CN112784706A - Oil testing test operation area safety control method based on image intelligent identification - Google Patents
Oil testing test operation area safety control method based on image intelligent identification Download PDFInfo
- Publication number
- CN112784706A CN112784706A CN202110004940.8A CN202110004940A CN112784706A CN 112784706 A CN112784706 A CN 112784706A CN 202110004940 A CN202110004940 A CN 202110004940A CN 112784706 A CN112784706 A CN 112784706A
- Authority
- CN
- China
- Prior art keywords
- image
- video image
- training
- type
- operation area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000012360 testing method Methods 0.000 title claims abstract description 82
- 238000000034 method Methods 0.000 title claims abstract description 41
- 238000012549 training Methods 0.000 claims abstract description 39
- 238000013135 deep learning Methods 0.000 claims abstract description 10
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 7
- 238000002485 combustion reaction Methods 0.000 claims description 21
- 238000012545 processing Methods 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims 2
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 239000003208 petroleum Substances 0.000 abstract description 2
- 238000012544 monitoring process Methods 0.000 description 5
- 238000004519 manufacturing process Methods 0.000 description 3
- 230000007812 deficiency Effects 0.000 description 1
- 238000005553 drilling Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000003129 oil well Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- 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
Abstract
The invention provides a safety control method for an oil testing test operation area based on intelligent image identification, and relates to the technical field of safety control methods for oil testing test operation fields in the petroleum industry. The method comprises the following steps: collecting an original video image and a training video image; extracting original features and training features; performing deep learning by using a convolutional neural network to obtain a background image and a training image; determining a threshold value according to the background image and the training image; collecting a field image of oil testing operation; comparing the field image with the corresponding background image to obtain a difference value of the field image and the corresponding background image; and comparing the difference value with a threshold value to determine the actual field condition of the oil testing operation area. The beneficial effects of the invention can include: the device can work efficiently, stably and for a long time; the safety problem of the oil testing operation area can be found in time, and an alarm is given to inform workers to deal with the safety problem in time, so that accidents are prevented.
Description
Technical Field
The invention relates to the technical field of safety control methods of oil testing operation sites in the petroleum industry, in particular to the technical field of safety control methods of oil testing operation areas based on intelligent image identification.
Background
During oil drilling operations, due to the fact that well depths are usually as high as thousands of meters, the pressure in a well is high, and a large amount of combustible gas exists, real-time monitoring and analysis of process data are needed. But the problems of how to effectively acquire production process data to serve production, improve production management efficiency, how to accurately acquire height data of open-flow flame in real time, how to effectively ensure the safety of oil wells and personnel and the like. Currently, sensor devices or video monitoring methods are generally used. The sensor equipment has the problem of measuring range, and when the sensor equipment is not in the measuring range, the sensor equipment cannot effectively detect and cannot be visualized directly. The video monitoring scheme is that on-site videos are collected, and an operator on duty checks a screen in real time to find problems and give an alarm in time. The mode has higher dependence degree on monitoring personnel, high labor intensity of the personnel, easy fatigue, excessive dependence on the monitoring personnel on timeliness and accuracy, and incapability of providing long-term stable and reliable guarantee. Furthermore, conventional "passive video" can only be forensically played back by calling video after an "event" has occurred, and losses have not been recoverable.
Disclosure of Invention
The present invention aims to address at least one of the above-mentioned deficiencies of the prior art.
In order to achieve the purpose, the invention provides a safety control method for an oil testing operation area based on intelligent image identification. The method may comprise the steps of: the method comprises the steps of collecting original video images and training video images of an oil testing operation area under different natural conditions, wherein the oil testing operation area comprises a combustion pool and/or a testing process area, the original video images comprise at least one of a first type video image, a second type video image and a third type video image, the training video images comprise at least one of a fourth type video image, a fifth type video image and a sixth type video image, the first type video image is a video image without flames of the combustion pool, the second type video image is a video image without personnel entering the testing process area, the third type video image is a video image without leakage of equipment, pipelines and valves, the fourth type video image is a video image with flames of the combustion pool, the fifth type video image is a video image without personnel entering the testing process area, and the sixth type video image is equipment, Video images of leaks in a pipeline or valve; extracting original features of an original video image to obtain a plurality of original features corresponding to different natural conditions, and performing deep learning on the original video image and the original features by using a convolutional neural network to obtain a plurality of background images corresponding to different natural conditions; extracting original features of a training video image to obtain a plurality of training features corresponding to different natural conditions, and performing deep learning on the training video image and the training features by using a convolutional neural network to obtain a plurality of training images corresponding to different natural conditions; determining threshold values for evaluating image differences under different natural conditions according to the plurality of background images and the training images; acquiring a field image of a test oil test operation, determining a corresponding background image according to field natural conditions, and comparing the field image with the corresponding background image to obtain a difference value of the field image and the corresponding background image; determining a threshold value according to natural conditions on site; and comparing the difference value with a threshold value to determine the actual field condition of the oil testing operation area.
Compared with the prior art, the beneficial effects of the invention can include: based on image intelligent identification, the system can work efficiently, stably and for a long time; safety problems in an oil testing operation area can be found in time, and an alarm is given to inform workers to deal with the safety problems in time, so that accidents are prevented; the safety of personnel can be effectively protected.
Detailed Description
Hereinafter, the method and system for controlling the safety of the oil testing operation area based on image intelligent identification according to the present invention will be described in detail with reference to exemplary embodiments.
As used herein, "first," "second," "third," "fourth," "fifth," "sixth," and the like are for convenience of description and of distinction only and are not to be construed as indicating or implying relative importance or order of magnitude.
Example 1
In an exemplary embodiment of the present invention, the test run test operating area may include a burn pool and/or a test flow area.
The oil testing test operation area safety control method based on image intelligent identification can comprise the following steps of:
collecting original video images and training video images of a test oil test operation area under different natural conditions;
extracting original features of an original video image to obtain a plurality of original features corresponding to different natural conditions;
performing deep learning on the original video image and the original features by using a convolutional neural network to obtain a plurality of background images corresponding to different natural conditions;
extracting original features of a training video image to obtain a plurality of training features corresponding to different natural conditions;
deep learning is carried out on the training video images and the training characteristics by using a convolutional neural network to obtain a plurality of training images corresponding to different natural conditions;
determining threshold values for evaluating image differences under different natural conditions according to the plurality of background images and the training images;
collecting a field image of oil testing operation;
determining a corresponding background image and a threshold value according to a field natural condition;
comparing the field image with the corresponding background image to obtain a difference value of the field image and the corresponding background image;
and comparing the difference value with a threshold value to determine the actual field condition of the oil testing operation area.
Wherein, the oil testing operation area comprises a combustion pool and/or a testing process area. The original video images comprise at least one of a first type video image, a second type video image and a third type video image, the first type video image is a video image without flame of the combustion pool, the second type video image is a video image without personnel entering the test process area, and the third type video image is a video image without leakage of equipment, pipelines and valves of the combustion pool and/or the test process area. The training video images comprise at least one of a fourth video image, a fifth video image and a sixth video image, the fourth video image is a video image with flames in the combustion pool, the fifth video image is a video image without personnel entering the test process area, and the sixth video image is a video image with leakage in equipment, pipelines or valves of the combustion pool and/or the test process area.
For example, when the test run area includes a combustion pool and a test flow area, the raw video images may include a first type video image, a second type video image, and a third type video image, and the training video images may include a fourth type video image, a fifth type video image, and a sixth type video image. At the moment, the third type of video image is a video image without leakage of equipment, pipelines and valves in the combustion pool and the test process area, and the sixth type of video image is a video image with leakage of the equipment, pipelines or valves in the combustion pool and the test process area.
When the test oil test operation area only comprises the combustion pool, the original video images can comprise the first type video images and/or the third type video images, and the training video images can comprise the first type video images and/or the third type video images. At the moment, the third type of video image is a video image of the combustion pool without leakage of equipment, pipelines and valves, and the sixth type of video image is a video image of the combustion pool with leakage of the equipment, pipelines or valves.
When the test oil test operation area only comprises the test flow area, the original video images can comprise the second type video images and/or the third type video images, and the training video images can comprise the fifth type video images and/or the sixth type video images. At this time, the third type of video image is a video image without leakage of the equipment, the pipeline and the valve in the test process area, and the sixth type of video image is a video image with leakage of the equipment, the pipeline or the valve in the test process area.
Further, in the step of determining the actual field condition of the oil testing operation area, a certain percentage (such as 80%) of the difference value larger than the threshold value or larger than the threshold value can be set, and then an alarm is given, so that the safety control of the oil testing operation area is realized. Further, in the step of determining the threshold, an artificial experience value can be considered, which is beneficial for the determined threshold to better conform to the field working condition of the oil testing operation area.
Further, the step of determining a threshold may comprise: performing image subtraction operation on the background image and the training image to obtain a difference value graph; and quantizing the difference value graph, and determining the threshold value. Furthermore, before the image subtraction operation is performed, the original features and the training features can be enhanced, so that the result of the image subtraction operation is more consistent with the field working condition of the oil testing test working area.
Further, the deep learning is supervised learning.
Further, the video image can be collected through a camera, and the collected video image can be stored in a video storage server.
Further, the deep learning step, the feature extraction step, the threshold determination step, the comparison step, and the like may be performed by a computer.
Further, the different natural conditions include: morning, noon, evening and night on any of cloudy, sunny and rainy days.
Further, the original features may include: a first feature of a first type of video image, a second feature of a second type of video image, and a third feature of a third type of video image. The first characteristic is the flameless condition of the combustion chamber, the second characteristic is the no personnel entry condition of the test flow area, and the third characteristic is the no leakage condition of the equipment, pipelines and valves. The training features may include: a fourth feature of a fourth type of video image, a fifth feature of a fifth type of video image, and a sixth feature of a sixth type of video image. The fourth characteristic is the flame condition of the combustion pool, the flame condition comprises flame, the fifth characteristic is the personnel condition of the test process area, the personnel condition comprises personnel entering, and the sixth characteristic is the leakage condition of equipment, pipelines and valves, and the leakage condition comprises leakage. Further, the flame condition also comprises the flame height, and the personnel condition also comprises the conditions of whether the personnel wear safety helmets or not and whether the personnel wear work clothes or not; the leakage condition may also include leakage and/or blow-by.
Further, the threshold may include: a first class threshold corresponding to the first feature and the fourth feature; a second class threshold corresponding to the second feature and the fifth feature; a third class threshold corresponding to the third feature and the sixth feature. The first threshold, the second threshold and the third threshold can be determined by performing image subtraction on the background image and the training image to obtain a difference figure, and then quantizing the difference figure. Further, the first, second and third threshold classes may also be determined in consideration of human experience and/or enhancement processing of the original and training features prior to the image subtraction operation.
While the present invention has been described above in connection with exemplary embodiments, it will be apparent to those of ordinary skill in the art that various modifications may be made to the above-described embodiments without departing from the spirit and scope of the claims.
Claims (9)
1. A safety control method for a test oil test operation area based on intelligent image identification is characterized by comprising the following steps:
the method comprises the steps of collecting original video images and training video images of an oil testing operation area under different natural conditions, wherein the oil testing operation area comprises a combustion pool and/or a testing process area, the original video images comprise at least one of a first type video image, a second type video image and a third type video image, the training video images comprise at least one of a fourth type video image, a fifth type video image and a sixth type video image, the first type video image is a video image without flames of the combustion pool, the second type video image is a video image without personnel entering the testing process area, the third type video image is a video image without leakage of equipment, pipelines and valves, the fourth type video image is a video image with flames of the combustion pool, the fifth type video image is a video image without personnel entering the testing process area, and the sixth type video image is equipment, Video images of leaks in a pipeline or valve;
extracting original features of an original video image to obtain a plurality of original features corresponding to different natural conditions, and performing deep learning on the original video image and the original features by using a convolutional neural network to obtain a plurality of background images corresponding to different natural conditions; extracting original features of a training video image to obtain a plurality of training features corresponding to different natural conditions, and performing deep learning on the training video image and the training features by using a convolutional neural network to obtain a plurality of training images corresponding to different natural conditions;
determining threshold values for evaluating image differences under different natural conditions according to the plurality of background images and the training images;
acquiring a field image of a test oil test operation, determining a corresponding background image according to field natural conditions, and comparing the field image with the corresponding background image to obtain a difference value of the field image and the corresponding background image; determining a threshold value according to natural conditions on site; and comparing the difference value with a threshold value to determine the actual field condition of the oil testing operation area.
2. The method for controlling the safety of the oil testing and testing operation area based on intelligent image recognition as claimed in claim 1, wherein in the step of determining the threshold value, an artificial experience value is also taken into account.
3. The oil testing test operation area safety control method based on intelligent image recognition as claimed in claim 1, wherein the step of determining the threshold value comprises:
performing image subtraction operation on the background image and the training image to obtain a difference value graph;
and quantizing the difference value graph, and determining the threshold value.
4. The oil test working area safety control method based on intelligent image recognition of claim 3, wherein before the image subtraction operation, the step of determining the threshold value further comprises:
and performing enhancement processing on the original features and the training features.
5. The oil testing test operation area safety control method based on intelligent image recognition is characterized in that the original characteristics comprise: a first feature of a first type of video image, a second feature of a second type of video image, and a third feature of a third type of video image; wherein the content of the first and second substances,
the first characteristic is the flameless condition of the combustion pool, the second characteristic is the condition that no person enters the test process area, and the third characteristic is the condition that the equipment, the pipeline and the valve have no leakage;
the training features include: a fourth feature of a fourth type of video image, a fifth feature of a fifth type of video image, a sixth feature of a sixth type of video image; wherein the content of the first and second substances,
the fourth characteristic is the flame condition of the combustion pool, and the flame condition comprises flame;
the fifth characteristic is that the personnel condition of the flow area is tested, and the personnel condition comprises the entering of personnel;
a sixth feature is the leakage of equipment, lines and valves, including leaks.
6. The oil testing test operation area safety control method based on intelligent image recognition is characterized in that the flame condition further comprises the flame height, and the personnel condition further comprises the conditions of whether personnel are provided with safety helmets and whether personnel wear work clothes; the leakage condition may also include leakage and/or blow-by.
7. The oil testing test operation area safety control method based on intelligent image recognition is characterized in that the threshold value comprises the following steps:
a first class threshold corresponding to the first feature and the fourth feature;
a second class threshold corresponding to the second feature and the fifth feature;
a third class threshold corresponding to the third feature and the sixth feature.
8. The oil test operation area safety control method based on intelligent image recognition is characterized in that the deep learning is supervised learning.
9. The oil testing test operation area safety control method based on intelligent image recognition is characterized in that the different natural conditions comprise: morning, noon, evening and night on any of cloudy, sunny and rainy days.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110004940.8A CN112784706B (en) | 2021-01-04 | 2021-01-04 | Oil testing test operation area safety control method based on image intelligent identification |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110004940.8A CN112784706B (en) | 2021-01-04 | 2021-01-04 | Oil testing test operation area safety control method based on image intelligent identification |
Publications (2)
Publication Number | Publication Date |
---|---|
CN112784706A true CN112784706A (en) | 2021-05-11 |
CN112784706B CN112784706B (en) | 2023-11-14 |
Family
ID=75755208
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110004940.8A Active CN112784706B (en) | 2021-01-04 | 2021-01-04 | Oil testing test operation area safety control method based on image intelligent identification |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112784706B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113655061A (en) * | 2021-09-23 | 2021-11-16 | 华志(福建)电子科技有限公司 | Method for identifying melting point of substance based on image and melting point instrument |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019127271A1 (en) * | 2017-12-28 | 2019-07-04 | 深圳市锐明技术股份有限公司 | Warning method, device, storage medium and server regarding physical conflict behavior |
CN112149513A (en) * | 2020-08-28 | 2020-12-29 | 成都飞机工业(集团)有限责任公司 | Industrial manufacturing site safety helmet wearing identification system and method based on deep learning |
-
2021
- 2021-01-04 CN CN202110004940.8A patent/CN112784706B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2019127271A1 (en) * | 2017-12-28 | 2019-07-04 | 深圳市锐明技术股份有限公司 | Warning method, device, storage medium and server regarding physical conflict behavior |
CN112149513A (en) * | 2020-08-28 | 2020-12-29 | 成都飞机工业(集团)有限责任公司 | Industrial manufacturing site safety helmet wearing identification system and method based on deep learning |
Non-Patent Citations (2)
Title |
---|
刘勇智;: "油田无线网络接入解决方案", 通信管理与技术, no. 02 * |
王文正;吴德松;李千登;: "钻井过程人员异常行为视频智能识别系统", 安全、健康和环境, no. 02 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113655061A (en) * | 2021-09-23 | 2021-11-16 | 华志(福建)电子科技有限公司 | Method for identifying melting point of substance based on image and melting point instrument |
Also Published As
Publication number | Publication date |
---|---|
CN112784706B (en) | 2023-11-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111930049A (en) | Forestry forest fire prevention safety intelligent monitoring management system based on thing networking | |
CN109815904B (en) | Fire identification method based on convolutional neural network | |
CN108051709A (en) | Transformer state online evaluation analysis method based on artificial intelligence technology | |
CN103968256B (en) | Piping for tank farm leakage detection method | |
CN106327062A (en) | Method for evaluating state of power distribution network equipment | |
CN110121053B (en) | Video monitoring method for drilling site risk grading early warning | |
CN113344326A (en) | System and method for identifying dynamic hidden danger and evaluating risk of multi-information fusion special operation site | |
CN109345148A (en) | Natural gas line and the intelligent diagnosing method of station process unit security risk identification | |
CN114444823A (en) | Method and system for predicting lost circulation risk in drilling operation process | |
CN110130875A (en) | Pumping unit unusual service condition monitoring method | |
CN112784706B (en) | Oil testing test operation area safety control method based on image intelligent identification | |
CN106706119B (en) | Vibration source identification method and system based on signal frequency domain characteristics | |
CN115949890A (en) | Urban gas pipe network leakage monitoring grading alarm and disposal method | |
CN116342999A (en) | Open space fire smoke detection and identification method for ultrahigh voltage transformer substation | |
CN111144433B (en) | Oil well working condition intelligent diagnosis and analysis method and device based on SVM model | |
CN112766772B (en) | Method and device for monitoring oil-gas-water well in real time, computer equipment and storage medium | |
CN110849938A (en) | Automatic warning and sample reserving method and device for over-standard pollution discharge | |
CN109902265B (en) | Underground early warning method based on hidden Markov model | |
CN114239215A (en) | Intelligent pipeline on-line monitoring system | |
CN116777085B (en) | Coal mine water damage prediction system based on data analysis and machine learning technology | |
CN113137221A (en) | Three-level gas leakage evaluation system and evaluation method for whole gas extraction system | |
CN110826762A (en) | Mine water disaster evolution trend prediction method suitable for complex groundwater environment | |
CN116164241A (en) | Intelligent detection method for leakage faults of gas extraction pipe network | |
CN115481941A (en) | Multifunctional area combined intelligent security management method and system | |
CN115689118A (en) | Oil and gas pipeline risk evaluation method and system, storage medium and electronic equipment |
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 |