CN111325721A - Gas leakage detection method and system based on infrared thermal imaging - Google Patents
Gas leakage detection method and system based on infrared thermal imaging Download PDFInfo
- Publication number
- CN111325721A CN111325721A CN202010090294.7A CN202010090294A CN111325721A CN 111325721 A CN111325721 A CN 111325721A CN 202010090294 A CN202010090294 A CN 202010090294A CN 111325721 A CN111325721 A CN 111325721A
- Authority
- CN
- China
- Prior art keywords
- image
- gas
- current frame
- gas leakage
- images
- 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.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 42
- 238000001931 thermography Methods 0.000 title claims abstract description 24
- 238000000034 method Methods 0.000 claims abstract description 25
- 238000007781 pre-processing Methods 0.000 claims abstract description 20
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000012937 correction Methods 0.000 claims description 27
- 238000001914 filtration Methods 0.000 claims description 19
- 230000002146 bilateral effect Effects 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 10
- 238000012706 support-vector machine Methods 0.000 claims description 10
- 230000011218 segmentation Effects 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 4
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 230000000007 visual effect Effects 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 2
- 238000012847 principal component analysis method Methods 0.000 claims description 2
- 238000010801 machine learning Methods 0.000 abstract description 11
- 238000005516 engineering process Methods 0.000 abstract description 9
- 239000007789 gas Substances 0.000 description 91
- 230000006872 improvement Effects 0.000 description 8
- 238000012549 training Methods 0.000 description 5
- 238000013106 supervised machine learning method Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000005259 measurement Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000001965 increasing effect Effects 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000002123 temporal effect Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 238000013107 unsupervised machine learning method Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 239000003245 coal Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000003331 infrared imaging Methods 0.000 description 1
- 230000001788 irregular Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 239000003208 petroleum Substances 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000005057 refrigeration Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- 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
-
- 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
- G06F18/213—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods
- G06F18/2135—Feature extraction, e.g. by transforming the feature space; Summarisation; Mappings, e.g. subspace methods based on approximation criteria, e.g. principal component analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/97—Determining parameters from multiple pictures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
Abstract
The invention discloses a gas leakage detection method and a system based on infrared thermal imaging, wherein the method comprises the following steps: collecting gas leakage infrared video sequence images to obtain a gas infrared image sequence; preprocessing the acquired current frame image and the acquired previous N frame images frame by frame; performing digital detail enhancement on the preprocessed current frame image and the preprocessed previous N frame images; acquiring a suspected gas leakage area in the current frame image by adopting combined inter-frame difference and background difference for the current frame image and the previous N frames of images after digital detail enhancement; extracting the gas cloud cluster characteristics of the gas leakage suspected area in the current frame image; carrying out feature selection on the extracted gas cloud cluster features of the suspected gas area, and selecting the main features most related to the detected gas; and inputting a pre-trained classifier to position, identify and detect the gas leakage area. The method combines infrared thermal imaging image processing and machine learning technology to realize accurate positioning and non-contact detection of the gas leakage position.
Description
Technical Field
The invention relates to the technical field of image processing and gas detection, in particular to a gas leakage detection method and system based on infrared thermal imaging.
Background
The increasing global demand for energy makes the industrialization scale continuously expanding, especially in the industries of petroleum, chemical industry, coal mine, automobile and the like, and the pollution to the environment is increasingly serious due to the increase of harmful gases. How to detect the existence of gas leakage fast, pinpoint the gas leakage source to prevent that the emergence of major gas leakage accident becomes the problem that needs to solve urgently.
The infrared thermal imaging technology has the obvious advantages of high efficiency, long distance, large range, dynamic intuition and the like, and becomes an important development direction of the gas leakage detection technology. However, the image formed after the gas infrared imaging has low concentration, low contrast, no fixed shape and volume, irregular fluidity and easy interference of various radiation source factors, and the detection difficulty of the leaked gas is increased. The existing infrared image enhancement algorithm is not sufficiently researched, and how to improve the gas leakage detection precision and realize the accurate measurement of the gas leakage becomes a challenging problem of the gas leakage detection.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, adopts a non-refrigeration infrared detector, provides a gas leakage detection method based on infrared thermal imaging, combines an infrared image processing technology and an artificial intelligence technology, and realizes real-time detection and high-precision positioning of leaked gas.
In order to achieve the above object, the present invention discloses a gas leakage detection method based on infrared thermal imaging, the method comprising:
collecting gas leakage infrared video sequence images to obtain a gas infrared image sequence; preprocessing the acquired current frame image and the acquired previous N frame images frame by frame;
performing digital detail enhancement on the preprocessed current frame image and the preprocessed previous N frame images;
acquiring a suspected gas leakage area in the current frame image by adopting combined inter-frame difference and background difference for the current frame image and the previous N frames of images after digital detail enhancement;
extracting the gas cloud cluster characteristics of the gas leakage suspected area in the current frame image; carrying out feature selection on the extracted gas cloud cluster features of the suspected gas area, and selecting the main features most related to the detected gas; and inputting the main characteristics most relevant to the detected gas into a pre-trained classifier, and positioning, identifying and detecting the gas leakage area.
As an improvement of the above method, the preprocessing the acquired current frame image and previous N frame images frame by frame specifically includes:
carrying out non-uniformity correction on the current frame image and the previous N frames of images frame by frame to remove fixed pattern noise;
and filtering the current frame image and the previous N frames of images with the fixed pattern noise removed one by one to remove noise interference.
As an improvement of the above method, the non-uniformity correction employs a one-point correction algorithm, a two-point correction algorithm, or a scene-based non-uniformity correction algorithm.
As an improvement of the above method, the filtering, frame by frame, the current frame image and the previous N frame image from which the fixed pattern noise is removed to remove the noise interference specifically includes:
for the image without fixed pattern noise, an improved bilateral filtering algorithm is adopted, and the corresponding improved bilateral filtering output image is expressed as follows:
w(i,j)＝ws(i,j)*wr(i,j)*wt(i,j)
wherein, Ic(I, j) is an image from which fixed pattern noise is removed, Is(i, j) is the output image signal of the bilateral filter, w (i, j) is the total template, ws(i, j) is a spatial template, wr(i, j) is a luminance template, wt(i, j) is the gradient template, (i, j) is the current image pixel location, sx,yRefers to the range of the filter window centered at (i, j)。
As an improvement of the method, the method comprises the steps of obtaining a suspected gas leakage area in the current frame image by adopting a combined interframe difference and a background difference for the current frame image and the previous N frames of images after the digital details are enhanced; the method specifically comprises the following steps:
averaging the previous N frames of images to obtain an initial background image;
updating the real-time background image by adopting a background updating algorithm to obtain an updated background image;
differentiating the current frame image and the updated background image to obtain a binary image, and performing morphological filtering on the binary image to obtain a background differential image DBk；
Performing interframe difference on the previous N frames OF images to obtain a multiframe interframe difference image OFk；
The background difference image DBkAnd multiframe interframe differential image OFkAnd performing logical AND operation to obtain a gas leakage suspected area in the current frame image.
As an improvement of the above method, the background update algorithm comprises a joint visual background extraction algorithm and an edge extraction algorithm.
As an improvement of the above method, the gas cloud cluster features include time domain features and wavelet domain features, wherein the time domain features include moment features, geometric features, gray scale features, texture features of gray scale co-occurrence matrices, and luminance histogram features; the wavelet domain features are wavelet high-frequency energy features.
As an improvement of the above method, the characteristic of the gas cloud in the extracted suspected gas area is selected, and a principal component analysis method is adopted.
As an improvement of the above method, the classifier is a non-linear prediction model of the support vector machine, the model uses the main features most relevant to the measured gas as input variables, uses the corresponding gas leakage type as output variables, and uses the kernel function of the support vector machine to find the non-linear mapping relation between the input variables and the output variables.
The invention also provides a gas leakage detection system based on infrared thermal imaging, which comprises: the system comprises an image preprocessing module, a digital detail enhancement module, a gas region segmentation module and a gas leakage detection module;
the image preprocessing module is used for acquiring gas leakage infrared video sequence images to obtain a gas infrared image sequence; preprocessing the acquired current frame image and the acquired previous N frame images frame by frame;
the digital detail enhancement module is used for carrying out digital detail enhancement on the preprocessed current frame image and the preprocessed previous N frames of images;
the gas area segmentation module is used for acquiring a gas leakage suspected area in the current frame image by adopting combined interframe difference and background difference for the current frame image and the previous N frames of images after digital detail enhancement;
the gas leakage detection module is used for extracting the gas cloud cluster characteristics of the suspected gas leakage area in the current frame image; carrying out feature selection on the extracted gas cloud cluster features of the suspected gas area, and selecting the main features most related to the detected gas; and inputting the main characteristics most relevant to the detected gas into a pre-trained classifier, and positioning, identifying and detecting the gas leakage area.
Compared with the prior art, the invention has the following advantages and prominent technical effects:
1. the method of the invention combines the infrared thermal imaging image processing technology and the machine learning technology to realize the accurate positioning and the non-contact detection of the gas leakage position;
2. the method provided by the invention effectively retains background information, detects and extracts a moving gas cloud target, performs gas cloud characteristic analysis on the basis, adopts a machine learning algorithm to realize high-precision positioning and measurement of gas, and overcomes the defects of low gas leakage detection precision and low efficiency.
Drawings
FIG. 1 is a schematic diagram of a gas leak detection method based on infrared thermal imaging according to the present invention;
fig. 2 is a flow chart of the digital detail enhancement algorithm of the present invention.
Detailed Description
The technical solution of the present invention will be described in detail below with reference to the accompanying drawings.
As shown in fig. 1, embodiment 1 of the present invention proposes a gas leak detection method based on infrared thermal imaging, including the steps of:
step 1) collecting infrared video sequence images of gas leakage to obtain a gas infrared image sequence; carrying out infrared image preprocessing on the obtained current frame image and the previous N frames of images frame by frame to remove noise in the images;
the preprocessing comprises non-uniformity correction and filtering, and specifically comprises the following steps:
step 2-1) carrying out non-uniformity correction on the infrared image, removing fixed pattern noise and obtaining an image I after non-uniformity correctionc；
The non-uniformity correction may employ, but is not limited to, a one-point correction algorithm, a two-point correction algorithm, and a scene-based non-uniformity correction algorithm.
Two-point correction algorithm is adopted:
the response characteristics of the array elements of the infrared detector are assumed to be linear. Then use the two temperature points phi2And phi1Calculating a correction gain k of the detector responseijAnd offset bijAnd (4) parameters.
The correction gain and offset can be expressed as:
and carrying out non-uniformity correction by using the obtained parameters, wherein the output of each pixel can be corrected to be an actual gray value under any radiation illumination phi.
Step 2-2) for image IcFiltering to remove noise and other interferences and obtain a denoised image Is；
The filtering processing is carried out by improving a bilateral filtering algorithm, namely gradient information is introduced into bilateral filtering, so that the accurate judgment of the algorithm on the image boundary is improved, and the denoising capability of the algorithm on a smooth region is enhanced.
Given an input image, its corresponding improved bilateral filtered output image is represented as follows:
w(i,j)＝ws(i,j)*wr(i,j)*wt(i,j)
wherein I (I, j) is the original input image signal, IBF(i, j) is the output image signal of the bilateral filter, w (i, j) is the total template, ws(i, j) is a spatial template, wr(i, j) is a luminance template, wt(i, j) is the gradient template, (i, j) is the current image pixel location, sx,yRefers to the range of the filter window centered at (i, j).
Step 2) for each frame of preprocessed image, adopting a digital detail enhancement algorithm, not only keeping whole scene perception in the low dynamic range image, but also highlighting detail information of important targets in the image;
the digital detail enhancement algorithm adopts a processing framework based on image layering to decompose a gas infrared image into a low-frequency signal and a high-frequency signal, the low-frequency signal and the high-frequency signal respectively correspond to a basic layer and a detail layer, then the basic layer adopts a self-adaptive double-platform histogram equalization algorithm, and linear weighting is carried out on the basic layer and the detail layer signals to obtain an image after detail enhancement.
Step 3) acquiring a suspected gas leakage area from the sequence infrared image processed by the preprocessing and the digital enhancement algorithm by adopting a method of combining interframe difference and background difference; the method comprises the following steps:
step 3-1), selecting the previous N frames of images to average to obtain an initial background image;
step 3-2) after the initial background is obtained, carrying out real-time background updating in subsequent operation, and continuously obtaining an updated background image;
the background updating algorithm adopts a background difference method, combines a visual background extraction (ViBE) algorithm and an edge extraction algorithm, and updates the background.
Step 3-3) differentiating the current frame image and the background image to obtain a binary image, and then performing morphological filtering processing on the binary image to obtain an image DBk；
Step 3-4) carrying out interframe difference on the previous N frames OF images to obtain an image OFk；
Step 3-5) background difference image DBkAnd multiframe interframe differential image OFkAnd performing logical AND operation to obtain a suspected gas leakage area.
Step 4) realizing accurate positioning and detection of the gas leakage position by adopting a machine learning method; the method specifically comprises the following steps:
step 4-1), performing high-dimensional feature extraction on the suspected gas area;
the high-dimensional features include, but are not limited to, time domain features and wavelet domain features, wherein the time domain features include, but are not limited to, moment features, geometric features, gray scale features, texture features of gray scale co-occurrence matrices, luminance histogram features, and the like. The wavelet domain features refer to wavelet high-frequency energy features. Specifically, as shown in table 1:
table 1: list of extracted feature parameters
Step 4-2) feature selection is carried out on the extracted high-dimensional features, and the main features most relevant to the gas to be detected are selected by adopting a principal component analysis algorithm;
and 4-3) training and learning the selected main characteristics, and identifying whether the suspected gas area is the gas leakage area to be detected by adopting a machine learning method. The machine learning method includes, but is not limited to, supervised machine learning and unsupervised machine learning methods.
By way of example of a supervised machine learning method, in the present embodiment, a Support Vector Machine (SVM) nonlinear prediction is used as a model, each main feature for distinguishing gas leakage from non-leakage is used as an input variable, a corresponding gas leakage type is used as an output variable, and a kernel function of the support vector machine is used to find a nonlinear mapping relationship between the input variable and the output variable.
Dividing sample data consisting of gas leakage types and main influence characteristic factors thereof into a training sample set and a testing sample set, performing optimization training of model parameters and construction of a detection model by using the training sample set, and analyzing detection and recognition results and evaluating detection performance by using the testing sample set.
The invention combines the infrared thermal imaging image processing technology and the machine learning technology to realize the accurate positioning and the non-contact detection of the gas leakage position. The method has the advantages that background information is effectively kept, meanwhile, the moving gas cloud target is detected and extracted, on the basis, gas cloud characteristic analysis is carried out, a machine learning algorithm is adopted, high-precision positioning and measurement of gas are achieved, and the defects of low gas leakage detection precision and low efficiency are overcome.
The embodiment 2 of the invention provides a gas leakage detection system based on infrared thermal imaging, which comprises an image preprocessing module, a digital detail enhancing module, a gas region segmentation module and a gas leakage detection module, wherein the image preprocessing module is used for preprocessing a gas region;
image preprocessing for removing fixed pattern noise and other redundant noise of the gas infrared image; the device comprises an infrared image non-uniformity correction unit and a single-frame image noise reduction unit;
the non-uniformity correction unit adopts a two-point correction method, which comprises the following steps:
the response characteristics of the array elements of the infrared detector are assumed to be linear. Then use the two temperature points phi2And phi1Calculating a correction gain k of the detector responseijAnd offset bijAnd (4) parameters.
The correction gain and offset can be expressed as:
and carrying out non-uniformity correction by using the obtained parameters, wherein the output of each pixel can be corrected to be an actual gray value under any radiation illumination phi.
And the single-frame image denoising unit is used for improving the bilateral filtering algorithm, namely introducing gradient information into the bilateral filtering algorithm to improve the accurate judgment of the algorithm on the image boundary and enhance the denoising capability of the algorithm on a smooth region. Given an input image, its corresponding improved bilateral filtered output image is represented as follows:
w(i,j)＝ws(i,j)*wr(i,j)*wt(i,j)
wherein I (I, j) is the original input image signal, IBF(i, j) is the output image signal of the bilateral filter, w (i, j) is the total template, ws(i, j) is a spatial template, wr(i, j) is a luminance template, wt(i, j) is a gradient template.
The digital detail enhancement module further highlights detail information of important targets in the image on the basis of image preprocessing; specifically, the digital detail enhancement algorithm adopts a processing framework based on image layering to decompose the gas infrared image into a low-frequency signal and a high-frequency signal, which respectively correspond to the base layer and the detail layer, and then adopts a self-adaptive dual-platform histogram equalization algorithm for the base layer, and then performs linear weighting with the detail layer signal to obtain an image after detail enhancement, as shown in fig. 2.
The gas area segmentation module is used for acquiring a suspected gas leakage area from the sequence infrared image processed by the preprocessing and the digital detail enhancement algorithm by adopting a method of combining interframe difference and background difference, and comprises the following steps of:
(1) selecting the previous N frames of images to average to obtain an initial background image;
(2) after the initial background is obtained, performing real-time background updating in subsequent operation to continuously obtain an updated background image;
the background updating algorithm adopts a combined visual background extraction algorithm and an edge extraction algorithm to update the background.
(3) Differentiating the current frame image and the background image to obtain a binary image, and performing morphological filtering processing on the binary image to obtain an image DBk；
(4) Performing interframe difference on the previous N frames OF images to obtain an image OFk；
(5) The background difference image DBkAnd multiframe interframe differential image OFkAnd performing logical AND operation to obtain a suspected gas leakage area.
And the gas leakage detection module is used for realizing accurate positioning and detection of a gas leakage position by adopting a machine learning method.
Specifically, the gas leakage detection module comprises the following specific steps:
(a) performing high-dimensional feature extraction on the suspected gas leakage area;
high-dimensional features include, but are not limited to, temporal domain features and wavelet domain features, wherein temporal domain features include, but are not limited to, moment features, geometric features, grayscale features, luminance histogram features, and the like. The wavelet domain features refer to wavelet high-frequency energy features. (b) Adopting a feature dimension reduction algorithm for the extracted high-dimensional features, and selecting the main features most relevant to the gas to be detected; and (c) adopting a principal component analysis algorithm as the high-dimensional feature dimension reduction algorithm in the step (b).
(c) And training and learning the selected main characteristics, and identifying whether the suspected gas leakage area is the gas leakage area to be detected by adopting a machine learning method. The machine learning method includes, but is not limited to, supervised machine learning and unsupervised machine learning methods.
By way of example of a supervised machine learning method, in the present embodiment, a Support Vector Machine (SVM) nonlinear prediction is used as a model, each main feature for distinguishing gas leakage from non-leakage is used as an input variable, a corresponding gas leakage type is used as an output variable, and a kernel function of the support vector machine is used to find a nonlinear mapping relationship between the input variable and the output variable.
Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention and are not limited. Although the present invention has been described in detail with reference to the embodiments, it will be understood by those skilled in the art that various changes may be made and equivalents may be substituted without departing from the spirit and scope of the invention as defined in the appended claims.
Claims (10)
1. A method of gas leak detection based on infrared thermal imaging, the method comprising:
collecting gas leakage infrared video sequence images to obtain a gas infrared image sequence; preprocessing the acquired current frame image and the acquired previous N frame images frame by frame;
performing digital detail enhancement on the preprocessed current frame image and the preprocessed previous N frame images;
acquiring a suspected gas leakage area in the current frame image by adopting combined inter-frame difference and background difference for the current frame image and the previous N frames of images after digital detail enhancement;
extracting the gas cloud cluster characteristics of the gas leakage suspected area in the current frame image; carrying out feature selection on the extracted gas cloud cluster features of the suspected gas area, and selecting the main features most related to the detected gas; and inputting the main characteristics most relevant to the detected gas into a pre-trained classifier, and positioning, identifying and detecting the gas leakage area.
2. The infrared thermal imaging-based gas leak detection method according to claim 1, wherein the preprocessing the acquired current frame image and the previous N frame images frame by frame specifically comprises:
carrying out non-uniformity correction on the current frame image and the previous N frames of images frame by frame to remove fixed pattern noise;
and filtering the current frame image and the previous N frames of images with the fixed pattern noise removed one by one to remove noise interference.
3. The infrared thermal imaging-based gas leak detection method according to claim 2, wherein the non-uniformity correction employs a one-point correction algorithm, a two-point correction algorithm, or a scene-based non-uniformity correction algorithm.
4. The infrared thermal imaging-based gas leakage detection method according to claim 2, wherein the filtering processing is performed on the current frame image and the previous N frame images from which the fixed pattern noise is removed frame by frame to remove noise interference, specifically comprising:
for the image without fixed pattern noise, an improved bilateral filtering algorithm is adopted, and the corresponding improved bilateral filtering output image is expressed as follows:
w(i,j)＝ws(i,j)*wr(i,j)*wt(i,j)
wherein, Ic(I, j) is an image from which fixed pattern noise is removed, Is(i, j) is the output image signal of the bilateral filter, w (i, j) is the total template, ws(i, j) is a spatial template, wr(i, j) is a luminance template, wt(i, j) is the gradient template, (i, j) is the current image pixel location, sx,yRefers to the range of the filter window centered at (i, j).
5. The infrared thermal imaging-based gas leakage detection method according to claim 1, wherein the gas leakage suspected area in the current frame image is obtained by using a combined inter-frame difference and background difference for the current frame image and the previous N frame images after digital detail enhancement; the method specifically comprises the following steps:
averaging the previous N frames of images to obtain an initial background image;
updating the real-time background image by adopting a background updating algorithm to obtain an updated background image;
differentiating the current frame image and the updated background image to obtain a binary image, and performing morphological filtering on the binary image to obtain a background differential image DBk；
Performing interframe difference on the previous N frames OF images to obtain a multiframe interframe difference image OFk；
The background difference image DBkAnd multiframe interframe differential image OFkAnd performing logical AND operation to obtain a gas leakage suspected area in the current frame image.
6. The infrared thermal imaging-based gas leak detection method according to claim 5, wherein the background update algorithm comprises a joint visual background extraction algorithm and an edge extraction algorithm.
7. The infrared thermal imaging-based gas leak detection method according to claim 1, wherein the gas cloud features include time domain features and wavelet domain features, wherein the time domain features include moment features, geometric features, grayscale features, texture features of grayscale co-occurrence matrices, and luminance histogram features; the wavelet domain features are wavelet high-frequency energy features.
8. The infrared thermal imaging-based gas leak detection method according to claim 1, wherein the gas cloud characteristics of the extracted suspected gas area are selected by a principal component analysis method.
9. The infrared thermal imaging-based gas leak detection method according to claim 1, wherein the classifier is a nonlinear predictive model of a support vector machine, the model uses the main features most relevant to the measured gas as input variables, uses the corresponding gas leak type as output variables, and uses a kernel function of the support vector machine to seek a nonlinear mapping relation between the input variables and the output variables.
10. A gas leak detection system based on infrared thermal imaging, the system comprising: the system comprises an image preprocessing module, a digital detail enhancement module, a gas region segmentation module and a gas leakage detection module;
the image preprocessing module is used for acquiring gas leakage infrared video sequence images to obtain a gas infrared image sequence; preprocessing the acquired current frame image and the acquired previous N frame images frame by frame;
the digital detail enhancement module is used for carrying out digital detail enhancement on the preprocessed current frame image and the preprocessed previous N frames of images;
the gas area segmentation module is used for acquiring a gas leakage suspected area in the current frame image by adopting combined interframe difference and background difference for the current frame image and the previous N frames of images after digital detail enhancement;
the gas leakage detection module is used for extracting the gas cloud cluster characteristics of the suspected gas leakage area in the current frame image; carrying out feature selection on the extracted gas cloud cluster features of the suspected gas area, and selecting the main features most related to the detected gas; and inputting the main characteristics most relevant to the detected gas into a pre-trained classifier, and positioning, identifying and detecting the gas leakage area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010090294.7A CN111325721A (en) | 2020-02-13 | 2020-02-13 | Gas leakage detection method and system based on infrared thermal imaging |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010090294.7A CN111325721A (en) | 2020-02-13 | 2020-02-13 | Gas leakage detection method and system based on infrared thermal imaging |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111325721A true CN111325721A (en) | 2020-06-23 |
Family
ID=71172737
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010090294.7A Pending CN111325721A (en) | 2020-02-13 | 2020-02-13 | Gas leakage detection method and system based on infrared thermal imaging |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111325721A (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112213244A (en) * | 2020-10-27 | 2021-01-12 | 中国科学院合肥物质科学研究院 | Motor vehicle exhaust ringer-Mannheim blackness measuring device and method based on machine learning |
| CN112444521A (en) * | 2020-10-23 | 2021-03-05 | 广西电网有限责任公司电力科学研究院 | SF (sulfur hexafluoride)6Leakage monitoring method and system |
| CN112634271A (en) * | 2021-03-09 | 2021-04-09 | 南京智谱科技有限公司 | Method and equipment for determining critical value of gas leakage amount detected by infrared camera |
| CN112672081A (en) * | 2020-12-02 | 2021-04-16 | 北京智慧共享技术服务有限公司 | Artificial intelligence online industrial gas leakage imaging monitoring system |
| CN112733646A (en) * | 2020-12-30 | 2021-04-30 | 大连海事大学 | Liquid medium leakage automatic detection method and system based on thermal imaging |
| CN112766084A (en) * | 2020-12-31 | 2021-05-07 | 安徽砺剑防务科技有限公司 | Gas cloud cluster real-time identification method for remotely monitoring toxic explosive gas |
| CN112781791A (en) * | 2020-12-30 | 2021-05-11 | 大连海事大学 | VOCs gas leakage detection method and system based on optical gas imaging |
| CN113654732A (en) * | 2021-08-17 | 2021-11-16 | 广西电网有限责任公司电力科学研究院 | Low-cost sulfur hexafluoride gas leakage infrared imaging detection system |
| CN113780135A (en) * | 2021-08-31 | 2021-12-10 | 中国科学技术大学先进技术研究院 | Cross-scene VOCs gas leakage detection method and system and storage medium |
| CN113963513A (en) * | 2021-10-13 | 2022-01-21 | 公安部第三研究所 | Robot system for realizing intelligent inspection in chemical industry and control method thereof |
| CN114459708A (en) * | 2022-02-10 | 2022-05-10 | 合肥永信科翔智能技术有限公司 | Gas leakage monitoring system based on intelligent gas sensing |
| CN115018737A (en) * | 2022-08-04 | 2022-09-06 | 四川迪晟新达类脑智能技术有限公司 | Infrared thermal image enhancement method and device |
| CN116091491A (en) * | 2023-04-03 | 2023-05-09 | 中国科学技术大学先进技术研究院 | VOCs gas detection method, device, equipment and computer readable storage medium |
| CN116543241A (en) * | 2023-07-07 | 2023-08-04 | 杭州海康威视数字技术股份有限公司 | Detection method and device for leakage gas cloud, storage medium and electronic equipment |
| CN116823839A (en) * | 2023-08-31 | 2023-09-29 | 梁山中维热力有限公司 | Pipeline leakage detection method based on thermal infrared image |
| CN117036347A (en) * | 2023-10-08 | 2023-11-10 | 山东柯林瑞尔管道工程有限公司 | Visual detection method and system for pipeline lining leakage based on image enhancement |
| CN117094995A (en) * | 2023-10-18 | 2023-11-21 | 中国恩菲工程技术有限公司 | Reaction kettle gas leakage detection method, device, medium and equipment |
| CN117132948A (en) * | 2023-10-27 | 2023-11-28 | 南昌理工学院 | Scenic spot tourist flow monitoring method, system, readable storage medium and computer |
| KR20230168243A (en) * | 2022-06-03 | 2023-12-13 | 울산과학기술원 | Method providing composite video for detection of gas leak and artificial intelligence based gas leak detection device performing the same |
| CN117788466A (en) * | 2024-02-26 | 2024-03-29 | 国科大杭州高等研究院 | Uncooled infrared video sequence dangerous gas imaging leakage detection method |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102798503A (en) * | 2012-08-02 | 2012-11-28 | 北京理工大学 | System for imaging detection of gas leakage based on broadband uncooled focal plane detector |
| CN103217397A (en) * | 2013-01-23 | 2013-07-24 | 山西省电力公司晋城供电分公司 | SF6 gas detection method based on infrared image processing |
| US20160320296A1 (en) * | 2015-04-28 | 2016-11-03 | Konica Minolta, Inc. | Gas detector |
-
2020
- 2020-02-13 CN CN202010090294.7A patent/CN111325721A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102798503A (en) * | 2012-08-02 | 2012-11-28 | 北京理工大学 | System for imaging detection of gas leakage based on broadband uncooled focal plane detector |
| CN103217397A (en) * | 2013-01-23 | 2013-07-24 | 山西省电力公司晋城供电分公司 | SF6 gas detection method based on infrared image processing |
| US20160320296A1 (en) * | 2015-04-28 | 2016-11-03 | Konica Minolta, Inc. | Gas detector |
Non-Patent Citations (3)
| Title |
|---|
| 刘路民根 等: "一种基于形状的红外图像泄漏气体检测方法", 《应用光学》 * |
| 张福弘: "基于FPGA的无挡片红外图像去噪及增强处理研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 王建平 等: "基于红外成像的乙烯气体泄漏检测", 《液晶与显示》 * |
Cited By (32)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112444521B (en) * | 2020-10-23 | 2023-06-13 | 广西电网有限责任公司电力科学研究院 | SF (sulfur hexafluoride) 6 Leakage monitoring method and system |
| CN112444521A (en) * | 2020-10-23 | 2021-03-05 | 广西电网有限责任公司电力科学研究院 | SF (sulfur hexafluoride)6Leakage monitoring method and system |
| CN112213244A (en) * | 2020-10-27 | 2021-01-12 | 中国科学院合肥物质科学研究院 | Motor vehicle exhaust ringer-Mannheim blackness measuring device and method based on machine learning |
| CN112213244B (en) * | 2020-10-27 | 2023-10-10 | 中国科学院合肥物质科学研究院 | Device and method for measuring ringeman blackness of motor vehicle tail gas based on machine learning |
| CN112672081A (en) * | 2020-12-02 | 2021-04-16 | 北京智慧共享技术服务有限公司 | Artificial intelligence online industrial gas leakage imaging monitoring system |
| CN112781791A (en) * | 2020-12-30 | 2021-05-11 | 大连海事大学 | VOCs gas leakage detection method and system based on optical gas imaging |
| CN112733646B (en) * | 2020-12-30 | 2021-11-23 | 大连海事大学 | Liquid medium leakage automatic detection method and system based on thermal imaging |
| CN112733646A (en) * | 2020-12-30 | 2021-04-30 | 大连海事大学 | Liquid medium leakage automatic detection method and system based on thermal imaging |
| CN112766084A (en) * | 2020-12-31 | 2021-05-07 | 安徽砺剑防务科技有限公司 | Gas cloud cluster real-time identification method for remotely monitoring toxic explosive gas |
| CN112634271B (en) * | 2021-03-09 | 2021-06-18 | 南京智谱科技有限公司 | Method and equipment for determining critical value of gas leakage amount detected by infrared camera |
| CN112634271A (en) * | 2021-03-09 | 2021-04-09 | 南京智谱科技有限公司 | Method and equipment for determining critical value of gas leakage amount detected by infrared camera |
| CN113654732A (en) * | 2021-08-17 | 2021-11-16 | 广西电网有限责任公司电力科学研究院 | Low-cost sulfur hexafluoride gas leakage infrared imaging detection system |
| CN113654732B (en) * | 2021-08-17 | 2023-12-01 | 广西电网有限责任公司电力科学研究院 | Low-cost sulfur hexafluoride gas leakage infrared imaging detection system |
| CN113780135B (en) * | 2021-08-31 | 2023-08-04 | 中国科学技术大学先进技术研究院 | Cross-scene VOCs gas leakage detection method, system and storage medium |
| CN113780135A (en) * | 2021-08-31 | 2021-12-10 | 中国科学技术大学先进技术研究院 | Cross-scene VOCs gas leakage detection method and system and storage medium |
| CN113963513A (en) * | 2021-10-13 | 2022-01-21 | 公安部第三研究所 | Robot system for realizing intelligent inspection in chemical industry and control method thereof |
| CN114459708A (en) * | 2022-02-10 | 2022-05-10 | 合肥永信科翔智能技术有限公司 | Gas leakage monitoring system based on intelligent gas sensing |
| KR102646239B1 (en) * | 2022-06-03 | 2024-03-12 | 울산과학기술원 | Method providing composite video for detection of gas leak and artificial intelligence based gas leak detection device performing the same |
| KR20230168243A (en) * | 2022-06-03 | 2023-12-13 | 울산과학기술원 | Method providing composite video for detection of gas leak and artificial intelligence based gas leak detection device performing the same |
| CN115018737A (en) * | 2022-08-04 | 2022-09-06 | 四川迪晟新达类脑智能技术有限公司 | Infrared thermal image enhancement method and device |
| CN116091491A (en) * | 2023-04-03 | 2023-05-09 | 中国科学技术大学先进技术研究院 | VOCs gas detection method, device, equipment and computer readable storage medium |
| CN116543241B (en) * | 2023-07-07 | 2023-09-15 | 杭州海康威视数字技术股份有限公司 | Detection method and device for leakage gas cloud, storage medium and electronic equipment |
| CN116543241A (en) * | 2023-07-07 | 2023-08-04 | 杭州海康威视数字技术股份有限公司 | Detection method and device for leakage gas cloud, storage medium and electronic equipment |
| CN116823839B (en) * | 2023-08-31 | 2023-12-01 | 梁山中维热力有限公司 | Pipeline leakage detection method based on thermal infrared image |
| CN116823839A (en) * | 2023-08-31 | 2023-09-29 | 梁山中维热力有限公司 | Pipeline leakage detection method based on thermal infrared image |
| CN117036347A (en) * | 2023-10-08 | 2023-11-10 | 山东柯林瑞尔管道工程有限公司 | Visual detection method and system for pipeline lining leakage based on image enhancement |
| CN117036347B (en) * | 2023-10-08 | 2024-02-02 | 山东柯林瑞尔管道工程有限公司 | Visual detection method and system for pipeline lining leakage based on image enhancement |
| CN117094995A (en) * | 2023-10-18 | 2023-11-21 | 中国恩菲工程技术有限公司 | Reaction kettle gas leakage detection method, device, medium and equipment |
| CN117094995B (en) * | 2023-10-18 | 2024-02-06 | 中国恩菲工程技术有限公司 | Reaction kettle gas leakage detection method, device, medium and equipment |
| CN117132948A (en) * | 2023-10-27 | 2023-11-28 | 南昌理工学院 | Scenic spot tourist flow monitoring method, system, readable storage medium and computer |
| CN117132948B (en) * | 2023-10-27 | 2024-01-30 | 南昌理工学院 | Scenic spot tourist flow monitoring method, system, readable storage medium and computer |
| CN117788466A (en) * | 2024-02-26 | 2024-03-29 | 国科大杭州高等研究院 | Uncooled infrared video sequence dangerous gas imaging leakage detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111325721A (en) | Gas leakage detection method and system based on infrared thermal imaging | |
| US11221107B2 (en) | Method for leakage detection of underground pipeline corridor based on dynamic infrared thermal image processing | |
| CN108827969B (en) | Method and device for detecting and identifying surface defects of metal parts | |
| CN111161222B (en) | Printing roller defect detection method based on visual saliency | |
| CN111611907B (en) | Image-enhanced infrared target detection method | |
| CN112308854B (en) | Automatic detection method and system for chip surface flaws and electronic equipment | |
| CN108318491A (en) | A kind of fabric defect detection method based on frequency spectrum curvature analysis | |
| CN110648330B (en) | Defect detection method for camera glass | |
| CN107341793A (en) | A kind of target surface image processing method and device | |
| CN112862744B (en) | Intelligent detection method for internal defects of capacitor based on ultrasonic image | |
| CN115841434A (en) | Infrared image enhancement method for gas concentration analysis | |
| Ma et al. | An automatic detection method of Mura defects for liquid crystal display | |
| CN112419261A (en) | Visual acquisition method and device with abnormal point removing function | |
| CN112927223A (en) | Glass curtain wall detection method based on infrared thermal imager | |
| CN116664565A (en) | Hidden crack detection method and system for photovoltaic solar cell | |
| CN113705564B (en) | Pointer type instrument identification reading method | |
| CN105608674B (en) | A kind of image enchancing method based on image registration, interpolation and denoising | |
| CN111353968A (en) | Infrared image quality evaluation method based on blind pixel detection and analysis | |
| CN113673396A (en) | Spore germination rate calculation method and device and storage medium | |
| CN115797314A (en) | Part surface defect detection method, system, equipment and storage medium | |
| Zhang et al. | A LCD screen Mura defect detection method based on machine vision | |
| Luo et al. | Adaptive canny and semantic segmentation networks based on feature fusion for road crack detection | |
| CN115330705A (en) | Skin paint surface defect detection method based on adaptive weighting template NCC | |
| Yi et al. | Tire body defect detection: From the perspective of industrial applications | |
| CN117649412B (en) | Aluminum material surface quality detection method |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200623 |
|
| RJ01 | Rejection of invention patent application after publication |