CN116758085B - Visual auxiliary detection method for infrared image of gas pollution - Google Patents
Visual auxiliary detection method for infrared image of gas pollution Download PDFInfo
- Publication number
- CN116758085B CN116758085B CN202311048437.8A CN202311048437A CN116758085B CN 116758085 B CN116758085 B CN 116758085B CN 202311048437 A CN202311048437 A CN 202311048437A CN 116758085 B CN116758085 B CN 116758085B
- Authority
- CN
- China
- Prior art keywords
- connected domain
- domains
- image
- domain
- pipeline
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 230000000007 visual effect Effects 0.000 title claims abstract description 14
- 238000013507 mapping Methods 0.000 claims abstract description 50
- 239000013598 vector Substances 0.000 claims abstract description 5
- 238000000638 solvent extraction Methods 0.000 claims abstract description 3
- 230000008859 change Effects 0.000 claims description 29
- 238000000034 method Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000012886 linear function Methods 0.000 claims description 6
- 238000003708 edge detection Methods 0.000 claims description 5
- 230000011218 segmentation Effects 0.000 claims description 5
- 238000004422 calculation algorithm Methods 0.000 claims description 4
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 238000004519 manufacturing process Methods 0.000 claims description 3
- 238000004088 simulation Methods 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 17
- 238000012545 processing Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 13
- 238000013528 artificial neural network Methods 0.000 description 3
- 238000009792 diffusion process Methods 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 230000002708 enhancing effect Effects 0.000 description 2
- 238000003331 infrared imaging Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000007796 conventional method Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N25/00—Investigating or analyzing materials by the use of thermal means
-
- 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
-
- 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
-
- 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
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- 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
- G06V20/70—Labelling scene content, e.g. deriving syntactic or semantic representations
-
- 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
- G06T2207/10024—Color 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
- G06T2207/10—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
-
- 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
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Analytical Chemistry (AREA)
- Pathology (AREA)
- Immunology (AREA)
- Biochemistry (AREA)
- Chemical & Material Sciences (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Biophysics (AREA)
- Quality & Reliability (AREA)
- Databases & Information Systems (AREA)
- Medical Informatics (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the technical field of image data processing, in particular to a gas pollution infrared image vision auxiliary detection method, which comprises the following steps: acquiring RGB images and infrared images of the product, and obtaining corresponding pipeline areas on the infrared images by utilizing the RGB images; vector classification is carried out by utilizing edge pixel points to obtain a plurality of connected domains after self-adaptive partitioning, and the connected domains are sequenced to obtain a connected domain sequence set; acquiring the temperature variation of each connected domain and fitting the temperature distance trend rate of each connected domain to obtain the contrast mapping parameter of each connected domain; and carrying out image enhancement on different connected domains according to the contrast mapping parameters to obtain enhanced connected domains, and completing auxiliary detection. According to the invention, through self-adaptive interval division, different degrees of image enhancement are carried out on different intervals, the influence of an image background in an image enhancement effect is reduced, and the visual auxiliary detection of the gas pollution infrared image is completed.
Description
Technical Field
The invention relates to the technical field of image data processing, in particular to a visual auxiliary detection method for an infrared image of gas pollution.
Background
In the gas detection field, infrared imaging detection has advantages of convenience, accuracy and high efficiency, but when the gas quantity is small or the background has stronger interference, the visual judgment of gas pollution can be caused, and the gas pollution judgment is difficult to effectively and accurately carry out. Furthermore, the detail perception of the infrared image for improving the gas pollution caused by the pipeline leakage is improved, the conventional method for enhancing the infrared image is commonly provided with a contrast-limited histogram equalization method, but the method depends on the data information of the whole image when the image is enhanced, and the enhancement effect on the gas part is not obvious when the image is enhanced.
Disclosure of Invention
The invention provides an infrared image vision auxiliary detection method for gas pollution, which aims to solve the existing problems.
The invention relates to a visual auxiliary detection method for an infrared image of gas pollution, which adopts the following technical scheme:
an embodiment of the invention provides a visual auxiliary detection method for an infrared image of gas pollution, which comprises the following steps:
acquiring an RGB image and an infrared image of a product through cameras arranged on a production line, performing semantic segmentation according to the RGB image to obtain a pipeline semantic edge on the RGB image, and corresponding the pipeline semantic edge on the RGB image to the infrared image to obtain the pipeline semantic edge of the infrared image;
obtaining a pipeline edge image by utilizing the pipeline semantic edge of the infrared image, and carrying out region division on different pixel points on the pipeline edge image to obtain a plurality of pipeline regions; obtaining a connected domain sequence set according to the pipeline region, obtaining a target connected domain of each connected domain in the connected domain sequence set, sequencing the target connected domain according to the distance between the target connected domain and the pipeline region to obtain a connected domain sequence of each connected domain, obtaining the temperature variation in the connected domain sequence of each connected domain, obtaining the temperature distance trend rate of each connected domain according to the temperature variation in the connected domain sequence of each connected domain, and obtaining the contrast mapping parameter of each connected domain according to the temperature variation and the temperature distance trend rate of each connected domain;
and carrying out image enhancement on different connected domains according to the contrast mapping parameters of each connected domain to obtain enhanced connected domains, recalculating the temperature distance trend rate of the enhanced connected domains, judging whether the temperature distance trend rate of the enhanced connected domains exceeds a preset enhancement threshold, if so, causing gas pollution, otherwise, causing no gas pollution.
Preferably, the step of obtaining the sequential set of connected domains according to the pipeline area includes the following specific steps:
and performing edge detection on the pipeline region by using an edge detection algorithm to obtain edge pixel points on the pipeline edge image, performing vector classification by using the edge pixel points to obtain a plurality of connected domains after self-adaptive partitioning, and sequencing the connected domains to obtain a connected domain sequence set.
Preferably, the sorting the connected domains to obtain a connected domain sequence set includes the following specific steps:
on the pipeline edge image, all pipeline semantic edge pixel points of the infrared image are obtained, the mass centers of all the connected domains are calculated, the distance between the mass center of each connected domain and the nearest edge pixel point is calculated and is recorded as the connected domain distance of each connected domain, and all the connected domains are arranged according to the sequence from large to small in the connected domain distance to obtain a connected domain sequence set.
Preferably, the obtaining the target connected domain of each connected domain in the connected domain sequence set includes the following specific steps:
and in the connected domain sequence set, acquiring the temperature variation of each connected domain and the mass center of each connected domain, calculating the center of the edge pixel point, and obtaining all connected domains passing by the straight line according to the fact that the mass center of each connected domain and the center of the edge pixel point are connected into the straight line, and marking the all connected domains as target connected domains of each connected domain.
Preferably, the specific method for obtaining the temperature variation of the connected domain is as follows:
and in each connected domain, carrying out difference on the image values of the infrared image and the infrared image of the previous frame, taking the absolute value of the difference value, obtaining the temperature variation corresponding to each pixel point, accumulating the temperature variation corresponding to all the pixel points in each connected domain, and recording the accumulated value as the temperature variation of each connected domain.
Preferably, the contrast mapping parameter of each connected domain is obtained according to the temperature variation and the temperature distance tendency rate of each connected domain, and the specific calculation formula is as follows:
wherein ,is->Contrast mapping parameters of the connected domains, +.>Is the +.o in the connected domain sequence set>Contrast mapping parameters of the immediately preceding connected domain, of the connected domains>Is->Temperature change corresponding to each connected domain, +.>Is->Temperature distance trend of individual connected domains, +.>Is a preset parameter adjusting coefficient, and is->Representation->A function.
Preferably, the method for obtaining the temperature distance trend rate of each connected domain according to the temperature variation in the connected domain sequence of each connected domain comprises the following specific steps:
for any connected domainCommunicating domain->The temperature change amounts of all the target connected domains in the pipeline are sequenced from large to small according to the distance values of the target connected domains and the pipeline points to obtain a connected domain sequence, wherein the connected domain sequence comprises connected domains>Obtaining the +.>Obtaining the temperature change values of the plurality of target connected domains according to the temperature change values of the plurality of target connected domains>Temperature distance trend of (2).
Preferably, the connected domain is obtained according to the temperature variation values of the plurality of target connected domainsComprises the following specific steps:
by means of communicating domainsThe temperature variation value of the target communicating region of (2) constitutes communicating region +.>Is to use the linear fitting method to connect the connected domain +.>Is fitted into a linear function, and the slope value of the linear function is obtained and is recorded as a connected domain +.>Temperature distance trend of (2).
Preferably, the image enhancement of different connected domains according to the contrast mapping parameter of each connected domain to obtain enhanced connected domains includes the following specific steps:
obtaining a fitting function according to the contrast mapping parameters of each connected domain; and obtaining the contrast values of different connected domains according to the fitting function, obtaining the connected domains with the contrast values larger than a preset contrast threshold, and carrying out image enhancement on the connected domains with the contrast values larger than the preset contrast threshold to obtain the enhanced connected domains.
Preferably, the fitting function is obtained according to the contrast mapping parameter of each connected domain, which comprises the following specific steps:
calculating contrast mapping parameters of all connected domains on the pipeline edge image according to the contrast mapping parameters of each connected domain, and constructing a contrast mapping parameter and a contrast mapping table by utilizing the contrast mapping parameters of all connected domains; and carrying out nonlinear fitting on the contrast mapping parameters and the contrast to obtain a fitting function between the contrast mapping parameters and the contrast.
The technical scheme of the invention has the beneficial effects that: through carrying out the division of different positions to infrared image area, according to the process that gas leakage caused the pollution, carried out the infrared image enhancement of self-adaptation for the enhancement effect has the pertinence to gas, and can carry out the enhancement adjustment according to the diffusion of gas, obtain stable accurate image enhancement effect, and then be used for the supplementary monitoring of gas pollution.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart showing the steps of a visual auxiliary detection method for gas pollution infrared images.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of specific implementation, structure, characteristics and effects of the method for visual auxiliary detection of gas pollution infrared images according to the invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the gas pollution infrared image vision auxiliary detection method provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for visually assisting in detecting a gas pollution infrared image according to an embodiment of the invention is shown, the method includes the following steps:
s001: the RGB image and the infrared image of the product are acquired through cameras arranged on the production line, semantic segmentation is carried out according to the RGB image to obtain the pipeline semantic edge on the RGB image, and the pipeline semantic edge on the RGB image is corresponding to the pipeline semantic edge on the infrared image to obtain the pipeline semantic edge of the infrared image.
Specifically, an infrared camera and a color industrial camera are arranged in a region to be polluted, and infrared images and color industrial cameras of the region to be monitored are collectedAn image. The infrared camera and the color industrial camera are connected with the data processing center in a wired mode for data transmission, and after the infrared camera and the color industrial camera acquire image data, the image data are sent to the data processing center. The data processing center obtains the infrared image and +.>After the image, will be composed of->The pipeline image acquired by the camera is input into a neural network after training, and the semantic segmentation network is utilized to perform +.>Identifying the image to obtainThe ducts in the region to be examined in the image, wherein the present embodiment specifically employs +.>The neural network performs semantic segmentation, the training set is acquired and labeled by related experience personnel, and the labeling process is pixel-level labelingMarking the pixel points belonging to the pipeline as 1, marking the background pixel points as 0, and adopting +.>Coding is carried out by a coding mode, whereinTraining procedure for neural networks and->The encoding method is well known, and this embodiment is not described here again. Obtain->The part of the pipeline in the region to be detected in the image, wherein the RGB image and the infrared imaging are calibrated in advance so that the pixels in the same infrared image are identical>The pixel points in the image represent the same position of the same object at the same view angle in the real world, so as to obtain the semantic edge of the pipeline in the infrared image.
Thus, the semantic edges of the pipeline in the infrared image are obtained.
S002: and obtaining a pipeline edge image by utilizing the pipeline semantic edge of the infrared image, and carrying out region division on different pixel points on the pipeline edge image to obtain a plurality of pipeline regions.
It should be noted that, under different positions apart from the pipeline, the gas amount is different, and the background is different, wherein when the gas amount is less obvious, the effect on gas pollution detection is often better, and when a large amount of gas is provided, the gas pollution detection can be completed without obvious image enhancement. When the gas pollution is detected, since the gas pollution starts to spread out from the pipe, when the infrared image enhancement is performed, the area closer to the pipe is preferentially detected, and the image enhancement is performed.
It should be further noted that after the pipeline area in the infrared image is obtained, since the gas pollution is diffused outwards from the pipeline, when the infrared image is enhanced, the area closer to the pipeline is preferentially ensured, the corresponding detail enhancement should be more obvious, and the problem that the gas data shot in the infrared image after the infrared image enhancement is still not obvious due to the fact that the gas proportion is smaller than the specific gravity of the whole image, when the infrared image is enhanced, the infrared image enhancement effect depends on global infrared information is solved, so that the pixel points are required to be adaptively divided according to the distance between the pixels in the image and the pipeline, a plurality of categories are obtained, and the image data are partitioned according to the categories to obtain a plurality of category areas. Meanwhile, as the camera is fixed and the pipeline position is unchanged, the position of the category area after the self-adaptive division is also fixed.
Specifically, all infrared pixel points in the semantic edge of the pipeline in the infrared image are extracted to a blank image to obtain a new image, and the new image is recorded as the pipeline edge image, and the pipeline semantic edge image is utilizedThe edge detection algorithm is used for carrying out edge detection to obtain pixel points belonging to edges in the pipeline edge image; further calculate +.>The Euclidean distance value from each pixel point to the corresponding pixel point of the semantic edge of the pipeline, wherein the number of the edge pixel points is more than one, and the Europe distance value is selected from the +.>The minimum value of the Euclidean distance values from the pixel point to all the edge pixel points is taken as the +.>Distance value of each pixel point to pipeline +.>The Euclidean distance calculation is performed by adopting the edge instead of the center line of the pipeline, because the surface texture of the pipeline is enhanced by the method, and the gas is influenced.
Further, in order to reduce background enhancement of imageThe present embodiment selects the coordinate information of each pixel point and the distance value corresponding to the pipeline, which interfere with the image gas enhancement effectForms a three-dimensional vector by +.>The algorithm classifies the three-dimensional vectors corresponding to all the pixels, so that the positions can be approximated and +.>The approximate values are divided into one type, a plurality of pipeline areas on images corresponding to different types are obtained, and area division among different pixel points is completed.
To this end, several pipe areas are obtained.
S003: obtaining the temperature variation of each connected domain in the connected domain sequence according to the pipeline region, obtaining the temperature distance trend rate of each connected domain according to the temperature variation of each connected domain in the connected domain sequence, obtaining the contrast mapping parameter of each connected domain according to the temperature variation of each connected domain and the temperature distance trend rate, and further obtaining the fitting function between the contrast mapping parameter and the contrast.
It should be noted that, after the adaptive region division is completed, in order to enable the infrared image enhancement effect of different regions to have different degrees, in this embodiment, a limited contrast histogram equalization method is selected to perform the image enhancement of different regions to different degrees. Wherein the larger the value of the contrast, the more pronounced the enhancement effect and should have a higher enhancement effect for the gas surrounding the pipe. However, it is necessary to calculate the temperature variation (i.e., frame difference image) between consecutive frames at the same position in the infrared image, but if the current gas pollution level is large or the variation is small when there is no pollution, the corresponding enhancement effect should be reduced, and if the current pollution level is slight, the temperature variation is small. However, since the gas is an out-diffusion process, if the temperature change amount is from near to far along the pipeline, the temperature is as high as possibleThe change of the degree increases gradually, and the farther the distance is, the more serious the current gas pollution degree is; if the temperature variation is reduced along with the pipeline from near to far, the current gas pollution degree is indicated to have a certain amount of leakage, but the gas is in a state of being difficult to observe and needs stronger infrared image enhancement effect. Furthermore, the embodiment combines the distance information between different areas and the pipelines and the corresponding temperature variation of different areas to obtain the contrast ratio coefficients required by different areas, and the contrast ratio coefficients are used for enhancing the image under different limiting contrast ratios so as to enhance the whole infrared image. By calculation to obtain the firstDistance weight of individual region->According to the change of the slope between the adjacent areas, the current +.>The individual regions are in regions with a large gas quantity, a gas quantity of 0 or a small gas quantity, and further enhancement is performed to different degrees.
Specifically, the image values of the current infrared image and the infrared image of the previous frame are subjected to difference, the absolute value of the difference is taken, the temperature variation corresponding to each pixel point is obtained, and then the current first pixel point is obtainedThe temperature change amounts corresponding to all pixel points in each area are accumulated and recorded as the current +.>Temperature variation corresponding to the individual regions->It->The larger the value is, the more easily the change amount is observed, the less the change amount is, the more difficult the observation is, and the limit contrast value corresponding to the region with the smaller all the change amounts is, the larger the limit contrast value corresponding to the region with the smaller change amount is. So use->The function performs a negative correlation mapping.
Further, on the pipeline edge image, all pipeline semantic edge pixel points of the infrared image are obtained, and the arithmetic mean value of the horizontal and vertical coordinates of all pipeline semantic edge pixels is calculated and recorded as the pipeline point; obtaining the first according to the definition of the connected domainThe communicating domains corresponding to the respective regions are a series of parallel communicating domains parallel to the pipeline region and have a shape similar to the pipeline region, and the +.>Temperature variation corresponding to the individual regions->As->Calculating the mass center of all the connected domains and calculating the +.>The distance between the centroid of each connected domain and the nearest edge pixel point is marked as +.>The communicating domain distances of the communicating domains are arranged according to the order of the communicating domain distances from large to small to obtain a communicating domain order set; computing the +.>Centroid of individual connected domain, will->Connecting the mass center of each connected domain with the nearest pipeline semantic edge pixel point to obtain a straight line, obtaining the connected domain passed by the straight line, marking all the connected domains passed by the straight line as target connected domains, obtaining the temperature variation of each connected domain passed by the straight line, sorting the temperature variation of the connected domains according to the distance value from the pipeline point from large to small to obtain a connected domain sequence, and obtaining the first line in the connected domain sequence>After the individual communicating domains->The temperature variation value of the connected domain constitutes +.>The temperature variation sequence of the individual communicating regions, wherein the present embodiment is described as +.>For the purposes of illustration, if->The number of the subsequent connected domains is less than 6, and the +.>The temperature variation value of all the connected domains after the connected domains constitutes +.>A temperature variation sequence of each connected domain; by means of a linear fit, the +.>The temperature change sequences of the connected domains are fitted into a linear function, and the slope value of the linear function is obtained and recorded as the +.>Temperature distance tendency of individual connected domains +.>And calculate->The specific calculation formula of the recurrence of the contrast mapping parameters of the connected domains is as follows:
wherein ,is->Contrast mapping parameters of the connected domains, +.>Is the +.o in the connected domain sequence set>Contrast mapping parameters of the immediately preceding connected domain, of the connected domains>Is->Temperature change corresponding to each connected domain, +.>Is->Temperature distance trend of individual connected domains, +.>Is a parameter tuning coefficient, wherein the present embodiment uses +.>For the purposes of illustration, add>Indicating +.>Function calculation, representing pair->Is inversely related mapped. If->The larger the value of (2) is greater than 0 and the larger the value is, the more indicates the current +.>The gas pollution amount corresponding to each region is a region with larger gas amount, but the region with larger gas amount is easy to observe, and excessive enhancement can be caused after enhancement, so that the enhancement effect of a smaller degree is required to be even not enhanced; if->The value of (2) is approximately 0, indicating the current +.>The gas pollution amount corresponding to each region is a region with the gas amount of 0, and the gas is a diffusion model, so that the gas pollution is not existed in the case; if->The smaller the value of (2) is smaller than 0 and the smaller the value is, the more indicates the current +.>The amount of gas pollution corresponding to each region is a region where the amount of gas is small, and the subsequent regions are gradually reduced, so that a large degree of enhancement effect is required here. It should be noted that when +.>When, the initial value of contrast map parameter set in this embodiment +.>A value of 0 indicates that the gas in this region does not need to be enhanced. />The larger the expression->The larger the contrast limit value is required for the individual regions in image enhancement, the more pronounced the gas becomes for image enhancement of the local regions.
Further, after the contrast mapping parameters corresponding to each connected domain are obtained, the adaptive image enhancement of different connected domains is realized by establishing the contrast mapping parameters and the contrast relation mapping table, so that the method is used for monitoring gas pollution. The method comprises the following specific steps:
the image enhancement under different contrast limits is carried out on different connected domains by relevant experience staff, the relevant experience staff selects the optimal contrast limit value corresponding to the different connected domains, the optimal contrast value corresponding to the different connected domains and the contrast mapping parameter are recorded, a contrast mapping parameter and a contrast mapping table are obtained, and further nonlinear fitting is carried out on the contrast mapping parameter and the contrast by using a least square method, so that a fitting function between the contrast mapping parameter and the contrast is obtained.
Thus, a fitting function between the contrast mapping parameters and the contrast is obtained.
S004: and carrying out image enhancement to different degrees according to the fitting function between the contrast mapping parameter and the contrast of each connected domain to obtain an enhanced gas infrared image, so as to realize the visual auxiliary detection of the gas pollution infrared image.
The embodiment obtains a contrast threshold according to practical experienceAnd an enhancement threshold->Wherein the present embodiment is +.> and />For the sake of example, the present embodiment is not limited in particular, and the two thresholds may be determined according to the specific implementation.
Specifically, in the subsequent gas pollution monitoring, the contrast mapping parameter of each connected domain is obtained by using a fitting function, and the contrast values corresponding to different connected domains are obtained by using the fitting function between the contrast mapping parameter and the contrast; at contrast values greater thanAnd (3) carrying out image enhancement on the connected domain by adopting histogram equalization to obtain infrared images with different enhancement effects on different connected domains.
Further, after obtaining the infrared images with different enhancement effects on different connected domains, recalculating the firstCorresponding->Absolute value of value, when->When the gas leakage is generated at present, the gas pollution is caused, otherwise, no leakage is generated at present, and infrared image enhancement with different degrees is carried out by the personnel with related experience, so that the auxiliary detection of the gas pollution is finished.
Thus, the auxiliary detection of gas pollution is completed.
The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.
Claims (4)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311048437.8A CN116758085B (en) | 2023-08-21 | 2023-08-21 | Visual auxiliary detection method for infrared image of gas pollution |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311048437.8A CN116758085B (en) | 2023-08-21 | 2023-08-21 | Visual auxiliary detection method for infrared image of gas pollution |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116758085A CN116758085A (en) | 2023-09-15 |
| CN116758085B true CN116758085B (en) | 2023-11-03 |
Family
ID=87957648
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311048437.8A Active CN116758085B (en) | 2023-08-21 | 2023-08-21 | Visual auxiliary detection method for infrared image of gas pollution |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116758085B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118314126B (en) * | 2024-06-05 | 2024-09-03 | 西安祥堃电气有限公司 | Mining oil on-line pollution detection system based on image processing |
| CN119417855B (en) * | 2025-01-08 | 2025-03-25 | 大连高聚新材料科技有限公司 | A method for screening chemical materials |
Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004044683A2 (en) * | 2002-11-06 | 2004-05-27 | Simmonds Precision Products, Inc. | Method for detection and recognition of fog presence within an aircraft compartment using video images |
| CN104484871A (en) * | 2014-11-27 | 2015-04-01 | 小米科技有限责任公司 | Edge extraction method and device |
| GB201711412D0 (en) * | 2016-12-30 | 2017-08-30 | Maxu Tech Inc | Early entry |
| CN109060144A (en) * | 2018-08-24 | 2018-12-21 | 电子科技大学 | The method that thermal infrared imager NETD is tested automatically |
| CN110047078A (en) * | 2019-04-18 | 2019-07-23 | 北京市商汤科技开发有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN110261436A (en) * | 2019-06-13 | 2019-09-20 | 暨南大学 | Rail deformation detection method and system based on infrared thermal imaging and computer vision |
| WO2020065241A1 (en) * | 2018-09-28 | 2020-04-02 | Olythe | Portable device for measuring the concentration of at least one component in a gas exhaled by a breath fluid |
| CN111914675A (en) * | 2020-07-09 | 2020-11-10 | 武汉万集信息技术有限公司 | Vehicle axle number determining method and device |
| CN113902641A (en) * | 2021-10-12 | 2022-01-07 | 西安交通大学 | A method and system for discriminating hot spots in data centers based on infrared images |
| WO2022110283A1 (en) * | 2020-11-24 | 2022-06-02 | 烟台艾睿光电科技有限公司 | Infrared temperature measurement method, apparatus, device, and storage medium |
| CN115965773A (en) * | 2023-03-16 | 2023-04-14 | 深圳市特安电子有限公司 | Gas leakage detection system based on big data of Internet of things |
| CN116311079A (en) * | 2023-05-12 | 2023-06-23 | 探长信息技术(苏州)有限公司 | Civil security engineering monitoring method based on computer vision |
| CN116433668A (en) * | 2023-06-14 | 2023-07-14 | 东营孚瑞特能源设备有限公司 | Intelligent hydraulic oil pipe oil leakage detection method |
| CN116486061A (en) * | 2023-06-20 | 2023-07-25 | 苏州德斯米尔智能科技有限公司 | Sewage treatment effect detection method based on machine vision |
-
2023
- 2023-08-21 CN CN202311048437.8A patent/CN116758085B/en active Active
Patent Citations (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2004044683A2 (en) * | 2002-11-06 | 2004-05-27 | Simmonds Precision Products, Inc. | Method for detection and recognition of fog presence within an aircraft compartment using video images |
| CN104484871A (en) * | 2014-11-27 | 2015-04-01 | 小米科技有限责任公司 | Edge extraction method and device |
| GB201711412D0 (en) * | 2016-12-30 | 2017-08-30 | Maxu Tech Inc | Early entry |
| CN109060144A (en) * | 2018-08-24 | 2018-12-21 | 电子科技大学 | The method that thermal infrared imager NETD is tested automatically |
| WO2020065241A1 (en) * | 2018-09-28 | 2020-04-02 | Olythe | Portable device for measuring the concentration of at least one component in a gas exhaled by a breath fluid |
| CN110047078A (en) * | 2019-04-18 | 2019-07-23 | 北京市商汤科技开发有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN110261436A (en) * | 2019-06-13 | 2019-09-20 | 暨南大学 | Rail deformation detection method and system based on infrared thermal imaging and computer vision |
| CN111914675A (en) * | 2020-07-09 | 2020-11-10 | 武汉万集信息技术有限公司 | Vehicle axle number determining method and device |
| WO2022110283A1 (en) * | 2020-11-24 | 2022-06-02 | 烟台艾睿光电科技有限公司 | Infrared temperature measurement method, apparatus, device, and storage medium |
| CN113902641A (en) * | 2021-10-12 | 2022-01-07 | 西安交通大学 | A method and system for discriminating hot spots in data centers based on infrared images |
| CN115965773A (en) * | 2023-03-16 | 2023-04-14 | 深圳市特安电子有限公司 | Gas leakage detection system based on big data of Internet of things |
| CN116311079A (en) * | 2023-05-12 | 2023-06-23 | 探长信息技术(苏州)有限公司 | Civil security engineering monitoring method based on computer vision |
| CN116433668A (en) * | 2023-06-14 | 2023-07-14 | 东营孚瑞特能源设备有限公司 | Intelligent hydraulic oil pipe oil leakage detection method |
| CN116486061A (en) * | 2023-06-20 | 2023-07-25 | 苏州德斯米尔智能科技有限公司 | Sewage treatment effect detection method based on machine vision |
Non-Patent Citations (5)
| Title |
|---|
| Rotation and Scale Invariant for Texture Analysis Based on Radon Transform and Wavelet Transform;Guangbin Yu 等;《 2008 Third International Conference on Pervasive Computing and Applications》;全文 * |
| 基于改进高低帽变换的钢轨裂纹红外图像增强;顾桂梅;刘丽;;激光与红外(01);全文 * |
| 基于红外图像处理的冰箱管路温度自动测量的研究;王媛;曾建风;;世界科技研究与发展(02);全文 * |
| 基于视觉对比度分辨率的红外图像增强算法;代少升;刘永朋;;半导体光电(06);全文 * |
| 颜色不均匀情况下光伏板红外图像分割算法研究;王楠;《中国优秀硕士学位论文全文数据库 (信息科学辑)》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116758085A (en) | 2023-09-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116758085B (en) | Visual auxiliary detection method for infrared image of gas pollution | |
| CN107578418B (en) | Indoor scene contour detection method fusing color and depth information | |
| CN112200045B (en) | Remote sensing image target detection model establishment method based on context enhancement and application | |
| CN110490914A (en) | It is a kind of based on brightness adaptively and conspicuousness detect image interfusion method | |
| CN103164692B (en) | A kind of special vehicle instrument automatic identification system based on computer vision and method | |
| CN105243667B (en) | The recognition methods again of target based on Local Feature Fusion | |
| CN109064418A (en) | A kind of Images Corrupted by Non-uniform Noise denoising method based on non-local mean | |
| CN104036493B (en) | No-reference image quality evaluation method based on multifractal spectrum | |
| CN112085675B (en) | Deep image denoising method, foreground segmentation method and human motion monitoring method | |
| CN109903331A (en) | A Convolutional Neural Network Object Detection Method Based on RGB-D Camera | |
| CN107743225B (en) | A Method for No-Reference Image Quality Prediction Using Multi-Layer Depth Representations | |
| CN114092478B (en) | Anomaly detection method | |
| CN106651829B (en) | A kind of non-reference picture method for evaluating objective quality based on energy and texture analysis | |
| CN106408526B (en) | A Visibility Detection Method Based on Multilayer Vector Graphics | |
| CN108053412B (en) | Determination method of optimal segmentation result of remote sensing image based on fusion of multiple quantitative evaluation indicators | |
| CN113945570A (en) | Detection and identification method for micro-leakage steam defects based on steam characteristic infrared characterization | |
| CN116524000B (en) | Rose size measurement method and system based on deep learning and stereoscopic vision | |
| CN111738984B (en) | Skin image spot evaluation method and system based on watershed and seed filling | |
| CN109978858B (en) | A dual-frame thumbnail image quality assessment method based on foreground detection | |
| CN103914840B (en) | A kind of human body contour outline extraction method for non-simple background | |
| CN114581475B (en) | Laser stripe segmentation method based on multi-scale salient features | |
| CN104392209B (en) | A kind of image complexity evaluation method of target and background | |
| CN116993733B (en) | Earphone sleeve appearance quality detection method and system | |
| CN118982754A (en) | A method for segmentation and identification of cracks in subway tunnels | |
| CN115222732B (en) | Injection molding process anomaly detection method based on big data analysis and color difference detection |
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 | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: Room A501, Science and Technology Entrepreneurship Park, Luoxi Street, High tech Industrial Development Zone, Linyi City, Shandong Province, 276000 Patentee after: Linyi Kunzhong Information Technology Service Co.,Ltd. Country or region after: China Address before: 276000 Room A510, Innovation Building, Linyi High-tech Zone, Shandong Province Patentee before: SHANDONG KUNZHONG INFORMATION TECHNOLOGY CO.,LTD. Country or region before: China |