CN107705288B - Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion - Google Patents

Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion Download PDF

Info

Publication number
CN107705288B
CN107705288B CN201710786152.2A CN201710786152A CN107705288B CN 107705288 B CN107705288 B CN 107705288B CN 201710786152 A CN201710786152 A CN 201710786152A CN 107705288 B CN107705288 B CN 107705288B
Authority
CN
China
Prior art keywords
area
image
pixel
pixel point
points
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
Application number
CN201710786152.2A
Other languages
Chinese (zh)
Other versions
CN107705288A (en
Inventor
洪汉玉
王敏
黄丽坤
洪梓铭
章秀华
王鹏程
刘忠
王万里
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan Institute of Technology
Original Assignee
Wuhan Institute of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Wuhan Institute of Technology filed Critical Wuhan Institute of Technology
Priority to CN201710786152.2A priority Critical patent/CN107705288B/en
Publication of CN107705288A publication Critical patent/CN107705288A/en
Application granted granted Critical
Publication of CN107705288B publication Critical patent/CN107705288B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G01—MEASURING; TESTING
    • G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
    • G01M3/00—Investigating fluid-tightness of structures
    • G01M3/02—Investigating fluid-tightness of structures by using fluid or vacuum
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G06F18/232—Non-hierarchical techniques
    • G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
    • G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/46—Descriptors for shape, contour or point-related descriptors, e.g. scale invariant feature transform [SIFT] or bags of words [BoW]; Salient regional features
    • G06V10/462—Salient features, e.g. scale invariant feature transforms [SIFT]
    • 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
    • G06T2207/10016—Video; Image sequence
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Probability & Statistics with Applications (AREA)
  • Multimedia (AREA)
  • Quality & Reliability (AREA)
  • Closed-Circuit Television Systems (AREA)

Abstract

The invention discloses a dangerous gas leakage infrared video detection method under strong interference of pseudo-target motion, which comprises the steps of firstly carrying out background modeling on gas infrared video images and carrying out differential processing on each frame of infrared image by utilizing a background obtained by modeling; then traversing the differential binary image by using a 4-neighborhood connected domain algorithm and primarily screening the found connected regions to obtain a plurality of alternative moving target regions; on the basis of improving a FAST characteristic point detection algorithm, FAST characteristic extraction is carried out on a plurality of alternative moving target areas in an infrared video image, and the characteristics are used for effectively distinguishing gas from interfering moving targets such as camera shake, pedestrian walking, tree shaking, cloud cluster movement and the like by utilizing different distribution rules in different kinds of moving target areas, so that the algorithm has good robustness under strong interference of the moving targets; and finally, the algorithm carries out pseudo-color processing on the detected gas area, so that a monitoring end can find and position the gas area in time conveniently.

Description

Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion
Technical Field
The invention belongs to the field of image processing and gas detection, relates to a dangerous gas leakage detection method based on infrared video image processing, and particularly relates to a dangerous gas leakage infrared video detection method under strong pseudo-target motion interference.
Background
Industrial production enterprises often inevitably use, produce or transport toxic, flammable and explosive gases during the production process. These gases are colorless and odorless and are indiscernible to the human eye, but are extremely dangerous and, in the event of a leak, are extremely explosive and cause a fire. In order to detect the leakage of dangerous gas in the production environment in time, industrial production enterprises often organize detection personnel to be equipped with professional instruments to patrol equipment regularly. The traditional mode of this kind of manual detection requires that the measurement personnel must carry out contact measurement, and it is many, the risk is high, the inefficiency not only to detect the dead angle, more importantly regularly patrols and examines that the hysteresis quality is big, can't in time discover and eliminate hidden danger after dangerous gas leaks. In order to solve the problem of timely detection of invisible gas leakage, many new theories and new methods are emerging continuously in recent years. Among them, the infrared imaging technology capable of capturing the gas trace becomes one of new gas detection technologies with great development potential.
The hazardous gas leakage detection technology based on infrared video image processing has the advantages of non-contact, all-weather, real-time and the like, and has a good application prospect in the intelligent development of future industrial production. However, the prior art is facing the serious challenge of difficult accurate detection of leaking gas under dynamic target disturbance. Compared with a tiny gas leakage area, complex moving targets such as pedestrians, vehicles, unavoidable background clouds, wind-blown trees and the like which randomly appear can form strong interference on detection of gas leakage. Under the interference, the prior art is difficult to accurately judge and position the leaked gas, the final detection result still needs manual auxiliary judgment, and the false detection rate is higher under the condition of separating from manual intervention.
By observing infrared videos of gas and common moving targets, the gas is different from common moving objects in the aspects of form and motion law, so that a method capable of extracting the characteristics of the gas and the common moving targets and distinguishing the gas from other moving target interferences according to the obtained characteristics needs to be designed, and a gas leakage detection system can independently and intelligently detect a gas region from other moving target interferences correctly.
Disclosure of Invention
The invention provides a real-time, accurate and strong-robustness infrared video detection method for dangerous gas leakage, aiming at the defects of low precision, poor real-time performance and the like of detection of leaked gas in a pseudo-target motion strong interference environment in the prior art.
The technical scheme adopted by the invention for solving the technical problems is as follows:
the method for detecting the dangerous gas leakage infrared video under the strong interference of the motion of the false target comprises the following steps:
s1, establishing an RGB color lookup table T;
s2, collecting an infrared video image of the leaked gas, and then performing background modeling by using the initialized data;
s3, capturing the current infrared image frame, and combining the current infrared image frame with the background image to obtain a differential image;
s4, performing binarization processing on the difference image through a preset threshold value, wherein the background is set to be 0, and the foreground is set to be 255;
s5, reducing the size of the binary image, traversing the binary image by using a 4-neighborhood connected domain algorithm to obtain the coordinates of each connected domain;
s6, mapping each connected domain coordinate in the reduced image back to the original size image, and screening a plurality of connected domains in the original size image by using the transverse-longitudinal ratio of the connected domains, the pixel points of the domains and the ratio of the connected domains to the to-be-tracked domain to obtain a plurality of candidate moving target domains;
s7, feature extraction is performed for each candidate region using the FAST feature point detection algorithm with 16 passes. Different from the FAST algorithm, the gray value intensity difference between continuous 8 pixels and the central pixel is considered to be larger than a preset threshold value in the algorithm, and the point can be defined as a characteristic point;
s8, screening each alternative area by combining PPP (Points-Per-Pixel) conditions and the characteristics extracted from the alternative areas, wherein Points correspond to the number of the characteristic Points extracted by a FAST characteristic point detection algorithm, and pixels correspond to the number of pixels of a single connected area. The areas that pass the screening are regarded as areas suspected of gas leakage, and these areas continue to execute step S9, and the areas that fail the screening condition are regarded as disturbance target areas. If all the candidate regions fail to pass the screening, the algorithm jumps back to step S3;
s9, carrying out binary clustering on the suspected gas leakage area by using a k-means clustering algorithm, setting the background to be 0 and the foreground to be 255;
and (5) after S10 and k-means clustering, the area still marked as 255 is the finally detected leakage gas area, pseudo color is added to the area by using a color lookup table T, and other areas are endowed with output display of corresponding gray values.
In the method of the present invention, in step S3, the first N frames of images after the image is captured are stabilized by the infrared camera, and the average value of the images is taken to obtain a stable infrared background image.
In the method of the invention, in S5, the connected domain algorithm pair of the 4 neighborhoods is narrowed to the original 1/4nThe size of the binary image is traversed, so that the time consumed by traversing can be greatly saved while the detection effect is not influenced, and the algorithm can achieve the performance of real-time detection in a general computer and an embedded system.
In step S7 of the present invention, the specific steps include:
s71, taking the current pixel point as the center, and having a radius of 3, totally 16 neighborhood pixel points on a circle, defining the gray value pt and the threshold Th of the center pixel point, and traversing the 16 neighborhood points p of the center pixel pointi,i＝1,2,…,16。
S72, sequentially traversing 16 pixel points p on the circumference of the central pixel pointiAnd each dot is given a label liIf p isiIf + Th is less than pt, corresponding to pixel point p on the circumferenceiMarker (a) ofi0, if piIf pt + Th is greater, the corresponding pixel point p on the circumferenceiMarker (a) ofi1, other cases, pixel point p on the circumferenceiMarker (a) ofi＝2；
S73, any pixel point p on the circumference of the central pixel pointiThe label l starting to traverse the pointiIf 8 continuous points are marked as 0 or 1, marking the central pixel point as a FAST characteristic point;
in the step S8 of the screening method for the moving target interference area and the gas area to be detected, a PPP (Points-Per-Pixel) judgment condition is adopted, and the total number Num of Pixel Points of each alternative moving target area is respectively calculatedpixelAnd the number Num of FAST characteristic pointsFAST(ii) a Then through PPP ═ NumFAST/NumpixelIs equal to a predetermined threshold ThPPPComparing, if the conditions are met, judging that the current area is a suspected gas area and reserving coordinates of the area for the next processing; if the condition is not met, the current area is considered to be an interference moving target, the area is not processed, and the next area is continuously judged; when all the regions are judged to be the interference moving targets, the current frame is abandoned, and the step S3 is skipped to process the next frame.
In step S10, the pseudo color display method according to the present invention is characterized in that, traversing an image, and assigning gray values of corresponding pixel points of an original image to 3 channels of the image by using original image information outside an unreserved connected domain area; in the reserved connected region, the gray value of the pixel point is 0 in the background region, the gray value of the original image is used for assignment, and for the gray value of the pixel point is 255 in the foreground region, the color lookup table T is used for assignment by combining the gray value of the current corresponding pixel point as an index.
The invention has the following beneficial effects:
1. the accuracy is high. By utilizing improved FAST characteristics and PPP region screening conditions, gas can be correctly identified/detected under the interference of a plurality of moving targets, other interference is eliminated, and the robustness is strong;
2. the real-time performance is strong. By the key steps of scaling the size and the like, the frame-by-frame processing speed is greatly improved, and the real-time requirement in practical application can be met.
Drawings
The invention will be further described with reference to the accompanying drawings and examples, in which:
FIG. 1 is a flow chart of a dangerous gas leakage infrared video detection method under strong interference of pseudo target motion according to an embodiment of the invention;
FIG. 2 is an infrared background model image;
FIG. 3 is an exemplary diagram of an infrared image with gas leakage, with an interfering moving object (person) and gas;
FIG. 4 is a difference image obtained by differentiating the infrared background models of FIG. 3 and FIG. 2 according to an embodiment of the present invention;
fig. 5 is a binary image obtained by performing a preset threshold process on a difference image;
fig. 6 is an exemplary diagram of an original image area corresponding to the alternative area in fig. 5 after FAST feature extraction (FAST feature points are white points in the diagram) according to an embodiment of the present invention;
FIG. 7 is a pseudo-color schematic of the results of a leak gas detection where a gas zone is detected and a false target zone of a person is identified as interfering and not shown with a pseudo-color logo;
FIG. 8 is a schematic diagram of a 16-way FAST feature detection template.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The method for detecting the dangerous gas leakage infrared video under the strong interference of the moving target, disclosed by the embodiment of the invention, as shown in figure 1, comprises the following steps of:
s1, establishing an RGB color lookup table T;
s2, collecting the gas leakage infrared video image, converting the image into a single-channel gray image, and establishing an infrared background image B by using the previous N frames of images, as shown in FIG. 2;
s3, capturing the current infrared image frame, as shown in fig. 3, converting the current infrared image frame into a single-channel grayscale image I, and obtaining a difference image D ═ absI-B |, as shown in fig. 4, using the background image B;
s4, binarizes the difference image by a preset threshold th, where the background is 0 and the foreground is 255. Because the spectrum range of the infrared camera for imaging becomes smaller and the imaging effect becomes worse after the light source is filtered by adding the optical filter, the method is obviously characterized in that the contrast of the infrared image is reduced and the signal-to-noise ratio is reduced, in the embodiment of the invention, a smaller preset threshold th is set to be 10, the difference image D shown in figure 4 is subjected to binarization processing through the threshold th to obtain a binary image A, and the result is shown in figure 5;
s5, reducing the size of the binary image, and utilizing a 4-neighborhood connected domain algorithm to perform the reduction on the binary image
Figure BDA0001398096010000054
Traversing to obtain the coordinates R of each connected region1,R2,…,RN. Since the proportion of the leakage gas in the image is large, the original image is traversed by the connected domain, although the effect is best, but the speed is greatly influenced, so in the embodiment of the invention, the original image 1/4 is adoptednThe thumbnail with the size of (n is 1), namely, odd-numbered rows and odd-numbered columns of pixel points of the original image are reserved, and even-numbered rows and even-numbered columns of pixel points of the original image are deleted;
in a preferred embodiment of the present invention, the traversal of the connected component in step S5 specifically includes the steps of:
s51 traversing the binary image
Figure BDA0001398096010000051
Until the current pixel point
Figure BDA0001398096010000052
S511, point-to-point
Figure BDA0001398096010000053
As an initial point, giving a label to the initial point, and then pressing all foreground pixel points with the gray value of 255 adjacent to the initial point into a stack; respectively assigning the point coordinates x and y to connected domain pole coordinates minX, maxX, minY and maxY;
s512, popping up the pixel points on the top of the stack, giving the same label to the pixel points, and then pressing all the foreground pixel points with the gray value of 255 adjacent to the current pixel point into the stack; comparing the current point coordinate with the connected domain pole coordinate, and storing the connected domain pole coordinates minX, minY, maxX and maxY; circulating the current step until the stack is empty;
s52, increasing label by 1, and if label is 255, continuing increasing label by 1; repeating the step S51 until the image traversal is finished;
s6, reducing the image
Figure BDA0001398096010000061
Mapping the coordinates of each connected domain back to the original size image A, and screening the existing connected domains by using the transverse-longitudinal ratio of the connected domains, the area pixel points and the ratio of the connected domains to the area to be tracked to obtain a plurality of alternative moving target areas;
because the proportion of the leakage gas in the differential image is large, the noise is relatively isolated, and the area of the connected region is small, the people count the pixels in the connected region, keep the connected region block which is 5 times the total number of the pixels, and the reserved connected region must meet the condition that the total number of the pixels is more than 400, otherwise, the pixels are eliminated; because the infrared camera shakes during framing, the image after difference is easy to generate the contour line of the object, and for the situation, a screening condition of the transverse-longitudinal ratio is provided; in the embodiment of the invention, the connected region with the aspect ratio larger than 10 is removed; the ratio of the gas leakage area to the tracking rectangular area is relatively fixed; in the embodiment of the invention, the area with the ratio of the foreground area to the tracking rectangular area in the communication area being more than 0.3 and less than 0.7 is reserved, and other communication areas are removed;
s7, extracting features of each candidate region by using a 16-communicated FAST (features From accessed Segment test) feature point detection algorithm, and judging the relationship between each point in the connected region and a neighborhood pixel point, wherein if the gray value intensity difference between continuous 8 pixel points in the 16 neighborhood pixel points of the point and a center pixel point is greater than a preset threshold Th, the point can be defined as a feature point; as shown in fig. 6, the white point in the image is the extracted FAST feature point;
in the embodiment of the present invention, step S7 specifically includes the steps of:
s71, centering on the current pixel point and taking the radius as3, as shown in fig. 8, defining a gray value pt and a threshold Th of the central pixel point, and traversing 16 neighborhood points p of the central pixel pointi,i＝1,2,…,16。
S72, sequentially traversing 16 pixel points p on the circumference of the central pixel pointiAnd each dot is given a label liIf p isiIf + Th is less than pt, corresponding to pixel point p on the circumferenceiMarker (a) ofi0, if piIf pt + Th is greater, the corresponding pixel point p on the circumferenceiMarker (a) ofi1, other cases, pixel point p on the circumferenceiMarker (a) ofi＝2；
S73, any pixel point p on the circumference of the central pixel pointiThe label l starting to traverse the pointiIf 8 continuous points are marked as 0 or 1, marking the central pixel point as a FAST characteristic point;
s8, screening each alternative area by combining PPP (Points-Per-Pixel) conditions and the characteristics extracted from the alternative areas, wherein Points correspond to the number of the characteristic Points extracted by a FAST characteristic point detection algorithm, and pixels correspond to the number of pixels of a single connected area. The areas that pass the screening are regarded as areas suspected of gas leakage, and these areas continue to execute step S9, and the areas that fail the screening condition are regarded as disturbance target areas. If all the candidate regions fail to pass the screening, the algorithm jumps back to step S3;
s9, carrying out binary clustering on the suspected gas leakage area by using a k-means clustering algorithm, setting the background to be 0 and the foreground to be 255;
the region which is still marked as 255 after being clustered by S10 and k-means is the finally detected leakage gas region, pseudo color is added to the region by using a color lookup table T, and the other regions are endowed with output display of corresponding gray values, as shown in FIG. 7.
The infrared video detection method for dangerous gas leakage under the strong interference of the moving target, provided by the invention, has the following advantages:
1. the accuracy is as follows: according to the method, the infrared difference image is traversed through the connected domain to obtain a plurality of connected regions, the regions to be tracked are screened through a series of judgment conditions, and interference objects which are easy to cause false detection are removed by combining with FAST characteristic points, so that the accuracy of the algorithm for detecting the leaked gas is ensured;
2. real-time performance: on a general computer, the running time of each frame of image is not more than 0.1 second, the embedded platform is used for processing, the processing time is shortened to be within 0.07 second, the real-time requirement of gas leakage detection is met, and a good visual effect is achieved;
3. the robustness is good: the method provided by the invention passes the test under the complex background environment, the time of processing the video and the real-time processing exceeds 1000 hours, the program runs stably, the robustness is strong, and the satisfactory gas detection effect is obtained.
It will be understood that modifications and variations can be made by persons skilled in the art in light of the above teachings and all such modifications and variations are intended to be included within the scope of the invention as defined in the appended claims.

Claims (4)

1. A dangerous gas leakage infrared video detection method under strong interference of pseudo-target motion is characterized by comprising the following steps:
s1, establishing an RGB color lookup table T;
s2, collecting an infrared video image of the leaked gas, and then performing background modeling by using the initialized data;
s3, capturing the current infrared image frame, and obtaining a difference image by combining the current infrared image frame with the background image;
s4, performing binarization processing on the difference image through a preset threshold value, wherein the background is set to be 0, and the foreground is set to be 255;
s5, reducing the size of the binary image, traversing the binary image by using a 4-neighborhood connected domain algorithm to obtain the coordinates of each connected domain;
s6, mapping each connected domain coordinate in the reduced image back to the original size image, and screening a plurality of connected domains in the original size image by using the transverse-longitudinal ratio of the connected domains, the area pixel points and the ratio of the connected domains to the area to be tracked to obtain a plurality of candidate target areas;
s7, extracting features of each candidate target area by using a 16-communicated FAST feature point detection algorithm, and defining a current pixel point as a feature point when the gray value intensity difference between continuous 8 pixels and a central pixel is greater than a preset threshold value;
s8, screening each alternative target area by combining Points-Per-Pixel judgment conditions and characteristics extracted from the alternative target areas, wherein Points correspond to the number of characteristic Points extracted by a FAST characteristic point detection algorithm, and pixels correspond to the number of pixels in a single connected area; the areas that pass the screening are regarded as areas suspected of gas leakage, and these areas continue to execute step S9, and the areas that fail the screening condition are regarded as disturbance target areas; if all the candidate target areas fail to pass the screening, the algorithm jumps back to step S3;
s9, carrying out binary clustering on the suspected gas leakage area by using a k-means clustering algorithm, setting the background to be 0 and the foreground to be 255;
s10, the area which is still marked as 255 after k-means clustering is the finally detected leakage gas area, the pseudo color is added to the area by using an RGB color lookup table T, and other areas are endowed with output display of corresponding gray values.
2. The method of claim 1, wherein the reduction to original 1/4 is performed in step S5 by a connected component area algorithm of 4 neighborhoodsnTraversing the binary image with the size, wherein n is a natural number.
3. The method according to claim 1, wherein the step S7 includes the following specific steps:
s71, taking the current pixel point as the center, and having a radius of 3, totally 16 neighborhood pixel points on a circle, defining the gray value pt and the threshold Th of the center pixel point, and traversing the 16 neighborhood points p of the center pixel pointi,i＝1,2,…,16；
S72, sequentially traversing 16 pixel points p on the circumference of the central pixel pointiAnd each dot is given a label liIf p isi+Th＜pt，Then corresponds to the pixel point p on the circumferenceiMarker (a) ofi0, if piIf pt + Th is greater, the corresponding pixel point p on the circumferenceiMarker (a) ofi1, other cases, pixel point p on the circumferenceiMarker (a) ofi＝2；
S73, any pixel point p on the circumference of the central pixel pointiMark of starting to traverse pixel pointiAnd if 8 continuous points are marked as 0 or 1, marking the central pixel point as a FAST characteristic point.
4. The method of claim 1, wherein the FAST feature extracted from each candidate target region in step S8 is determined by means of Points-Per-Pixel decision condition, and the total number Num of pixels of each candidate target region is calculatedpixelAnd the number Num of FAST characteristic pointsFAST(ii) a Then through PPP ═ NumFAST/NumpixelIs equal to a predetermined threshold ThPPPComparing, if the conditions are met, judging that the current area is a suspected gas area and reserving coordinates of the area for the next processing; if the condition is not met, the current area is considered to be an interference moving target area, the area is not processed, and the next area is continuously judged; when all the regions are judged as the interference moving target regions, the current frame is abandoned, and the step S3 is skipped to process the next frame.
CN201710786152.2A 2017-09-04 2017-09-04 Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion Active CN107705288B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201710786152.2A CN107705288B (en) 2017-09-04 2017-09-04 Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201710786152.2A CN107705288B (en) 2017-09-04 2017-09-04 Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion

Publications (2)

Publication Number Publication Date
CN107705288A CN107705288A (en) 2018-02-16
CN107705288B true CN107705288B (en) 2021-06-01

Family

ID=61171954

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201710786152.2A Active CN107705288B (en) 2017-09-04 2017-09-04 Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion

Country Status (1)

Country Link
CN (1) CN107705288B (en)

Families Citing this family (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108921878A (en) * 2018-04-30 2018-11-30 武汉工程大学 Hazardous gas spillage infrared video detection method under moving-target low contrast
CN110557522A (en) * 2018-05-31 2019-12-10 阿里巴巴集团控股有限公司 Method and device for removing video jitter
CN109115418B (en) * 2018-07-12 2020-01-07 浙江工业大学 Pressure container air tightness detection method based on region division
CN109115417B (en) * 2018-07-12 2020-01-14 浙江工业大学 Pressure container air tightness detection method based on connected domain judgment
CN109029861B (en) * 2018-07-12 2020-01-07 浙江工业大学 Pressure container air tightness detection method based on background modeling and centroid clustering
CN109102540B (en) * 2018-08-16 2022-01-28 杭州电子科技大学 FPGA-based lower limit separation and lane division method for marked area blocks
CN109146953B (en) * 2018-09-11 2021-12-10 杭州电子科技大学 FPGA-based upper limit separation and lane separation method for marked area blocks
JP7289684B2 (en) * 2019-03-25 2023-06-12 Ｅｎｅｏｓ株式会社 DETECTION DEVICE, DETECTION SYSTEM AND DETECTION METHOD
CN110487797A (en) * 2019-08-31 2019-11-22 中国石油集团川庆钻探工程有限公司 A kind of separator leakage fluid dram loss gas monitoring method for Oil testing
CN110853077B (en) * 2019-10-17 2023-05-30 广西电网有限责任公司电力科学研究院 Self-adaptive infrared dynamic frame feature extraction method based on morphological change estimation
CN111210405A (en) * 2019-12-27 2020-05-29 山东省特种设备检验研究院有限公司 Volatile organic compound leakage detection method based on infrared imaging
CN112488153B (en) * 2020-11-06 2022-07-05 中国人民解放军国防科技大学 Autonomous searching and positioning method for odor sources based on clustering tendency principle
CN113076883B (en) * 2021-04-08 2022-05-06 西南石油大学 Blowout gas flow velocity measuring method based on image feature recognition
CN113588176B (en) * 2021-07-08 2024-03-08 浙江焜腾红外科技有限公司 Infrared imaging system for monitoring volatile gas
CN115909180B (en) * 2023-01-04 2024-01-26 中环碳和(北京)科技有限公司 Sulfur hexafluoride measuring method

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101432440B1 (en) * 2013-04-29 2014-08-21 홍익대학교 산학협력단 Fire smoke detection method and apparatus
CN105701474B (en) * 2016-01-15 2019-02-05 西安交通大学 A kind of video smoke recognition methods of color combining and external physical characteristic
CN106897720A (en) * 2017-01-11 2017-06-27 济南中维世纪科技有限公司 A kind of firework detecting method and device based on video analysis

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
Machine learning for high-speed corner detection;Edward Rosten and Tom Drummond;《In European Conference on Computer Vision》;20060513;第1-14页 *

Also Published As

Publication number Publication date
CN107705288A (en) 2018-02-16

Similar Documents

Publication Publication Date Title
CN107705288B (en) Infrared video detection method for dangerous gas leakage under strong interference of pseudo-target motion
CN105788142B (en) A kind of fire detection system and detection method based on Computer Vision
KR101403876B1 (en) Method and Apparatus for Vehicle License Plate Recognition
CN107085714B (en) Forest fire detection method based on video
CN103186904B (en) Picture contour extraction method and device
GB2569751A (en) Static infrared thermal image processing-based underground pipe leakage detection method
CN111915704A (en) Apple hierarchical identification method based on deep learning
CN109918971B (en) Method and device for detecting number of people in monitoring video
CN103150549B (en) A kind of road tunnel fire detection method based on the early stage motion feature of smog
CN102982313B (en) The method of Smoke Detection
CN104978567B (en) Vehicle checking method based on scene classification
CN111292228B (en) Lens defect detection method
CN106096603A (en) A kind of dynamic flame detection method merging multiple features and device
CN109544535B (en) Peeping camera detection method and system based on optical filtering characteristics of infrared cut-off filter
CN106228541A (en) Screen positioning method and device in visual inspection
CN113221881B (en) Multi-level smart phone screen defect detection method
CN108133488A (en) A kind of infrared image foreground detection method and equipment
CN113327255A (en) Power transmission line inspection image processing method based on YOLOv3 detection, positioning and cutting and fine-tune
CN112052730B (en) 3D dynamic portrait identification monitoring equipment and method
CN114387269B (en) Fiber yarn defect detection method based on laser
CN107818583A (en) Cross searching detection method and device
CN115880476A (en) Belt deviation detection method based on visible light and infrared image fusion
CN111583341B (en) Cloud deck camera shift detection method
Ning et al. Flame object segmentation by an improved frame difference method
CN107273804A (en) Pedestrian recognition method based on SVMs and depth characteristic

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