CN116486307B - Chemical industry park safety early warning method and system based on video analysis technology - Google Patents
Chemical industry park safety early warning method and system based on video analysis technology Download PDFInfo
- Publication number
- CN116486307B CN116486307B CN202310494640.1A CN202310494640A CN116486307B CN 116486307 B CN116486307 B CN 116486307B CN 202310494640 A CN202310494640 A CN 202310494640A CN 116486307 B CN116486307 B CN 116486307B
- Authority
- CN
- China
- Prior art keywords
- video image
- spectrum
- spectral
- parabola
- chemical industry
- 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
- 239000000126 substance Substances 0.000 title claims abstract description 64
- 238000000034 method Methods 0.000 title claims abstract description 44
- 238000005516 engineering process Methods 0.000 title claims abstract description 13
- 238000004458 analytical method Methods 0.000 title claims abstract description 12
- 238000001228 spectrum Methods 0.000 claims abstract description 111
- 238000001514 detection method Methods 0.000 claims abstract description 67
- 238000010586 diagram Methods 0.000 claims abstract description 36
- 238000012544 monitoring process Methods 0.000 claims abstract description 12
- 230000002159 abnormal effect Effects 0.000 claims abstract description 9
- 230000003595 spectral effect Effects 0.000 claims description 121
- 238000012545 processing Methods 0.000 claims description 10
- 230000011218 segmentation Effects 0.000 claims description 10
- 238000000605 extraction Methods 0.000 claims description 7
- 238000009792 diffusion process Methods 0.000 claims description 5
- 230000005856 abnormality Effects 0.000 claims description 4
- 238000005259 measurement Methods 0.000 claims description 4
- 239000013598 vector Substances 0.000 claims description 4
- 208000037170 Delayed Emergence from Anesthesia Diseases 0.000 claims description 3
- 238000012423 maintenance Methods 0.000 abstract description 7
- 239000007789 gas Substances 0.000 description 49
- 238000007781 pre-processing Methods 0.000 description 3
- 238000010183 spectrum analysis Methods 0.000 description 3
- 230000002457 bidirectional effect Effects 0.000 description 2
- 230000004927 fusion Effects 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/02—Alarms for ensuring the safety of persons
- G08B21/12—Alarms for ensuring the safety of persons responsive to undesired emission of substances, e.g. pollution alarms
-
- 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
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- 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
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video 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
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- 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
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- General Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Molecular Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Toxicology (AREA)
- Business, Economics & Management (AREA)
- Environmental & Geological Engineering (AREA)
- Emergency Management (AREA)
- Image Processing (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The application discloses a chemical industry park safety early warning method and a system based on a video analysis technology, wherein the method comprises the following steps: carrying out standardized treatment on the chemical industry park area to obtain a rectangular area diagram of the monitoring area; taking the position of the central point of the rectangular area diagram as the origin of coordinates, and completing the establishment of a plane rectangular coordinate system; drawing a shape parabola U1 and a shape parabola U2 in a plane rectangular coordinate system, disposing a first spectrum video image collector at the coordinate origin position of the rectangular region diagram, disposing m third spectrum video image collectors on the U1 parabola and the U2 parabola at equal intervals; based on the spectrum video image collected by the spectrum video image collector, the detection result of gas leakage in the chemical industry park is obtained through recognition, and abnormal alarm is carried out according to the detection result. The data acquisition range that makes the cover chemical industry garden is more extensive, and the maintenance of later stage collector is more convenient, has practicality and rationality more.
Description
Technical Field
The application relates to the field of video image processing, in particular to a chemical industry park safety early warning method and system based on a video analysis technology.
Background
The effective monitoring of dangerous gases in the chemical industry park, railway station, airport and other fields plays a vital role in safe production and living order. With the rapid development of science and technology, environmental data are acquired by using a collector, and a detection model is input for gas detection, so that the method is a hot spot technology of current research; however, the current arrangement of data collectors in chemical industry parks is mostly free from fixed rules and can be circulated, the arrangement strategies of each park are different, and unified efficient management cannot be realized, so that monitoring of the data collectors in a special period can be caused, and particularly, the condition that the management and control maintenance of the data collectors is reduced in efficiency is caused in the working handover/alternation/replacement process of workers. For example, CN111965116a (publication No. 20201120) discloses a hyperspectral-based airport gas detection system, which comprises a hyperspectral image data acquisition module, a data preprocessing module and a spectrum analysis module; the hyperspectral image data acquisition module comprises a plurality of single-band hyperspectral cameras with different shooting angles; the single-band hyperspectral camera is used for collecting hyperspectral image data of airport gas in real time; the data preprocessing module is used for preprocessing the collected airport gas hyperspectral image data; the spectrum analysis module is used for analyzing and processing the preprocessed airport gas hyperspectral image data, a bidirectional circulating neural network model is arranged in the spectrum analysis module, the bidirectional circulating neural network model inputs the airport gas hyperspectral image data, and content information of target gas in an airport is output. The method does not reasonably set the position of the data acquisition module, which leads to high maintenance cost, waste of manpower and material resources and low maintenance efficiency in the later period. Meanwhile, the target detection task based on the traditional computer vision is very dependent on the related acquisition characteristics, and partial characteristics of the gas leakage image are often ignored, so that the accuracy of dangerous gas monitoring is low, whether the target gas leaks can not be rapidly and accurately identified, and the requirements of the current complex environment can not be met.
Disclosure of Invention
Aiming at the defects in the prior art, the application aims to provide a chemical industry park safety early warning method and system based on a video analysis technology, which are characterized in that a spectrum video image collector is arranged on a parabola by drawing a plurality of symmetrical parabolas and combining with the intersection point position of the rectangle chemical industry park, the detection result of gas leakage in the chemical industry park is identified and obtained based on the spectrum video image collected by the spectrum video image collector, and the abnormal alarm is carried out according to the detection result.
In order to achieve the aim of the application, the application adopts the following technical scheme:
in a first aspect, the application provides a chemical industry park safety pre-warning method based on a video analysis technology, which comprises the following steps:
step 1, carrying out standardization treatment on a chemical industry park area to obtain a rectangular area diagram of a monitoring area;
step 2, taking the position of the central point of the rectangular area diagram as a coordinate origin, taking the horizontal direction as an X axis and the vertical direction as a Y axis, and completing the establishment of a plane rectangular coordinate system;
step 3, drawing in the plane rectangular coordinate systemShape parabolas U1 and->The shape of the parabola U2, the parabola U1 and the parabola U2 are symmetrical, wherein n is a natural number;
step 4, adjusting n to enable two intersection points of the U1 parabola and the rectangular long side to be the length of the rectangular long sideAt such a position that the two points of intersection of the U2 parabola with the rectangular long side are +.>A location;
step 5, the following operations are carried out within the scope of the rectangular area diagram: disposing first spectrum video image collectors at the coordinate origin position of the rectangular region diagram, disposing m second spectrum video image collectors on the U1 parabola at equal intervals, and disposing m third spectrum video image collectors on the U2 parabola at equal intervals;
and 6, identifying and obtaining a detection result of gas leakage in the chemical industry park based on the spectral video images acquired by the first spectral video image acquirer, the spectral video images acquired by the m second spectral video image acquirers and the spectral video images acquired by the m third spectral video image acquirers, and carrying out abnormal alarm according to the detection result.
Further, step 6, based on the spectral video images collected by the first spectral video image collector, the spectral video images collected by the m second spectral video image collectors, and the spectral video images collected by the m third spectral video image collectors, identifies and obtains a detection result of gas leakage in the chemical industry park, specifically includes:
the spectral video images acquired by the first spectral video image acquirer, the spectral video images acquired by the m second spectral video image acquirers and the spectral video images acquired by the m third spectral video image acquirers are respectively subjected to video frame feature extraction, and feature segmentation is respectively performed according to nodes of a fixed time period;
then, according to the nodes of the corresponding time periods, the characteristics of the video images of different spectrums after being segmented are combined, and the combined time series characteristics are expressed as P= (x) 1 ,x 2 ,...,x i ) The method comprises the steps of carrying out a first treatment on the surface of the Wherein x is i Representing the characteristics of the spectrum video images acquired by the first spectrum video image acquirer, the spectrum video images acquired by the m second spectrum video image acquirers and the spectrum video images acquired by the m third spectrum video image acquirers in an ith time period node after being combined in the ith time period node;
and inputting the combined time sequence characteristics into a ResNet50 network to obtain a detection result of gas leakage in the chemical industry park.
Further, the method further comprises the following steps:
based on the step 5, the following operations are performed within the scope of the rectangular area map: disposing first spectrum video image collectors at the coordinate origin position of the rectangular region diagram, disposing m second spectrum video image collectors on the U1 parabola at equal intervals, and disposing m third spectrum video image collectors on the U2 parabola at equal intervals;
extracting the spatial position characteristic of each spectrum video image collector in the step 5, wherein the spatial position characteristic is the position coordinate of the spectrum video image collector in the plane rectangular coordinate system;
the time sequence features and the spatial position features are overlapped and combined to obtain the fused enhancement features;
and inputting the enhanced characteristics into a ResNet50 network to obtain a detection result of gas leakage in the chemical industry park.
Further, step 6, based on the spectral video images collected by the first spectral video image collector, the spectral video images collected by the m second spectral video image collectors, and the spectral video images collected by the m third spectral video image collectors, identifies and obtains a detection result of gas leakage in the chemical industry park, specifically includes:
the spectral video images acquired by the first spectral video image acquirer, the spectral video images acquired by the m second spectral video image acquirers and the spectral video images acquired by the m third spectral video image acquirers are respectively subjected to video frame feature extraction, and feature segmentation is respectively carried out according to nodes in a fixed time period to obtain time sequence features corresponding to each spectral image acquirer;
aiming at the current time sequence characteristic x corresponding to each spectrum image collector a The similarity measurement is carried out, and specifically comprises the following steps:
for the spectrum image collector at the current position, the h time sequence section in the history database is assumed to be in a state without target dangerous gas, and the image time sequence characteristic of the corresponding spectrum image collector is x h The method comprises the steps of carrying out a first treatment on the surface of the The j-th time sequence segment is assumed to be in a state with target dangerous gas, and the image time sequence characteristic of the corresponding spectrum image collector is x j ;
The degree of similarity D1 is calculated and,,/>representing the 2 norms of the vectors;
the degree of similarity D2 is calculated and,;
comparing the similarity D1 and the similarity D2 with set thresholds respectively to obtain detection results of target dangerous gas leakage, wherein the detection results specifically comprise:
when D1 is smaller than the first threshold value and D2 is larger than the second threshold value, indicating that the target dangerous gas leaks, and performing primary alarm;
when D1 is larger than the first threshold value and D2 is smaller than the second threshold value, no target dangerous gas leakage is indicated;
when D1 is larger than the first threshold value and D2 is larger than the second threshold value, or when D1 is smaller than the first threshold value and D2 is smaller than the second threshold value, the dangerous abnormality is indicated to exist in a large probability, and the secondary alarm is carried out.
Further, the method further comprises the following steps:
step 7, drawing in the plane rectangular coordinate systemShape parabolic sum->The shape parabola specifically further comprises:
n is a plurality of numerical values, and a plurality of parabolas which are symmetrical up and down are drawn;
step 8, arranging spectrum video image collectors on the parabolas which are vertically symmetrical in the range of the rectangular area diagram;
and 9, identifying and obtaining a detection result of gas leakage in the chemical industry park based on the spectrum video image acquired by the spectrum video image acquisition device in the step 8.
Further, the method further comprises the following steps:
respectively extracting video frame characteristics of the spectrum video images acquired by the spectrum video image acquirer in the step 8, and respectively carrying out characteristic segmentation according to nodes in a fixed time period;
and merging the video image features of different spectrums after the nodes are segmented according to the corresponding time period, and inputting the merged time sequence features into a ResNet50 network to obtain a detection result of gas leakage in the chemical industry park.
Further, the method further comprises the following steps: the detection result specifically comprises detection and identification of gas leakage position, concentration distribution and diffusion trend information in the spectrum video.
Further, the method further comprises the following steps: the ResNet50 network is an improved network structure, and specifically comprises the following components: a mixed attention mechanism CBAM is added to the res net50 network, specifically the CBAM is placed in the residual block of the res net50 network and named res net50+cbam.
In a second aspect, the present application also provides a chemical industry park safety pre-warning system based on a video analysis technology, the system comprising:
the standardized processing module is used for carrying out standardized processing on the chemical industry park area to obtain a rectangular area diagram of the monitoring area;
the coordinate system establishment module is used for completing establishment of a plane rectangular coordinate system by taking the position of the central point of the rectangular area diagram as a coordinate origin, taking the horizontal direction as an X axis and taking the vertical direction as a Y axis;
a shape drawing module for drawing in the plane rectangular coordinate systemShape parabolas U1 and->The shape of the parabola U2, the parabola U1 and the parabola U2 are symmetrical, wherein n is a natural number;
the adjusting module is used for adjusting n to enable two intersection points of the U1 parabola and the rectangular long side to be the length of the rectangular long sideAt such a position that the two points of intersection of the U2 parabola with the rectangular long side are +.>A location;
the first deployment module is used for carrying out the following operations within the scope of the rectangular area diagram: disposing first spectrum video image collectors at the coordinate origin position of the rectangular region diagram, disposing m second spectrum video image collectors on the U1 parabola at equal intervals, and disposing m third spectrum video image collectors on the U2 parabola at equal intervals;
the first detection module is used for identifying and obtaining detection results of gas leakage in the chemical industry park based on the spectral video images collected by the first spectral video image collector, the spectral video images collected by the m second spectral video image collectors and the spectral video images collected by the m third spectral video image collectors, and carrying out abnormal alarm according to the detection results.
Further, the method further comprises the following steps:
a plurality of symmetrical parabolic drawing modules for drawing in the plane rectangular coordinate systemShape parabolic sum->The shape parabola specifically further comprises:
n is a plurality of numerical values, and a plurality of parabolas which are symmetrical up and down are drawn;
the second deployment module is used for arranging the spectrum video image collectors on the parabolas which are symmetrical up and down in the range of the rectangular area diagram;
and the second detection module is used for identifying and obtaining a detection result of gas leakage in the chemical industry park based on the spectral video image acquired by the spectral video image acquisition device in the second deployment module.
The beneficial effects are that:
1. the application has strong variability based on parabola, large coverage, easy adjustment of curve shape and easy realization of good symmetry; the first innovation provides a mode of drawing a plurality of symmetrical parabolas, arranging a spectrum video image collector on the parabolas in combination with the intersection point position of the rectangle chemical industry park, identifying and obtaining the detection result of gas leakage in the chemical industry park based on the spectrum video image collected by the spectrum video image collector, and carrying out abnormal alarm according to the detection result. The data acquisition range that makes the cover chemical industry garden is more extensive for the maintenance of later stage collector is more convenient, has practicality and rationality more, and provides the reference value of high performance price ratio for the arrangement of the collector in the chemical industry garden.
2. The time feature and the space feature can well ensure the integrity of the extracted feature, and based on space-time feature fusion, the mixed attention mechanism CBAM is further added in the ResNet50 network, so that the accuracy of a network model is improved, and the information such as the gas leakage position, concentration distribution, diffusion trend and the like in the spectrum video can be efficiently and accurately detected and identified.
3. According to the application, the characteristic distance between the current time sequence characteristic corresponding to the spectrum image collector and the time sequence characteristic under the two presupposed states is calculated, and the two characteristic distances are comprehensively compared to obtain the target dangerous gas leakage detection result, so that the detection efficiency and accuracy are greatly improved.
Drawings
Fig. 1 is a schematic flow chart of a chemical industry park safety pre-warning method based on a video analysis technology.
Detailed Description
The application is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting of the application. It should be noted that, for convenience of description, only the portions related to the application are shown in the drawings.
As shown in fig. 1, this embodiment provides a chemical industry park safety early warning method based on video analysis technology, which includes:
step 1, carrying out standardization treatment on a chemical industry park area to obtain a rectangular area diagram of a monitoring area;
specifically, based on the edge points of the chemical industry park, the rectangular frame covers all edge range points of the park as the reference, and the rectangular area map is obtained through marking processing.
Step 2, taking the position of the central point of the rectangular area diagram as a coordinate origin, taking the horizontal direction as an X axis and the vertical direction as a Y axis, and completing the establishment of a plane rectangular coordinate system;
step 3, drawing in the plane rectangular coordinate systemShape parabolas U1 and->The shape of the parabola U2, the parabola U1 and the parabola U2 are symmetrical, wherein n is a natural number;
step 4, adjusting n to enable two intersection points of the U1 parabola and the rectangular long side to be the length of the rectangular long sideAt such a position that the two points of intersection of the U2 parabola with the rectangular long side are +.>A location;
specifically, the parabola U1 intersects with the long side of the rectangle, the long side of the rectangle is quartered, and n is adjusted so that two intersection points of the parabola U1 and the long side of the rectangle are located at two quarter points on the left and right sides.
Step 5, the following operations are carried out within the scope of the rectangular area diagram: disposing first spectrum video image collectors at the coordinate origin position of the rectangular region diagram, disposing m second spectrum video image collectors on the U1 parabola at equal intervals, and disposing m third spectrum video image collectors on the U2 parabola at equal intervals;
and 6, identifying and obtaining a detection result of gas leakage in the chemical industry park based on the spectral video images acquired by the first spectral video image acquirer, the spectral video images acquired by the m second spectral video image acquirers and the spectral video images acquired by the m third spectral video image acquirers, and carrying out abnormal alarm according to the detection result.
In particular, the parabola has the advantages of strong variability and large coverage, compared with other curves, and the curve shape is easy to adjust and good symmetry is easy to realize. Therefore, the method is provided that a plurality of symmetrical parabolas are drawn, the spectrum video image collectors are arranged on the parabolas in combination with the intersection point positions of the rectangular chemical industry park, and the detection result of the gas leakage in the chemical industry park is identified and obtained based on the spectrum video images collected by the spectrum video image collectors.
In an optional embodiment, the step 6, based on the spectral video images collected by the first spectral video image collector, the spectral video images collected by the m second spectral video image collectors, and the spectral video images collected by the m third spectral video image collectors, identifies and obtains a detection result of gas leakage in the chemical industry park, specifically includes:
the spectral video images acquired by the first spectral video image acquirer, the spectral video images acquired by the m second spectral video image acquirers and the spectral video images acquired by the m third spectral video image acquirers are respectively subjected to video frame feature extraction, and feature segmentation is respectively performed according to nodes of a fixed time period;
then, according to the nodes of the corresponding time periods, the characteristics of the video images of different spectrums after being segmented are combined, and the combined time series characteristics are expressed as P= (x) 1 ,x 2 ,...,x i ) The method comprises the steps of carrying out a first treatment on the surface of the Wherein x is i Representing the characteristics of the spectrum video images acquired by the first spectrum video image acquirer, the spectrum video images acquired by the m second spectrum video image acquirers and the spectrum video images acquired by the m third spectrum video image acquirers in an ith time period node after being combined in the ith time period node;
specifically, according to each time period node, the spectral video image features corresponding to each time period node are overlapped and combined.
And inputting the combined time sequence characteristics into a ResNet50 network to obtain a detection result of gas leakage in the chemical industry park.
In an alternative embodiment, the method further comprises:
based on the step 5, the following operations are performed within the scope of the rectangular area map: disposing first spectrum video image collectors at the coordinate origin position of the rectangular region diagram, disposing m second spectrum video image collectors on the U1 parabola at equal intervals, and disposing m third spectrum video image collectors on the U2 parabola at equal intervals;
extracting the spatial position characteristic of each spectrum video image collector in the step 5, wherein the spatial position characteristic is the position coordinate of the spectrum video image collector in the plane rectangular coordinate system;
the time sequence features and the spatial position features are overlapped and combined to obtain the fused enhancement features;
and inputting the enhanced characteristics into a ResNet50 network to obtain a detection result of gas leakage in the chemical industry park.
The time feature and the space feature can well ensure the integrity of the extracted feature, and the method is based on space-time feature fusion, and further improves the accuracy of a network model by adding a mixed attention mechanism CBAM in a ResNet50 network, so that the information such as gas leakage position, concentration distribution, diffusion trend and the like in a spectrum video can be efficiently and accurately detected and identified.
In an optional embodiment, the step 6, based on the spectral video images collected by the first spectral video image collector, the spectral video images collected by the m second spectral video image collectors, and the spectral video images collected by the m third spectral video image collectors, identifies and obtains a detection result of gas leakage in the chemical industry park, specifically includes:
the spectral video images acquired by the first spectral video image acquirer, the spectral video images acquired by the m second spectral video image acquirers and the spectral video images acquired by the m third spectral video image acquirers are respectively subjected to video frame feature extraction, and feature segmentation is respectively carried out according to nodes in a fixed time period to obtain time sequence features corresponding to each spectral image acquirer;
aiming at the current time sequence characteristic x corresponding to each spectrum image collector a The similarity measurement is carried out, and specifically comprises the following steps:
for the spectrum image collector at the current position, the h time sequence section in the history database is assumed to be in a state without target dangerous gas, and the image time sequence characteristic of the corresponding spectrum image collector is x h The method comprises the steps of carrying out a first treatment on the surface of the The j-th time sequence segment is assumed to be in a state with target dangerous gas, and the image time sequence characteristic of the corresponding spectrum image collector is x j ;
The degree of similarity D1 is calculated and,,/>representing the 2 norms of the vectors;
the degree of similarity D2 is calculated and,;
comparing the similarity D1 and the similarity D2 with set thresholds respectively to obtain detection results of target dangerous gas leakage, wherein the detection results specifically comprise:
when D1 is smaller than the first threshold value and D2 is larger than the second threshold value, indicating that the target dangerous gas leaks, and performing primary alarm;
when D1 is larger than the first threshold value and D2 is smaller than the second threshold value, no target dangerous gas leakage is indicated;
when D1 is larger than the first threshold value and D2 is larger than the second threshold value, or when D1 is smaller than the first threshold value and D2 is smaller than the second threshold value, the dangerous abnormality is indicated to exist in a large probability, and the secondary alarm is carried out.
Specifically, the emergency processing capability and efficiency can be improved by carrying out alarm distinction in a hierarchical alarm mode.
According to the application, the characteristic distance between the current time sequence characteristic corresponding to the spectrum image collector and the time sequence characteristic under the two presupposed states is calculated, and the two characteristic distances are comprehensively compared to obtain the target dangerous gas leakage detection result, so that the detection efficiency and accuracy are greatly improved.
In an alternative embodiment, the method further comprises:
step 7, drawing in the plane rectangular coordinate systemShape parabolic sum->The shape parabola specifically further comprises:
n is a plurality of numerical values, and a plurality of parabolas which are vertically symmetrical and are not intersected are drawn;
step 8, arranging spectrum video image collectors on the parabolas which are vertically symmetrical in the range of the rectangular area diagram;
and 9, identifying and obtaining a detection result of gas leakage in the chemical industry park based on the spectrum video image acquired by the spectrum video image acquisition device in the step 8.
In an alternative embodiment, the method further comprises:
respectively extracting video frame characteristics of the spectrum video images acquired by the spectrum video image acquirer in the step 8, and respectively carrying out characteristic segmentation according to nodes in a fixed time period;
and merging the video image features of different spectrums after the nodes are segmented according to the corresponding time period, and inputting the merged time sequence features into a ResNet50 network to obtain a detection result of gas leakage in the chemical industry park.
Specifically, the spatial position coordinates of the spectral video image collector in the step 8 may be input into a res net50 network together, so as to obtain a detection result of gas leakage in the chemical industry park.
In an alternative embodiment, the method further comprises: the detection result specifically comprises a plurality of information such as gas leakage positions, concentration distribution, diffusion trends and the like detected and identified in the spectrum video.
In an alternative embodiment, the method further comprises: the ResNet50 network is an improved network structure, and specifically comprises the following components: a mixed attention mechanism CBAM is added to the res net50 network, specifically the CBAM is placed in the residual block of the res net50 network and named res net50+cbam.
Specifically, the ResNet50 network is trained by random gradient descent (SGD) as an optimizer.
Based on the same inventive concept, the embodiment provides a chemical industry park safety early warning system based on a video analysis technology, which comprises:
the standardized processing module is used for carrying out standardized processing on the chemical industry park area to obtain a rectangular area diagram of the monitoring area;
the coordinate system establishment module is used for completing establishment of a plane rectangular coordinate system by taking the position of the central point of the rectangular area diagram as a coordinate origin, taking the horizontal direction as an X axis and taking the vertical direction as a Y axis;
a shape drawing module for drawing in the plane rectangular coordinate systemShape parabolas U1 and->The shape of the parabola U2, the parabola U1 and the parabola U2 are symmetrical, wherein n is a natural number;
the adjusting module is used for adjusting n to enable two intersection points of the U1 parabola and the rectangular long side to be the length of the rectangular long sideAt such a position that the two points of intersection of the U2 parabola with the rectangular long side are +.>A location;
the first deployment module is used for carrying out the following operations within the scope of the rectangular area diagram: disposing first spectrum video image collectors at the coordinate origin position of the rectangular region diagram, disposing m second spectrum video image collectors on the U1 parabola at equal intervals, and disposing m third spectrum video image collectors on the U2 parabola at equal intervals;
the first detection module is used for identifying and obtaining detection results of gas leakage in the chemical industry park based on the spectral video images collected by the first spectral video image collector, the spectral video images collected by the m second spectral video image collectors and the spectral video images collected by the m third spectral video image collectors, and carrying out abnormal alarm according to the detection results.
In an alternative embodiment, the method further comprises:
a plurality of symmetrical parabolic drawing modules for drawing in the plane rectangular coordinate systemShape parabolic sum->The shape parabola specifically further comprises:
n is a plurality of numerical values, and a plurality of parabolas which are symmetrical up and down are drawn;
the second deployment module is used for arranging the spectrum video image collectors on the parabolas which are symmetrical up and down in the range of the rectangular area diagram;
and the second detection module is used for identifying and obtaining a detection result of gas leakage in the chemical industry park based on the spectral video image acquired by the spectral video image acquisition device in the second deployment module.
Aiming at the arrangement of the current chemical industry park data collectors mentioned in the background art, most of the arrangement is free from fixed rules and can be circulated, the arrangement strategies of each park are different, and unified efficient management cannot be realized, so that monitoring of the data collectors in a special period can be caused, and particularly, the condition that efficiency is reduced in the management and maintenance of the data collectors in the working handover/alternation/replacement process of workers is caused. The application aims at the problem, and provides a spectrum video image collector which is arranged on a parabola by combining the intersection point position with a rectangular chemical industry park by drawing a plurality of symmetrical parabolas and adopting a common parabolic quadratic equation formula. And the quantity that the parabola was drawn can combine the size scope of chemical industry garden to set up, and the position around the origin of coordinates was located can be the district key monitoring area, and under the condition that this area collector quantity was many, the timeliness and the validity of early warning help realizing simultaneously. The method enables the data acquisition range covering the chemical industry park to be wider, enables the maintenance of the later-stage collector to be more convenient, has higher practicability and rationality, and provides a high cost performance reference value for the arrangement of the collectors in the chemical industry park.
Claims (9)
1. The chemical industry park safety early warning method based on the video analysis technology is characterized by comprising the following steps:
step 1, carrying out standardization treatment on a chemical industry park area to obtain a rectangular area diagram of a monitoring area;
step 2, taking the position of the central point of the rectangular area diagram as a coordinate origin, taking the horizontal direction as an X axis and the vertical direction as a Y axis, and completing the establishment of a plane rectangular coordinate system;
step 3, drawing in the plane rectangular coordinate systemShape parabolas U1 and->The shape of the parabola U2, the parabola U1 and the parabola U2 are symmetrical, wherein n is a natural number;
step 4, adjusting n to enable two intersection points of the U1 parabola and the rectangular long side to be the length of the rectangular long sideAt the position that the two intersection points of the U2 parabola and the long side of the rectangle are momentsLength of long side of shape->A location;
step 5, the following operations are carried out within the scope of the rectangular area diagram: disposing first spectrum video image collectors at the coordinate origin position of the rectangular region diagram, disposing m second spectrum video image collectors on the U1 parabola at equal intervals, and disposing m third spectrum video image collectors on the U2 parabola at equal intervals;
and 6, based on the spectral video images collected by the first spectral video image collector, the spectral video images collected by the m second spectral video image collectors and the spectral video images collected by the m third spectral video image collectors, identifying and obtaining a detection result of gas leakage in the chemical industry park, and carrying out abnormal alarm according to the detection result, wherein the method specifically comprises the following steps:
the spectral video images acquired by the first spectral video image acquirer, the spectral video images acquired by the m second spectral video image acquirers and the spectral video images acquired by the m third spectral video image acquirers are respectively subjected to video frame feature extraction, and feature segmentation is respectively carried out according to nodes in a fixed time period to obtain time sequence features corresponding to each spectral image acquirer;
aiming at the current time sequence characteristic x corresponding to each spectrum image collector a The similarity measurement is carried out, and specifically comprises the following steps:
for the spectrum image collector at the current position, the h time sequence section in the history database is assumed to be in a state without target dangerous gas, and the image time sequence characteristic of the corresponding spectrum image collector is x h The method comprises the steps of carrying out a first treatment on the surface of the The j-th time sequence segment is assumed to be in a state with target dangerous gas, and the image time sequence characteristic of the corresponding spectrum image collector is x j ;
The degree of similarity D1 is calculated and,,/>representing the 2 norms of the vectors;
the degree of similarity D2 is calculated and,;
comparing the similarity D1 and the similarity D2 with set thresholds respectively to obtain detection results of target dangerous gas leakage, wherein the detection results specifically comprise:
when D1 is smaller than the first threshold value and D2 is larger than the second threshold value, indicating that the target dangerous gas leaks, and performing primary alarm;
when D1 is larger than the first threshold value and D2 is smaller than the second threshold value, no target dangerous gas leakage is indicated;
when D1 is larger than the first threshold value and D2 is larger than the second threshold value, or when D1 is smaller than the first threshold value and D2 is smaller than the second threshold value, the dangerous abnormality is indicated to exist in a large probability, and the secondary alarm is carried out.
2. The method according to claim 1, wherein the step 6 is based on the spectral video images collected by the first spectral video image collector, the spectral video images collected by the m second spectral video image collectors, and the spectral video images collected by the m third spectral video image collectors, and the identifying and obtaining the detection result of the gas leakage in the chemical industry park specifically includes:
the spectral video images acquired by the first spectral video image acquirer, the spectral video images acquired by the m second spectral video image acquirers and the spectral video images acquired by the m third spectral video image acquirers are respectively subjected to video frame feature extraction, and feature segmentation is respectively performed according to nodes of a fixed time period;
then, according to the nodes of the corresponding time periods, the characteristics of the video images of different spectrums after being segmented are combined, and the combined time series characteristics are expressed as P= (x) 1 ,x 2 ,...,x i ) The method comprises the steps of carrying out a first treatment on the surface of the Wherein x is i Representing the first spectral video in an ith time period nodeThe spectrum video images acquired by the image acquisition device, the spectrum video images acquired by the m second spectrum video image acquisition devices and the spectrum video images acquired by the m third spectrum video image acquisition devices are combined in the ith time period node;
and inputting the combined time sequence characteristics into a ResNet50 network to obtain a detection result of gas leakage in the chemical industry park.
3. The method as recited in claim 2, further comprising:
based on the step 5, the following operations are performed within the scope of the rectangular area map: disposing first spectrum video image collectors at the coordinate origin position of the rectangular region diagram, disposing m second spectrum video image collectors on the U1 parabola at equal intervals, and disposing m third spectrum video image collectors on the U2 parabola at equal intervals;
extracting the spatial position characteristic of each spectrum video image collector in the step 5, wherein the spatial position characteristic is the position coordinate of the spectrum video image collector in the plane rectangular coordinate system;
the time sequence features and the spatial position features are overlapped and combined to obtain the fused enhancement features;
and inputting the enhanced characteristics into a ResNet50 network to obtain a detection result of gas leakage in the chemical industry park.
4. A method according to any one of claims 1-3, further comprising:
step 7, drawing in the plane rectangular coordinate systemShape parabolic sum->The shape parabola specifically further comprises:
n is a plurality of numerical values, and a plurality of parabolas which are symmetrical up and down are drawn;
step 8, arranging spectrum video image collectors on the parabolas which are vertically symmetrical in the range of the rectangular area diagram;
and 9, identifying and obtaining a detection result of gas leakage in the chemical industry park based on the spectrum video image acquired by the spectrum video image acquisition device in the step 8.
5. The method as recited in claim 4, further comprising:
respectively extracting video frame characteristics of the spectrum video images acquired by the spectrum video image acquirer in the step 8, and respectively carrying out characteristic segmentation according to nodes in a fixed time period;
and merging the video image features of different spectrums after the nodes are segmented according to the corresponding time period, and inputting the merged time sequence features into a ResNet50 network to obtain a detection result of gas leakage in the chemical industry park.
6. The method as recited in claim 1, further comprising: the detection result specifically comprises detection and identification of gas leakage position, concentration distribution and diffusion trend information in the spectrum video.
7. A method according to claim 2 or 3, wherein the res net50 network is a modified network structure, in particular: a mixed attention mechanism CBAM is added to the res net50 network, specifically the CBAM is placed in the residual block of the res net50 network and named res net50+cbam.
8. Chemical industry park safety precaution system based on video analysis technique, characterized in that, this system includes:
the standardized processing module is used for carrying out standardized processing on the chemical industry park area to obtain a rectangular area diagram of the monitoring area;
the coordinate system establishment module is used for completing establishment of a plane rectangular coordinate system by taking the position of the central point of the rectangular area diagram as a coordinate origin, taking the horizontal direction as an X axis and taking the vertical direction as a Y axis;
a shape drawing module for drawing in the plane rectangular coordinate systemShape parabolas U1 and->The shape of the parabola U2, the parabola U1 and the parabola U2 are symmetrical, wherein n is a natural number;
the adjusting module is used for adjusting n to enable two intersection points of the U1 parabola and the rectangular long side to be the length of the rectangular long sideAt such a position that the two points of intersection of the U2 parabola with the rectangular long side are +.>A location;
the first deployment module is used for carrying out the following operations within the scope of the rectangular area diagram: disposing first spectrum video image collectors at the coordinate origin position of the rectangular region diagram, disposing m second spectrum video image collectors on the U1 parabola at equal intervals, and disposing m third spectrum video image collectors on the U2 parabola at equal intervals;
the first detection module is used for identifying and obtaining detection results of gas leakage in the chemical industry park based on the spectral video images collected by the first spectral video image collector, the spectral video images collected by the m second spectral video image collectors and the spectral video images collected by the m third spectral video image collectors, and carrying out abnormal alarm according to the detection results, and specifically comprises the following steps:
the spectral video images acquired by the first spectral video image acquirer, the spectral video images acquired by the m second spectral video image acquirers and the spectral video images acquired by the m third spectral video image acquirers are respectively subjected to video frame feature extraction, and feature segmentation is respectively carried out according to nodes in a fixed time period to obtain time sequence features corresponding to each spectral image acquirer;
aiming at the current time sequence characteristic x corresponding to each spectrum image collector a The similarity measurement is carried out, and specifically comprises the following steps:
for the spectrum image collector at the current position, the h time sequence section in the history database is assumed to be in a state without target dangerous gas, and the image time sequence characteristic of the corresponding spectrum image collector is x h The method comprises the steps of carrying out a first treatment on the surface of the The j-th time sequence segment is assumed to be in a state with target dangerous gas, and the image time sequence characteristic of the corresponding spectrum image collector is x j ;
The degree of similarity D1 is calculated and,,/>representing the 2 norms of the vectors;
the degree of similarity D2 is calculated and,;
comparing the similarity D1 and the similarity D2 with set thresholds respectively to obtain detection results of target dangerous gas leakage, wherein the detection results specifically comprise:
when D1 is smaller than the first threshold value and D2 is larger than the second threshold value, indicating that the target dangerous gas leaks, and performing primary alarm;
when D1 is larger than the first threshold value and D2 is smaller than the second threshold value, no target dangerous gas leakage is indicated;
when D1 is larger than the first threshold value and D2 is larger than the second threshold value, or when D1 is smaller than the first threshold value and D2 is smaller than the second threshold value, the dangerous abnormality is indicated to exist in a large probability, and the secondary alarm is carried out.
9. The system of claim 8, further comprising:
a plurality of symmetrical parabolic drawing modules for drawing in the plane rectangular coordinate systemShape parabolaThe shape parabola specifically further comprises:
n is a plurality of numerical values, and a plurality of parabolas which are symmetrical up and down are drawn;
the second deployment module is used for arranging the spectrum video image collectors on the parabolas which are symmetrical up and down in the range of the rectangular area diagram;
and the second detection module is used for identifying and obtaining a detection result of gas leakage in the chemical industry park based on the spectral video image acquired by the spectral video image acquisition device in the second deployment module.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310494640.1A CN116486307B (en) | 2023-05-05 | 2023-05-05 | Chemical industry park safety early warning method and system based on video analysis technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310494640.1A CN116486307B (en) | 2023-05-05 | 2023-05-05 | Chemical industry park safety early warning method and system based on video analysis technology |
Publications (2)
Publication Number | Publication Date |
---|---|
CN116486307A CN116486307A (en) | 2023-07-25 |
CN116486307B true CN116486307B (en) | 2023-11-21 |
Family
ID=87213657
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310494640.1A Active CN116486307B (en) | 2023-05-05 | 2023-05-05 | Chemical industry park safety early warning method and system based on video analysis technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116486307B (en) |
Families Citing this family (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN118365079A (en) * | 2024-05-06 | 2024-07-19 | 吉林化工学院 | Industrial park risk early warning system based on image recognition |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101751744A (en) * | 2008-12-10 | 2010-06-23 | 中国科学院自动化研究所 | Detection and early warning method of smoke |
CN109716753A (en) * | 2016-06-17 | 2019-05-03 | 亚历山大·库尔泰 | Use the image capture method and system of virtual-sensor |
CN110133202A (en) * | 2019-05-20 | 2019-08-16 | 梁志鹏 | A kind of method and device of the food materials monitoring freshness of intelligent kitchen |
CN111122490A (en) * | 2020-01-19 | 2020-05-08 | 北京戴纳实验科技有限公司 | Indoor gas leakage information acquisition method and device |
CN111965116A (en) * | 2020-07-21 | 2020-11-20 | 天津大学 | Hyperspectrum-based airport gas detection system and method |
CN112767644A (en) * | 2020-12-31 | 2021-05-07 | 千方捷通科技股份有限公司 | Method and device for early warning of fire in highway tunnel based on video identification |
CN113688801A (en) * | 2021-10-22 | 2021-11-23 | 南京智谱科技有限公司 | Chemical gas leakage detection method and system based on spectrum video |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US10955355B2 (en) * | 2017-02-22 | 2021-03-23 | Rebellion Photonics, Inc. | Systems and methods for monitoring remote installations |
-
2023
- 2023-05-05 CN CN202310494640.1A patent/CN116486307B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101751744A (en) * | 2008-12-10 | 2010-06-23 | 中国科学院自动化研究所 | Detection and early warning method of smoke |
CN109716753A (en) * | 2016-06-17 | 2019-05-03 | 亚历山大·库尔泰 | Use the image capture method and system of virtual-sensor |
CN110133202A (en) * | 2019-05-20 | 2019-08-16 | 梁志鹏 | A kind of method and device of the food materials monitoring freshness of intelligent kitchen |
CN111122490A (en) * | 2020-01-19 | 2020-05-08 | 北京戴纳实验科技有限公司 | Indoor gas leakage information acquisition method and device |
CN111965116A (en) * | 2020-07-21 | 2020-11-20 | 天津大学 | Hyperspectrum-based airport gas detection system and method |
CN112767644A (en) * | 2020-12-31 | 2021-05-07 | 千方捷通科技股份有限公司 | Method and device for early warning of fire in highway tunnel based on video identification |
CN113688801A (en) * | 2021-10-22 | 2021-11-23 | 南京智谱科技有限公司 | Chemical gas leakage detection method and system based on spectrum video |
Non-Patent Citations (1)
Title |
---|
"工业气体泄漏红外成像检测及差分光谱滤波检测方法研究";袁盼等;《红外与激光工程》;第51卷(第8期);全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN116486307A (en) | 2023-07-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Jiang et al. | Insulator fault detection in aerial images based on ensemble learning with multi-level perception | |
CN109858367B (en) | Visual automatic detection method and system for worker through supporting unsafe behaviors | |
Su et al. | RCAG-Net: Residual channelwise attention gate network for hot spot defect detection of photovoltaic farms | |
Wang et al. | High-voltage power transmission tower detection based on faster R-CNN and YOLO-V3 | |
Luo et al. | An ultrasmall bolt defect detection method for transmission line inspection | |
CN116486307B (en) | Chemical industry park safety early warning method and system based on video analysis technology | |
Zhao et al. | Detection and location of safety protective wear in power substation operation using wear-enhanced YOLOv3 algorithm | |
Lianqiao et al. | Recognition and application of infrared thermal image among power facilities based on yolo | |
Wang et al. | Rail fastener automatic recognition method in complex background | |
CN116846059A (en) | Edge detection system for power grid inspection and monitoring | |
CN118053118B (en) | Comprehensive monitoring, evaluating and maintaining method and system for photovoltaic performance | |
Wang et al. | Research and application of artificial technology for substation environment surveillance system | |
Wang et al. | Worker’s helmet recognition and identity recognition based on deep learning | |
CN114372922A (en) | Insulator infrared defect diagnosis method based on unmanned aerial vehicle routing inspection under night environment | |
Dai et al. | Cascade centernet: Robust object detection for power line surveillance | |
CN113724247B (en) | Intelligent substation inspection method based on image discrimination technology | |
Mu et al. | UAV image defect detection method for steel structure of high-speed railway bridge girder | |
Li et al. | Automatic infrared image recognition method for substation equipment based on a deep self-attention network and multi-factor similarity calculation | |
Zou et al. | Research on detection of transmission line corridor external force object containing random feature targets | |
Hou et al. | A Method of Shear Line Automatic Identification on High-precision Numerical Data | |
Wang et al. | Foreign Object Detection Network for Transmission Lines from Unmanned Aerial Vehicle Images. | |
Yu et al. | Research on defect recognition method of substation inspection images based on faster R-CNN | |
Ma et al. | Research on fault identification of ship PV system based on aerial photography image recognition | |
Zhan | Electric equipment inspection on high voltage transmission line via Mobile Net-SSD | |
Huang et al. | Detection of abnormal dangerous behavior in electrical power grids based on the fusion of contextual features |
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 |