014-08-19 | Video based indoor leak detection |
| PCT/CA2015/050789 WO2016026043A1 (en) | 2014-08-19 | 2015-08-19 | Video based leak detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CA2860014A CA2860014C (en) | 2014-08-19 | 2014-08-19 | Video based indoor leak detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CA2860014A1 CA2860014A1 (en) | 2016-02-19 |
| CA2860014C true CA2860014C (en) | 2021-11-23 |
Family
ID=55346396
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CA2860014A Active CA2860014C (en) | 2014-08-19 | 2014-08-19 | Video based indoor leak detection |
Country Status (2)
| Country | Link |
|---|---|
| CA (1) | CA2860014C (en) |
| WO (1) | WO2016026043A1 (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11607654B2 (en) | 2019-12-30 | 2023-03-21 | Marathon Petroleum Company Lp | Methods and systems for in-line mixing of hydrocarbon liquids |
| US11578638B2 (en) | 2021-03-16 | 2023-02-14 | Marathon Petroleum Company Lp | Scalable greenhouse gas capture systems and methods |
| US11655940B2 (en) | 2021-03-16 | 2023-05-23 | Marathon Petroleum Company Lp | Systems and methods for transporting fuel and carbon dioxide in a dual fluid vessel |
| US11447877B1 (en) | 2021-08-26 | 2022-09-20 | Marathon Petroleum Company Lp | Assemblies and methods for monitoring cathodic protection of structures |
| US11686070B1 (en) | 2022-05-04 | 2023-06-27 | Marathon Petroleum Company Lp | Systems, methods, and controllers to enhance heavy equipment warning |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4905187B2 (en) * | 2007-03-09 | 2012-03-28 | ソニー株式会社 | Image processing apparatus, imaging apparatus, image processing method, and computer program |
| CN101846576B (en) * | 2010-05-27 | 2011-12-14 | 王巍 | Video-based liquid leakage analyzing and alarming system |
| JP2013101474A (en) * | 2011-11-08 | 2013-05-23 | Tohoku Electric Power Co Inc | Oil leakage detector |
| US9025817B2 (en) * | 2012-09-05 | 2015-05-05 | Critical Imaging, LLC | System and method for leak detection |
-
2014
- 2014-08-19 CA CA2860014A patent/CA2860014C/en active Active
-
2015
- 2015-08-19 WO PCT/CA2015/050789 patent/WO2016026043A1/en active Application Filing
Also Published As
| Publication number | Publication date |
|---|---|
| WO2016026043A1 (en) | 2016-02-25 |
| CA2860014A1 (en) | 2016-02-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10943357B2 (en) | Video based indoor leak detection | |
| CN109886130B (en) | Target object determination method and device, storage medium and processor | |
| CA2860014C (en) | Video based indoor leak detection | |
| US7961953B2 (en) | Image monitoring system | |
| AU2021201574A1 (en) | Security system and method | |
| CN103270536B (en) | Stopped object detection | |
| US9648285B2 (en) | Monitoring method and camera | |
| US10713798B2 (en) | Low-complexity motion detection based on image edges | |
| CN111654700B (en) | Privacy mask processing method and device, electronic equipment and monitoring system | |
| US9001206B2 (en) | Cascadable camera tampering detection transceiver module | |
| KR20080058171A (en) | Camera tampering detection | |
| AU2012340862A1 (en) | Geographic map based control | |
| CN105100689A (en) | Automatic teller machine (ATM) video surveillance method and apparatus | |
| WO2018032270A1 (en) | Low complexity tamper detection in video analytics | |
| Kong et al. | A new quality model for object detection using compressed videos | |
| KR102366544B1 (en) | Vision-based Rainfall Information System and Methodology Using Deep Learning | |
| EP3629577B1 (en) | Data transmission method, camera and electronic device | |
| KR20180089249A (en) | Method for diagnosing fault of camera | |
| KR101970611B1 (en) | Method and Apparatus for Detecting Moving Object Modifying Learning Rate According to Frame Rate | |
| KR101169631B1 (en) | Video analysis camera, video analysis system having detection interface and detecting adjustment method of the same | |
| CN107845103B (en) | Alarm method and device based on texture information and computer readable storage medium | |
| CN111191593A (en) | Image target detection method and device, storage medium and sewage pipeline detection device | |
| CN108419047B (en) | Remote monitoring method based on cloud computing | |
| TWI512638B (en) | Intelligent area method and automatic camera state judgment method | |
| CN111091024A (en) | Small target filtering method and system based on video recognition result |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| EEER | Examination request |
Effective date: 20190807 |