CN109145732B - Black smoke vehicle detection method based on Gabor projection - Google Patents

Black smoke vehicle detection method based on Gabor projection Download PDF

Info

Publication number
CN109145732B
CN109145732B CN201810781463.4A CN201810781463A CN109145732B CN 109145732 B CN109145732 B CN 109145732B CN 201810781463 A CN201810781463 A CN 201810781463A CN 109145732 B CN109145732 B CN 109145732B
Authority
CN
China
Prior art keywords
gabor
black smoke
vehicle
projection
key
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
CN201810781463.4A
Other languages
Chinese (zh)
Other versions
CN109145732A (en
Inventor
路小波
陶焕杰
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.)
Southeast University
Original Assignee
Southeast University
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 Southeast University filed Critical Southeast University
Priority to CN201810781463.4A priority Critical patent/CN109145732B/en
Publication of CN109145732A publication Critical patent/CN109145732A/en
Application granted granted Critical
Publication of CN109145732B publication Critical patent/CN109145732B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
    • G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
    • 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
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/25—Fusion techniques
    • G06F18/253—Fusion techniques of extracted features
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
    • G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • General Physics & Mathematics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Multimedia (AREA)
  • Computational Linguistics (AREA)
  • Software Systems (AREA)
  • Traffic Control Systems (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a black smoke vehicle detection method based on Gabor projection, which comprises the following steps: step 1: detecting a moving target from a vehicle monitoring video; step 2: determining the position and the size of a key area by utilizing filtered integral projection and data fitting; and step 3: extracting Gabor projection characteristics of a key region of the vehicle based on the established model, and performing multi-sequence fusion to form a final characteristic vector; and 4, step 4: the extracted feature vectors are classified by using an SVM classifier, and black smoke frames are identified, so that the black smoke vehicle is further detected, the monitoring efficiency of the black smoke vehicle can be improved by using the technical scheme of the invention, and meanwhile, the key area is extracted by using a data fitting method, so that the size of the key area can be acquired in a self-adaptive manner, and the false alarm rate is reduced; the Gabor projection characteristic provided by the invention can reduce the interference of vehicle shadow and further reduce the false alarm rate of black smoke vehicles.

Description

Black smoke vehicle detection method based on Gabor projection
Technical Field
The invention belongs to the technical field of computer vision, and relates to a black tobacco car detection method based on Gabor projection.
Background
In 2018, "environmental management annual newspaper for motor vehicles in china (2018)" issued by the ministry of ecological environment states that along with the continuous high-speed development of world economy and the rapid increase of the number of motor vehicles in cities, motor vehicle pollution becomes an important source of environmental air pollution in China and is an important reason for causing pollution of fine particles and photochemical smog. The urgency for pollution control of motor vehicles is increasingly becoming more prominent. Among the motor vehicle pollution, the pollution of heavy diesel vehicles is of central importance. Heavy diesel vehicle also called black smoke vehicle, generally refers to a highly polluted vehicle with dense black smoke in the tail gas hole of the vehicle, the tail gas emission of the vehicle not only pollutes the air, but also damages the human health, and causes adverse effects on the respiratory system, cardiovascular system and the like of the contacter, therefore, the research on how to accurately, rapidly and intelligently capture the black smoke vehicle from the traffic flow is carried out, and the environmental protection department carries out further quantitative detection and corresponding treatment and correction, which has very important significance for controlling and reducing the tail gas emission of the motor vehicle and improving the urban air quality
The early black smoke vehicle detection method mainly depends on manual work, including people reporting, periodic road inspection, night patrol, manual video monitoring and the like, and the method usually needs to consume a large amount of manpower and financial resources, influences traffic, has low efficiency and is not beneficial to acquiring and storing subsequent law enforcement evidences. Since the last 70 s, various methods for collecting exhaust emission data of motor vehicles appeared in succession, mainly including a chassis dynamometer method, an infrared remote sensing test method, a real-time vehicle-mounted exhaust detection method (PEMS), and the like. The adoption of a vehicle exhaust analysis device to detect black smoke vehicles has many disadvantages. (1) Vehicle exhaust analysis devices tend to be expensive and require a significant amount of capital for subsequent maintenance and upkeep. (2) For a vehicle exhaust analysis device installed on the roadside, the device is often influenced by the driving behavior of a driver passing through a detection interval, and the uncertainty of detection data is high. The conditions of the equipment (light intensity, noise, etc.) and the professional degree of the operator (installation and debugging) can all influence the final detection result. (3) For roadside detection devices, such as multiple vehicles running in parallel, too close to the vehicle, severe weather (wind speed is too high, rain and snow weather), road environment background, nearby vehicle emission, height of an exhaust pipe, and working conditions at the moment of driving, the reliability of monitoring data can be influenced. In addition, the exhaust pipe is hard to detect large vehicles with complex vehicle body structures, strict requirements are placed on the position of the exhaust pipe, and the exhaust pipe needs to be in a light path coverage range. Has more rigorous requirements on traffic conditions and arrangement places, and the use is greatly influenced by environmental factors. (4) For the vehicle-mounted exhaust gas analysis device, due to the increase of the number of vehicles, the feasibility of configuration for each vehicle is difficult, and the maintenance and the inspection of subsequent equipment are time-consuming and labor-consuming.
In recent years, with the continuous improvement of urban road video monitoring systems, the continuous development of computer vision technology and the continuous improvement of computer performance, it becomes possible to detect black smoke vehicles from massive road monitoring videos by using video analysis technology. However, no specific implementation method is available at present.
Disclosure of Invention
In order to solve the problems, the invention discloses a black smoke vehicle detection method based on Gabor projection, which can make up the defect that the traditional method for manually monitoring the black smoke vehicle has low efficiency, reduce the false alarm rate and have certain robustness on shadows.
In order to achieve the purpose, the invention provides the following technical scheme:
a black smoke vehicle detection method based on Gabor projection comprises the following steps:
step 1: detecting a moving target from a vehicle monitoring video;
step 2: determining the position and the size of a key area by utilizing filtered integral projection and data fitting;
and step 3: extracting Gabor projection characteristics of a key region of the vehicle based on the established model, and performing multi-sequence fusion to form a final characteristic vector;
and 4, step 4: and classifying the extracted feature vectors by using an SVM classifier, and identifying black smoke frames so as to further detect the black smoke vehicle.
Further, the step 1 adopts a Gaussian mixture model to perform background modeling.
Further, the step 2 specifically includes the following steps:
step 2.1: the top position coordinates x of the critical area are calculated bykeyI.e. by
Figure GDA0003273035740000021
Wherein, Iobj(x, y) is a vehicle object image IobjCoordinates at point (x, y), w is the width of the vehicle target image, function norm () is data normalization, Δ x is a parameter related to the vehicle tail coordinate calculation;
step 2.2: the width and height of the critical area are calculated as follows,
width＝round(0.8w)
height＝round(H0(xkey))
Figure GDA0003273035740000022
where w is the width of the vehicle object image, HframeReferring to the height of the current frame, round () is a rounding function, H0(xkey) Is a function of height variation, xkeyIndicating the top position ordinate of the key area.
Further, the height variation function is obtained by labeling a plurality of real samples and by linear fitting.
Further, the model established in the step 3 is:
Ismoke＝λ1uvT+λ2
u＝(1,2,...,M)T；v＝(1,2,...,N)T
wherein, IsmokeRefers to a local black smoke image with resolution of MxN, λ1And λ2The two coefficients represent the smoke density attenuation coefficient and the image brightness, respectively.
Further, the calculation of the Gabor projection features in step 3 includes the following steps:
step 3.1: designing Gabor filters with different directions and frequencies, calculating a Gabor filter h (x, y) of a space domain by adopting the following formula,
Figure GDA0003273035740000031
where phi and u0For the phase and frequency, σ, of the plane-of-view wave along the z-coordinate axisyAnd σxSpace constants of the two-dimensional Gaussian envelope along the y axis and the x axis respectively;
setting phi to 0, the frequency domain Gabor filter H (u, v) can be calculated using the following formula,
Figure GDA0003273035740000032
wherein σu＝1/2πσx,σv＝1/2πσy，A＝2πσxσy；
Taking 45 degrees as intervals, dividing 180 degrees into four, adopting two empirical values of 2 and 4 for wavelength, and extracting a Gabor energy diagram of a key region;
step 3.2: some post-processing is performed on the Gabor energy diagram, including Gaussian blur, spatial information addition, normalization and principal component analysis:
the Gabor energy diagram is subjected to Gaussian blur processing by using the following formula to remove noise,
Figure GDA0003273035740000033
wherein σ is a parameter for controlling the degree of blurring;
adding a position map and adding spatial information, so that 4 Gabor characteristics and two spatial position map characteristics can be obtained;
normalizing the features to a 0 mean and a 1 variance using z-score normalization;
and PCA is implemented to obtain main Gabor characteristics;
step 3.3: making vertical integral projection and horizontal integral projection on main Gabor characteristics, and recording GPVFor a vertical integral projection vector of 1 × 120, take GPHIs a horizontal integral projection vector with the size of 1 × 80, and is recorded with FGPIs a Gabor projection feature, then
Figure GDA0003273035740000034
Further, the final feature vector formed by the multi-sequence fusion in step 3 is as follows:
FFINAL(t)＝{FGP(t-1),FGP(t),...,FGP(t+k)}
wherein, FFINAL(t) final feature vector extracted from key region of t-th frame, FGP(t) is the Gabor projection characteristic of the key region of the t-th frame, and k +2 represents the number of sequence analysis.
Further, the step 4 specifically includes the following steps:
step 4.1: classifying all key areas in the current frame image by using a trained SVM classifier, and if at least one key area is identified as a black smoke area, identifying the current frame as a black smoke frame;
step 4.2: if K frames are identified as black smoke frames in every continuous 100 frames and K satisfies K > alpha, the black smoke vehicle is considered to be present in the current video sequence, wherein alpha is an adjusting coefficient for controlling recall rate and precision rate.
Compared with the prior art, the invention has the following advantages and beneficial effects:
1. the method of the invention can greatly improve the law enforcement efficiency. The system can be remotely monitored, does not hinder traffic, realizes all-antenna online watching, is suitable for various road environments such as double lanes and multiple lanes, is convenient to install, is suitable for large-range distribution and control of urban roads, and is easier to form an online monitoring network for high-pollution black smoke vehicles.
2. The method adopts a data fitting method to extract the key area, adaptively obtains the size of the key area and reduces the false alarm rate.
3. The invention provides a novel black smoke model which can be used for depicting the characteristic that the concentration of black smoke tail gas of a black smoke vehicle gradually diffuses from a tail gas hole to the rear of the vehicle.
4. The Gabor projection characteristics provided by the invention have certain robustness to vehicle shadows, can avoid false detection caused by the vehicle shadows to a certain extent, and fully utilize the dynamic characteristics of black smoke by combining a multi-sequence analysis strategy, thereby further increasing the robustness of the characteristics, improving the detection rate and reducing the false alarm rate.
Drawings
FIG. 1 is a flow chart of the method of the present invention.
FIG. 2 is a schematic diagram of the calculation process of Gabor projection features of the present invention.
Fig. 3 is a black smoke vehicle detected from the traffic stream according to the present invention.
Detailed Description
The technical solutions provided by the present invention will be described in detail below with reference to specific examples, and it should be understood that the following specific embodiments are only illustrative of the present invention and are not intended to limit the scope of the present invention.
The invention relates to an intelligent method for automatically analyzing and detecting a black smoke vehicle for a monitoring video, which is characterized in that a high-definition camera arranged on a road monitoring rod is used for collecting the road real-time monitoring video, the road real-time monitoring video is transmitted to a management platform through a network, the intelligent identification is carried out according to the image characteristics of radiation information of visible light on a CCD (Charge-coupled Device) sensor by adopting digital image processing, mode identification and computer vision technology, the key information of the black smoke vehicle in a video source is automatically analyzed and extracted, the massive data in the video image is analyzed at high speed by means of the powerful data processing function of a computer, the information which is not concerned by a user is filtered, and the useful key information of the black smoke vehicle is only provided for a monitor. Once a suspected black smoke vehicle is found, the system automatically gives an alarm, tracks and snapshot the black smoke vehicle running on the road (pollutants are mainly particulate matter PM2.5 and PM10), automatically identifies the license plate, and stores important information such as the vehicle passing time and the vehicle passing place, thereby more effectively assisting the working personnel to find abnormal conditions and process the high-pollution vehicle, and reducing the phenomena of misinformation and missing report to the maximum extent.
Specifically, as shown in fig. 1, the method comprises the following specific steps:
step 1: detecting a moving target from a vehicle monitoring video; in the step, a background modeling is carried out based on a Gaussian mixture model.
Step 2: determining the position and the size of a key area by utilizing filtered integral projection and data fitting; the method specifically comprises the following steps:
step 2.1: the top position coordinates x of the critical area are calculated bykeyI.e. by
Figure GDA0003273035740000051
Wherein, Iobj(x, y) is a vehicle object image IobjCoordinates at point (x, y), w is the width of the vehicle target image, function norm () is data normalization, and Δ x is a parameter related to the vehicle tail coordinate calculation.
Step 2.2: the width and height of the critical area are calculated as follows,
width＝round(0.8w)
height＝round(H0(xkey))
Figure GDA0003273035740000052
where w is the width of the vehicle object image, HframeReferring to the height of the current frame, round () is a rounding function, H0(xkey) Refers to the height variation function. The function is obtained by labeling 400 real samples and by linear fitting, xkeyIndicating the top position ordinate of the key area.
And step 3: extracting Gabor projection characteristics of a key region of the vehicle based on the established model, and performing multi-sequence fusion to form a final characteristic vector;
the black smoke model established in the step is as follows:
Ismoke＝λ1uvT+λ2
u＝(1,2,...,M)T；v＝(1,2,...,N)T
wherein, IsmokeRefers to a local black smoke image with resolution of MxN, λ1And λ2The two coefficients represent the smoke density attenuation coefficient and the image brightness, respectively.
And the calculation of the Gabor projection characteristics comprises the following procedures:
step 3.1: designing Gabor filters with different directions and frequencies, calculating a Gabor filter h (x, y) of a space domain by adopting the following formula,
Figure GDA0003273035740000061
where phi and u0For the phase and frequency, σ, of the plane-of-view wave along the z-coordinate axisyAnd σxSpace constants of the two-dimensional Gaussian envelope along the y axis and the x axis respectively;
setting phi to 0, the frequency domain Gabor filter H (u, v) can be calculated using the following formula,
Figure GDA0003273035740000062
wherein σu＝1/2πσx,σv＝1/2πσy，A＝2πσxσy；
Taking 45 degrees as intervals, dividing 180 degrees into four, adopting two empirical values of 2 and 4 for wavelength, and extracting a Gabor energy diagram of a key region;
step 3.2: some post-processing is performed on the Gabor energy diagram, including Gaussian blur, spatial information addition, normalization and principal component analysis:
the Gabor energy diagram is subjected to Gaussian blur processing by using the following formula to remove noise,
Figure GDA0003273035740000063
wherein σ is a parameter for controlling the degree of blurring;
adding a position map and spatial information, so that 4 Gabor characteristics and two spatial position map characteristics can be obtained
These features were normalized to 0 mean and 1 variance using z-score normalization,
and PCA is implemented to obtain main Gabor characteristics;
step 3.3: making vertical integral projection and horizontal integral projection on main Gabor characteristics, and recording GPVFor a vertical integral projection vector of 1 × 120, take GPHIs a horizontal integral projection vector with the size of 1 × 80, and is recorded with FGPIs a Gabor projection feature, then
Figure GDA0003273035740000064
Fig. 2 is a schematic flow chart of extracting Gabor projection features.
Specifically, the multiple sequences in step 3 are fused to form the final feature vector finger,
FFINAL(t)＝{FGP(t-1),FGP(t),...,FGP(t+k)}
wherein, FFINAL(t) final feature vector extracted from key region of t-th frame, FGP(t) is the Gabor projection characteristic of the key region of the t-th frame, and k +2 represents the number of sequence analysis, generally taking 3-6.
And 4, step 4: the SVM classifier is used for classifying the extracted feature vectors and recognizing black smoke frames, so that the black smoke vehicle is further detected, and the method specifically comprises the following steps:
step 4.1: classifying all key areas in the current frame image by using a trained SVM classifier, and if at least one key area is identified as a black smoke area, identifying the current frame as a black smoke frame;
step 4.2: if K frames are identified as black smoke frames in every 100 continuous frames and K satisfies K > alpha (alpha is an adjusting coefficient for controlling recall rate and precision rate), the current video sequence is considered to have black smoke cars.
Fig. 3 shows an example of a black smoke car detected from the traffic stream by the method of the present invention, the thin black rectangle representing the detected moving object and the thick black rectangle representing the position of the key area. Obviously, the method can detect the black smoke car through the video image.
The technical means disclosed in the invention scheme are not limited to the technical means disclosed in the above embodiments, but also include the technical scheme formed by any combination of the above technical features. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and such improvements and modifications are also considered to be within the scope of the present invention.

Claims (5)

1. A black smoke vehicle detection method based on Gabor projection is characterized by comprising the following steps:
step 1: detecting a moving target from a vehicle monitoring video;
step 2: determining the position and the size of a key area by utilizing filtered integral projection and data fitting, which specifically comprises the following processes:
step 2.1: the top position ordinate x of the critical area is calculated in the following mannerkeyI.e. by
Figure FDA0003273035730000011
Wherein, Iobj(x, y) is a vehicle object image IobjCoordinates at point (x, y), w is the width of the vehicle target image, function norm () is data normalization, Δ x is a parameter related to the vehicle tail coordinate calculation;
step 2.2: the width and height of the critical area are calculated as follows,
width＝round(0.8w)
height＝round(H0(xkey))
Figure FDA0003273035730000012
where w is the width of the vehicle object image, HframeReferring to the height of the current frame, round () is a rounding function, H0(xkey) Is a function of height variation, xkeyA top position ordinate representing a key region;
and step 3: extracting Gabor projection characteristics of a key region of the vehicle based on the established model, and performing multi-sequence fusion to form a final characteristic vector, wherein the established model is as follows:
Ismoke＝λ1uvT+λ2
u＝(1,2,...,M)T；v＝(1,2,...,N)T
wherein, IsmokeRefers to a local black smoke image with resolution of MxN, λ1And λ2Two coefficients respectively representing the smoke density attenuation coefficient and the image brightness;
the calculation of the Gabor projection characteristics comprises the following procedures:
step 3.1: designing Gabor filters with different directions and frequencies, calculating a Gabor filter h (x, y) of a space domain by adopting the following formula,
Figure FDA0003273035730000013
where phi and u0Is viewed as a plane wave edgePhase and frequency of the z-axis, σyAnd σxSpace constants of the two-dimensional Gaussian envelope along the y axis and the x axis respectively;
setting phi to 0, the frequency domain Gabor filter H (u, v) can be calculated using the following formula,
Figure FDA0003273035730000021
wherein σu＝1/2πσx,σv＝1/2πσy，A＝2πσxσy；
Taking 45 degrees as intervals, dividing 180 degrees into four, adopting two empirical values of 2 and 4 for wavelength, and extracting a Gabor energy diagram of a key region;
step 3.2: some post-processing is performed on the Gabor energy diagram, including Gaussian blur, spatial information addition, normalization and principal component analysis:
the Gabor energy diagram is subjected to Gaussian blur processing by using the following formula to remove noise,
Figure FDA0003273035730000022
wherein σ is a parameter for controlling the degree of blurring;
adding a position map and adding spatial information, so that 4 Gabor characteristics and two spatial position map characteristics can be obtained;
normalizing the features to a 0 mean and a 1 variance using z-score normalization;
and PCA is implemented to obtain main Gabor characteristics;
step 3.3: making vertical integral projection and horizontal integral projection on main Gabor characteristics, and recording GPVFor a vertical integral projection vector of 1 × 120, take GPHIs a horizontal integral projection vector with the size of 1 × 80, and is recorded with FGPIs a Gabor projection feature, then
Figure FDA0003273035730000023
And 4, step 4: and classifying the extracted feature vectors by using an SVM classifier, and identifying black smoke frames so as to further detect the black smoke vehicle.
2. The method for detecting the black smoke based on the Gabor projection according to claim 1, wherein the step 1 adopts a Gaussian mixture model for background modeling.
3. The black smoke vehicle detection method based on Gabor projection according to claim 1, wherein: the height variation function is obtained by labeling a plurality of real samples and by linear fitting.
4. The black smoke vehicle detection method based on Gabor projection according to claim 1, wherein: the final feature vector formed by the multi-sequence fusion in the step 3 is as follows:
FFINAL(t)＝{FGP(t-1),FGP(t),...,FGP(t+k)}
wherein, FFINAL(t) final feature vector extracted from key region of t-th frame, FGP(t) is the Gabor projection characteristic of the key region of the t-th frame, and k +2 represents the number of sequence analysis.
5. The black smoke vehicle detection method based on Gabor projection according to claim 1, wherein: the step 4 specifically comprises the following steps:
step 4.1: classifying all key areas in the current frame image by using a trained SVM classifier, and if at least one key area is identified as a black smoke area, identifying the current frame as a black smoke frame;
step 4.2: if K frames are identified as black smoke frames in every continuous 100 frames and K satisfies K > alpha, the black smoke vehicle is considered to be present in the current video sequence, wherein alpha is an adjusting coefficient for controlling recall rate and precision rate.
CN201810781463.4A 2018-07-17 2018-07-17 Black smoke vehicle detection method based on Gabor projection Active CN109145732B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810781463.4A CN109145732B (en) 2018-07-17 2018-07-17 Black smoke vehicle detection method based on Gabor projection

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810781463.4A CN109145732B (en) 2018-07-17 2018-07-17 Black smoke vehicle detection method based on Gabor projection

Publications (2)

Publication Number Publication Date
CN109145732A CN109145732A (en) 2019-01-04
CN109145732B true CN109145732B (en) 2022-02-15

Family

ID=64800661

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810781463.4A Active CN109145732B (en) 2018-07-17 2018-07-17 Black smoke vehicle detection method based on Gabor projection

Country Status (1)

Country Link
CN (1) CN109145732B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110517280B (en) * 2019-08-14 2021-11-16 武汉理工大学 Ship tail gas monitoring system based on infrared recognition

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103324958B (en) * 2013-06-28 2016-04-27 浙江大学苏州工业技术研究院 Based on the license plate locating method of sciagraphy and SVM under a kind of complex background
CN105809954B (en) * 2014-12-30 2018-03-16 清华大学 Traffic incidents detection method and system
KR20180073327A (en) * 2016-12-22 2018-07-02 삼성전자주식회사 Display control method, storage medium and electronic device for displaying image
CN106951821A (en) * 2016-12-27 2017-07-14 湘潭大学 A kind of black smoke car intelligent monitoring recognition methods based on image processing techniques

Also Published As

Publication number Publication date
CN109145732A (en) 2019-01-04

Similar Documents

Publication Publication Date Title
CN105744232B (en) A kind of method of the transmission line of electricity video external force damage prevention of Behavior-based control analytical technology
CN103714363B (en) A kind of motor vehicle exhaust smoke video identification system
CN103093249B (en) A kind of taxi identification method based on HD video and system
CN107478590A (en) A kind of method of combination motor vehicle intelligent vision identification and remote exhaust emission detection
CN109409337B (en) Muck vehicle feature identification method based on convolutional neural network
CN105809679A (en) Mountain railway side slope rockfall detection method based on visual analysis
CN108269407B (en) Security robot capable of automatically managing people stream and logistics
CN109448365A (en) Across the scale space base land regions road traffic system integrated supervision method of one kind
CN111126171A (en) Vehicle reverse running detection method and system
CN109086682B (en) Intelligent video black smoke vehicle detection method based on multi-feature fusion
CN105915840A (en) Factory smoke emission automatic monitoring method based on video signal
CN112818834B (en) Method, device and medium for judging avoidance of emergency vehicle at intersection
CN107516423B (en) Video-based vehicle driving direction detection method
CN110691224A (en) Transformer substation perimeter video intelligent detection system
CN112289022B (en) Black smoke vehicle detection and judgment method and system based on space-time background comparison
CN111783700B (en) Automatic recognition and early warning method and system for pavement foreign matters
CN109191492B (en) Intelligent video black smoke vehicle detection method based on contour analysis
CN113450573A (en) Traffic monitoring method and traffic monitoring system based on unmanned aerial vehicle image recognition
Khan et al. Cyber physical system for vehicle counting and emission monitoring
CN109145732B (en) Black smoke vehicle detection method based on Gabor projection
CN115100555B (en) Forest fire smoke detection method based on multi-motion characteristics in cloud and fog environment
Patel et al. An algorithm for automatic license plate detection from video using corner features
Zhao et al. Research on vehicle detection and vehicle type recognition under cloud computer vision
Jehad et al. Developing and validating a real time video based traffic counting and classification
CN114266837A (en) Road sign detection method, device, equipment and computer storage medium

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