CN106682566A - Traffic accident detection method, traffic accident detection device and electronic device - Google Patents
Traffic accident detection method, traffic accident detection device and electronic device Download PDFInfo
- Publication number
- CN106682566A CN106682566A CN201510755728.XA CN201510755728A CN106682566A CN 106682566 A CN106682566 A CN 106682566A CN 201510755728 A CN201510755728 A CN 201510755728A CN 106682566 A CN106682566 A CN 106682566A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- pixel value
- accident
- relevant position
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
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
-
- 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
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20224—Image subtraction
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30241—Trajectory
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The embodiments of the invention provide a traffic accident detection method, a traffic accident detection device and an electronic device. The electronic device comprises a first detection unit used to detect the foreground image of a traffic monitoring image; a first setting unit used to set the pixel values of the pixels at the corresponding positions in the foreground counting image according to the pixel values of the pixels in the foreground image; a second setting unit used to set the pixel values of the pixels at the corresponding positions in the traffic accident image according to the pixel values of the pixels at the corresponding positions in the foreground counting image; and a first processing unit used to conduct morphological treatment on the traffic accident image. According to the embodiments of the invention, with a simple method, the position of the traffic accident in a traffic monitoring image could be detected; and the detection accuracy is also increased.
Description
Technical field
The application is related to a kind of detection side of traffic accident in areas of information technology, more particularly to detection traffic monitoring image
Method, device and electronic equipment.
Background technology
With expanding economy, increasing vehicle enters the life of people, but, by the rapid increasing of vehicle fleet size
Plus the problems such as the traffic jam brought and traffic safety it is also even more serious.Have benefited from the development of information technology, Ren Menti
The concept of intelligent transportation (Intelligent Transport) is gone out, it is desirable to which traffic problems are solved by technological means.
Traffic condition detection is a part for intelligent transportation, and it can provide important information for traffic administration.In intelligence
In traffic technique, image processing techniques can be generally adopted, traffic monitoring image is analyzed, so as to obtain traffic
Status information.
In the prior art, generally using machine learning method (Machine learning based method) and method of loci
(Trajectory based method) to detect traffic monitoring image in traffic accident information.Wherein, machine learning
Method obtains traffic accident information by the anomalous condition in detection traffic monitoring image；Method of loci by detect vehicle be
No long-time is detained, and the collision between vehicle is obtaining traffic accident information.
It should be noted that the introduction of technical background is intended merely to above the convenient technical scheme to the application carry out it is clear,
Complete explanation, and facilitate the understanding of those skilled in the art and illustrate.Can not be merely because these schemes be in this Shen
Background section please is set forth and thinks that above-mentioned technical proposal is known to those skilled in the art.
The content of the invention
Inventors herein have recognized that, in above-mentioned machine learning method, need to obtain substantial amounts of training sample to carry out
Training, it is complicated and time-consuming；In above-mentioned method of loci, testing result depends on tracing algorithm, it can be difficult to obtaining foot
Enough good tracing algorithms, therefore the accuracy for detecting is limited.
The application provides a kind of traffic accident detection method, detection means and electronic equipment, and the detection method is according to traffic
The position of the pixel with predetermined pixel value in the foreground image of monitoring image, arranges the picture of respective pixel in accident image
Element value, the position thus, it is possible to detect traffic accident in simple method, also, by processing accident image,
The degree of accuracy of detection can be improved.
According to the first aspect of the embodiment of the present application, there is provided a kind of traffic accident detection means, for detecting traffic monitoring
There is the position of traffic accident in image, including：
First detector unit, it is used to detect the foreground image of traffic monitoring image；
First setting unit,, according to the pixel value of pixel in the foreground image, the prospect that arranges counts corresponding in image for it
The pixel value of the pixel of position；
Second setting unit, it counts the pixel value of the pixel of relevant position in image according to the prospect, arranges accident
The pixel value of the pixel of relevant position in image；
First processing units, it is used to carry out Morphological scale-space to the accident image,
Wherein, the foreground image, the prospect count image and the accident image is all of the same size,
And the foreground image and the accident image are bianry image.
According to the second aspect of the embodiment of the present application, there is provided a kind of electronic equipment, the electronic equipment includes real as described above
Apply the traffic accident detection means described in a first aspect.
According to the third aspect of the embodiment of the present application, there is provided a kind of traffic accident detection method, for detecting traffic monitoring
There is the position of traffic accident in image, including：
The foreground image of detection traffic monitoring image；
According to the pixel value of pixel in the foreground image, the pixel that prospect counts the pixel of relevant position in image is set
Value；
The pixel value of the pixel of relevant position in image is counted according to the prospect, relevant position in accident image is set
The pixel value of pixel；
Morphological scale-space is carried out to the accident image, wherein, the foreground image, the prospect count image, with
And the accident image is all of the same size, and the foreground image and the accident image are bianry image.
The beneficial effect of the application is：Can detect in simple method and occur in traffic monitoring image traffic accident
Position, further, it is possible to improve the degree of accuracy of detection.
With reference to explanation hereinafter and accompanying drawing, only certain exemplary embodiments of this invention is disclose in detail, specify the original of the present invention
Reason can be in adopted mode.It should be understood that embodiments of the present invention are not so limited in scope.
In the range of the spirit and terms of claims, embodiments of the present invention include many changes, modifications and equivalent.
The feature for describing for a kind of embodiment and/or illustrating can be in same or similar mode one or more
It is combined with the feature in other embodiment used in individual other embodiment, or substitute in other embodiment
Feature.
It should be emphasized that term "comprises/comprising" refers to the presence of feature, one integral piece, step or component when using herein,
But it is not precluded from the presence of one or more further features, one integral piece, step or component or additional.
Description of the drawings
Included accompanying drawing is used for providing being further understood from the embodiment of the present invention, which constitutes of specification
Point, for illustrating embodiments of the present invention, and come together to explain the principle of the present invention with word description.Obviously
Ground, drawings in the following description are only some embodiments of the present invention, for those of ordinary skill in the art,
Without having to pay creative labor, can be with according to these other accompanying drawings of accompanying drawings acquisition.In the accompanying drawings：
Fig. 1 is a schematic flow sheet of the traffic accident detection method of the embodiment of the present application；
Fig. 2 is a schematic flow sheet using the traffic accident detection method detection traffic accident position of the present embodiment；
Fig. 3 is a composition schematic diagram of the traffic accident detection means of the embodiment of the present application；
Fig. 4 is a composition schematic diagram of the electronic equipment of the embodiment of the present application.
Specific embodiment
Referring to the drawings, by description below, the aforementioned and further feature of the present invention will be apparent from.In explanation
In book and accompanying drawing, only certain exemplary embodiments of this invention is specifically disclosed, which show can wherein adopt the original of the present invention
Some embodiments then, it will thus be appreciated that the invention is not restricted to described embodiment, conversely, present invention bag
Include whole modifications, modification and the equivalent for falling within the scope of the appended claims.
Embodiment 1
The embodiment of the present application 1 provides a kind of traffic accident detection method, for detecting traffic monitoring image in there is traffic
The position of accident.Fig. 1 is a schematic flow sheet of the traffic accident detection method of embodiment 1, as shown in figure 1,
The method includes：
The foreground image of S101, detection traffic monitoring image；
S102, according to the pixel value of pixel in the foreground image, the pixel that prospect counts relevant position in image is set
Pixel value；
S103, the pixel value that the pixel of relevant position in image is counted according to the prospect, arrange corresponding in accident image
The pixel value of the pixel of position；
S104, Morphological scale-space is carried out to the accident image.
In the present embodiment, prospect (Foreground) image, prospect count (Foreground counter) figure
Picture and accident (Incident) image are all of the same size.
In the present embodiment, according to the position of the pixel with predetermined pixel value in the foreground image of traffic monitoring image,
The pixel value of the pixel of relevant position in accident image is set, the position thus, it is possible to detect traffic accident in simple method
Put, and make accident that position to occur and be intuitively embodied on accident image；Also, by carrying out morphology to accident image
Process, testing result can be made more accurate.
The step of the present embodiment in S101, can adopt specific algorithm, by the present frame of traffic monitoring image with
Background image is compared, and to obtain the foreground image of the frame, and the foreground image can not only present fortune at a high speed
Dynamic object, can also be presented the static object of the object and long-time of low-speed motion, so as to be adapted to detect for due to traffic
Situations such as traffic jam caused by accident.The specific algorithm for example may be referred to non-patent document " An Enhanced
Background Estimation Algorithm for Vehicle Detection in Urban Traffic Scenes”(Jose
Manuel Milla et al, IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, VOL.59,
NO.8, OCTOBER 2010), or can be other algorithms.The foreground image can be bianry image, its
In, pixel value is that 1 pixel can correspond to the foreground objects such as the vehicle in traffic monitoring image in addition to background.
The step of the present embodiment in S102, for each frame foreground image, each of which pixel can be detected
Pixel value, when the pixel has the first pixel value, can make prospect count the pixel of the pixel of relevant position in image
Value increases, wherein, first pixel value for example can be 1.For example, if coordinate is in the i-th frame foreground image
The pixel p at (X=100, Y=101) placefi(100,101)Pixel value be 1, then, make prospect count image in coordinate be
The pixel p at (X=100, Y=101) placec(100,101)Pixel value increase by 1, thus, prospect count image in each
The pixel value of pixel can reflect the accumulation result of the pixel value of the pixel of relevant position in multiframe foreground image, prospect meter
The higher region of the higher pixel correspondence car blockage degree of pixel value in number image, also, prospect counts picture in image
The possibility that region corresponding to the higher pixel of element value occurs traffic accident is also higher.
The step of the present embodiment in S103, this can be judged according to the pixel value of pixel in prospect counting image
Whether the corresponding region of pixel there is traffic accident, for example, prospect can be counted the pixel value of each pixel in image
It is compared with first threshold T1, if the pixel value of the pixel is more than or equal to first threshold T1, can be by
The pixel value of the pixel of relevant position is set to the second pixel value in accident image, wherein, the accident image can be two
Value image, and second pixel value can be 1.For example, if prospect count image in coordinate for (X=100,
Y=101) the pixel p at placec(100,101)Pixel value be more than first threshold T1, then be by coordinate in accident image
The pixel p at (X=100, Y=101) placei(100,101)Pixel value be set to 1.Thus, have second in accident image
The pixel of pixel value can correspond to the region for being judged as traffic accident, also, because the accident image is two-value
Image, therefore, it is possible to clear and intuitively reflect that this is judged as the region that traffic accident occurs.
The step of the present embodiment in S104, Morphological scale-space (Morphological can be carried out to accident image
Processing), the pixel with the second pixel value during the Morphological scale-space for example can be to the accident image is carried out
Delete processing and/or expansion process, so as to delete the accident image in some isolated pixels with the second pixel value,
And/or connect multiple pixels with the second pixel value.For example, certain can be deleted according to the algorithm of Morphological scale-space
A little pixel values are 1 pixel, and/or connect the pixel that neighbouring multiple pixel values are 1, are formed by continuous picture
Element value is the region that 1 pixel is constituted.Thus, the noise spot in accident image, Neng Goufang are removed by delete processing
Only error detection；Also, neighbouring pixel is connected by expansion process, is easy to be identified incident area and position.
In the present embodiment, the concrete grammar for carrying out Morphological scale-space may be referred to prior art, and the embodiment of the present application is no longer gone to live in the household of one's in-laws on getting married
State.
In the present embodiment, can to carrying out Morphological scale-space after accident image be further processed, for example, can
In to recognize the accident image by the region that constituted of pixel continuously with the second pixel value, and extract the region
Position and/or the information such as area, be identified thereby, it is possible to the region automatically to there is traffic accident, and
More reference informations are provided for intelligent transportation.In the present embodiment, recognize the method for above-mentioned zone and extract the area
The method of relevant information may be referred to prior art in domain, and here is omitted.
In the present embodiment, can be combined with counting the auxiliary counting that image has same size with prospect
(Flick_counter) pixel of the pixel of relevant position during image is to arrange prospect counting image and accident image
Value.
In above-mentioned steps S102 of the present embodiment, for the pixel in each frame foreground image, when the pixel has
During four pixel values, can also make the pixel value of the pixel of relevant position in the auxiliary counting image increases, wherein, should
4th pixel value for example can be 0, thus, have the pixel of higher pixel value in the auxiliary counting image and hand over
The relatively low region correspondence of logical accident probability；Also, when the pixel value of the pixel of relevant position in the auxiliary counting image
During more than or equal to Second Threshold T2, it is possible to determine that traffic thing does not occur in a period of time for the relevant position
Therefore, so, the pixel value of the pixel of the relevant position in the auxiliary counting image can be set to the 3rd pixel value,
The pixel value that prospect counts the pixel of relevant position in image is set to into the 4th pixel value, wherein, the 3rd pixel value
Can be 0, thus, by the counting of auxiliary counting image, prospect can be made to count in image corresponding to without generation
The pixel value zero setting of the pixel of traffic accident position, to update the testing result corresponding to prospect counting image, lays equal stress on
Newly counted.
For example, based on the testing result to (i-2*n) frame to (i-n) frame traffic monitoring image, before judging this
Scape counts pixel p in imagec(100,101)Pixel value be more than first threshold T1, i.e. be judged as the pixel pc(100,101)It is right
The position of traffic accident should occur；Based on the inspection to subsequent (i-n+1) frame to (i) frame traffic monitoring image
Result is surveyed, pixel p in the auxiliary counting image is judgedfc(100,101)Pixel value be more than Second Threshold T2, i.e. judge
For the pixel pfc(100,101)Corresponding to the position that traffic accident does not occur during (i-n+1) frame to (i) frame,
Therefore, by pixel p in the auxiliary counting imagefc(100,101)Pixel value be set to 0, and prospect is counted into pixel in image
pc(100,101)Pixel value be placed in 0, to update testing result, and enable prospect to count image and auxiliary counting image
Enough re-start counting.
Additionally, in above-mentioned steps S103 of the present embodiment, can be with the relevant position in the auxiliary counting image
When the pixel value of pixel is more than or equal to Second Threshold T2, by the pixel value of the pixel of relevant position in the accident image
The 4th pixel value is set to, thus, by the counting of auxiliary counting image, can make to be corresponded in accident image
There is no the pixel value zero setting of the pixel of traffic accident position, to update the testing result that accident image is presented.
For example, based on the testing result to (i-2*n) frame to (i-n) frame traffic monitoring image, hazard plot is made
The pixel p as ini(100,101)Pixel value be 1, i.e. be judged as the pixel pi(100,101)There is the position of traffic accident in correspondence；
Based on the testing result to subsequent (i-n+1) frame to (i) frame traffic monitoring image, the auxiliary counting is judged
Pixel p in imagefc(100,101)Pixel value be more than Second Threshold T2, i.e. be judged as the pixel pfc(100,101)Correspond to
There is no the position of traffic accident during (i-n+1) frame to (i) frame, therefore, by picture in accident image
Plain pi(100,101)Pixel value be placed in 0, to update the testing result that accident image is presented.
Additionally, in the present embodiment, when the pixel in foreground image has the first pixel value, can be by auxiliary counting
The pixel of relevant position is set to have the 3rd pixel value in image, wherein, the 3rd pixel value can be 0, thus,
The pixel for enabling relevant position in auxiliary counting image re-starts counting.
Fig. 2 is to detect occur in traffic monitoring image traffic accident using the traffic accident detection method of the present embodiment
One flow chart of position.As described in Figure 2, the flow process includes：
S201, the traffic monitoring image for obtaining present frame；
The foreground image of S202, detection traffic monitoring image；
S203, for each pixel in foreground image, judge whether its pixel value is 1, be judged as that "Yes" then flows
Journey to S204, be judged as "No" then flow process to S210；
S204, the pixel value increase for making the pixel of relevant position in prospect counting image；
S205, the pixel value of the pixel of relevant position in auxiliary counting image is set to into 0；
S206, when prospect count image in relevant position pixel pixel value >=first threshold T1 when, by hazard plot
The pixel value of the pixel of relevant position is set to 1 as in；
S2061, judge in the foreground image of traffic monitoring image of present frame, if all of pixel all carried out inspection
Survey, if it is determined that "Yes", then flow process is to S207, if it is determined as no, then flow process is to S2062；
S2062, the next pixel for choosing current pixel, using the next pixel as new current pixel, continue
Detected；
S207, Morphological scale-space is carried out to accident image；
The region being made up of 1 pixel continuous pixel value in S208, identification accident image；
S209, judge whether to terminate, if it is determined as no, then return S201, obtain the friendship of next present frame
Logical monitoring image；
S210, increase the pixel value of the pixel of relevant position in auxiliary counting image；
S211, judge the pixel value of the pixel of relevant position in auxiliary counting image whether more than or equal to Second Threshold
T2, be judged as "Yes" then flow process to S212, be judged as "No" then flow process to S2061；
S212, the pixel that auxiliary counting image, prospect are counted the pixel of relevant position in image and accident image
Value is all set to 0.
In the present embodiment, according to the position of the pixel with predetermined pixel value in the foreground image of traffic monitoring image,
The pixel value of the pixel of relevant position in accident image is set, the position thus, it is possible to detect traffic accident in simple method
Put, and make accident that position to occur and be intuitively embodied on accident image；Also, by carrying out morphology to accident image
Process, testing result can be made more accurate.
Embodiment 2
The embodiment of the present application 2 provides a kind of traffic accident detection means, the traffic accident detection method pair with embodiment 1
Should.Fig. 3 is a composition schematic diagram of the traffic accident detection means of the present embodiment, as shown in figure 3, traffic accident
Detection means 300 can include the first detector unit 301, the first setting unit 302, the and of the second setting unit 303
First processing units 304.
Wherein, the first detector unit 301 is used to detect the foreground image of traffic monitoring image；First setting unit 302
According to the pixel value of pixel in the foreground image, the pixel value that prospect counts the pixel of relevant position in image is set；
Second setting unit 303 counts the pixel value of the pixel of relevant position in image according to the prospect, arranges accident image
The pixel value of the pixel of middle relevant position；First processing units 304 are used to carry out at morphology the accident image
Reason, wherein, the foreground image, the prospect count image and the accident image is all of the same size,
And the foreground image and the accident image are bianry image.
In the present embodiment, the first detector unit 301 is compared each frame of traffic monitoring image and background image
Compared with to obtain foreground image described in each frame.
First setting unit 302 for the pixel in foreground image described in each frame, when the pixel has the first pixel value
When, increase the pixel value that the prospect counts the pixel of relevant position in image, and will be corresponding in auxiliary counting image
The pixel of position is set to have the 3rd pixel value；When the pixel has four pixel values, the auxiliary counting is also made
The pixel value of the pixel of relevant position increases in image, also, the relevant position described in the auxiliary counting image
When the pixel value of pixel is more than or equal to Second Threshold, by the picture of the pixel of the relevant position in the auxiliary counting image
Plain value is set to the 3rd pixel value, the pixel value that the prospect counts the pixel of relevant position in image is set to described
4th pixel value.
Second setting unit 303 pixel value of the pixel of relevant position in the prospect counts image is more than or equal to
During first threshold, the pixel value of the pixel of relevant position in the accident image is set to into the second pixel value；Also, work as
When the pixel value of the pixel of relevant position is more than or equal to Second Threshold in auxiliary counting image, will be corresponding in accident image
The pixel value of the pixel of position is set to the 4th pixel value.
The pixel with second pixel value in 304 pairs of accident images of first processing units is carried out at deletion
Reason and/or expansion process.
The traffic accident detection means 300 can also include recognition unit (not shown), and it is used to recognize through described
In the accident image after Morphological scale-space, by the area that continuously there is the pixel of second pixel value to be constituted
Domain.
In the present embodiment, with regard to accident detection device 300 each unit detailed description, may be referred to embodiment 1
In explanation to corresponding steps, the present embodiment is not repeated explanation.
By the present embodiment, according to the position of the pixel with predetermined pixel value in the foreground image of traffic monitoring image,
The pixel value of the pixel of relevant position in accident image is set, the position thus, it is possible to detect traffic accident in simple method
Put, and make accident that position to occur and be intuitively embodied on accident image；Also, by carrying out morphology to accident image
Process, testing result can be made more accurate.
Embodiment 3
The embodiment of the present application 3 provides a kind of electronic equipment, and the electronic equipment includes：Traffic as described in Example 2
Accident detection device.
Fig. 4 is a composition schematic diagram of the electronic equipment of the embodiment of the present application.As shown in figure 4, electronic equipment 400
Can include：Central processing unit (CPU) 401 and memory 402；Memory 402 is coupled to central processing unit 401.
Wherein the memory 402 can store various data；The program that additionally storage information is processed, and in central processing unit
The program is performed under 401 control.
In one embodiment, the function of traffic accident detection means 300 can be integrated into central processing unit 401
In.Wherein, central processing unit 401 can be configured to be controlled the electronic equipment, to realize embodiment 1
Described traffic accident detection method.
In another embodiment, traffic accident detection means 300 can with the separate configuration of central processing unit 401,
For example traffic accident detection means 300 can be configured to the chip that is connected with central processing unit 401, by centre
The control of reason device 401 is realizing the function of traffic accident detection means 300.
Additionally, as shown in figure 4, electronic equipment 400 can also include：Input-output unit 403 and display unit
404 etc.；Wherein, similarly to the prior art, here is omitted for the function of above-mentioned part.It should be noted that electric
Sub- equipment 400 is also not necessary to include all parts shown in Fig. 4；Additionally, electronic equipment 400 can be with
Including the part being not shown in Fig. 4, prior art is may be referred to.
The embodiment of the present invention also provides a kind of computer-readable program, wherein when performing described program in the electronic device
When, described program causes computer that traffic accident detection side as described in Example 1 is performed in the electronic equipment
Method.
The embodiment of the present invention also provides a kind of storage medium of the computer-readable program that is stored with, wherein the computer can
Reader causes computer to perform traffic accident detection method as described in Example 1 in the electronic device.
Apparatus and method more than of the invention can be realized by hardware, it is also possible to be realized by combination of hardware software.The present invention
It is related to such computer-readable program, when the program is performed by logical block, realizes can the logical block
Devices described above or component parts, or make the logical block realize various methods or step mentioned above.This
The bright storage medium further related to for storing procedure above, such as hard disk, disk, CD, DVD, flash memory.
For one or more combinations of one or more in the function box described in accompanying drawing and/or function box,
Can be implemented as performing the general processor of function described herein, digital signal processor (DSP), specially
With integrated circuit (ASIC), field programmable gate array (FPGA) or other PLDs, discrete
Door either transistor logic, discrete hardware components or it is arbitrarily appropriately combined.For the function side of Description of Drawings
One or more combinations of one or more and/or function box in frame, are also implemented as the combination of computing device,
For example, the combination of DSP and microprocessor, multi-microprocessor communicate one or more the micro- places for combining with DSP
Reason device or any other this configuration.
Above in association with specific embodiment, invention has been described, it will be appreciated by those skilled in the art that this
A little descriptions are all exemplary, are not limiting the scope of the invention.Those skilled in the art can be according to this
The spirit and principle of invention makes various variants and modifications to the present invention, and these variants and modifications are also in the scope of the present invention
It is interior.
With regard to including the embodiment of above example, following note being also disclosed：
Note 1, a kind of traffic accident detection means, for detecting traffic monitoring image in occur traffic accident position,
Including：
First detector unit, it is used to detect the foreground image of traffic monitoring image；
First setting unit,, according to the pixel value of pixel in the foreground image, the prospect that arranges counts corresponding in image for it
The pixel value of the pixel of position；
Second setting unit, it counts the pixel value of the pixel of relevant position in image according to the prospect, arranges accident
The pixel value of the pixel of relevant position in image；
First processing units, it is used to carry out Morphological scale-space to the accident image,
Wherein, the foreground image, the prospect count image and the accident image is all of the same size,
And the foreground image and the accident image are bianry image.
Note 2, the traffic accident detection means as described in note 1, wherein,
First detector unit is compared each frame of traffic monitoring image with background image, to obtain each frame
The foreground image.
Note 3, the traffic accident detection means as described in note 1, wherein,
First setting unit for the pixel in foreground image described in each frame, when the pixel has the first pixel value
When, increase the pixel value that the prospect counts the pixel of relevant position in image.
Note 4, the traffic accident detection means as described in note 3, wherein,
Second setting unit, the pixel value of the pixel of relevant position is more than or equal in the prospect counts image
During first threshold, the pixel value of the pixel of relevant position in the accident image is set to into the second pixel value.
Note 5, the traffic accident detection means as described in note 3, wherein, the traffic accident detection means is also wrapped
Include：
Recognition unit, it is used to recognize in the accident image after the Morphological scale-space, by continuously having
The region that the pixel of second pixel value is constituted.
Note 6, the traffic accident detection means as described in note 1, wherein,
The first processing units to the accident image in the pixel with second pixel value carry out at deletion
Reason and/or expansion process.
Note 7, the traffic accident detection means as described in note 3, wherein,
First setting unit for the pixel in foreground image described in each frame, when the pixel has the 4th pixel value
When, also making the pixel value of the pixel of relevant position in the auxiliary counting image increases,
Also, the pixel value of the pixel of the relevant position described in the auxiliary counting image is more than or equal to Second Threshold
When, the pixel value of the pixel of the relevant position in the auxiliary counting image is set to into the 3rd pixel value, before described
Scape counts the pixel value of the pixel of relevant position in image and is set to the 4th pixel value,
Wherein, the auxiliary counting image counts image and is of the same size with the prospect.
Note 8, the traffic accident detection means as described in note 7, wherein,
The pixel value of the pixel of second setting unit relevant position described in the auxiliary counting image be more than or
During equal to Second Threshold, also the pixel value of the pixel of relevant position in the accident image is set to into the 4th pixel
Value.
Note 9, the traffic accident detection means as described in note 7, wherein,
First setting unit also counts the auxiliary when the pixel in the foreground image has the first pixel value
The pixel of relevant position is set to have the 3rd pixel value in number image.
Note 10, a kind of electronic equipment, the electronic equipment includes the traffic accident as any one of being attached 1-9
Detection means.
Note 11, a kind of traffic accident detection method, for detecting traffic monitoring image in occur traffic accident position
Put, including：
The foreground image of detection traffic monitoring image；
According to the pixel value of pixel in the foreground image, the pixel that prospect counts the pixel of relevant position in image is set
Value；
The pixel value of the pixel of relevant position in image is counted according to the prospect, relevant position in accident image is set
The pixel value of pixel；
Morphological scale-space is carried out to the accident image,
Wherein, the foreground image, the prospect count image and the accident image is all of the same size,
And the foreground image and the accident image are bianry image.
Note 12, the traffic accident detection method as described in note 11, wherein, detect the prospect of traffic monitoring image
Image includes：
Each frame of traffic monitoring image is compared with background image, to obtain foreground image described in each frame.
Note 13, the traffic accident detection method as described in note 11, wherein, according to pixel in the foreground image
Pixel value, setting prospect counts the pixel value of pixel of relevant position in image to be included：
For the pixel in foreground image described in each frame, when the pixel has the first pixel value, the prospect meter is made
The pixel value of the pixel of relevant position increases in number image.
Note 14, the traffic accident detection method as described in note 13, wherein, counted in image according to the prospect
The pixel value of the pixel of relevant position, arranging the pixel value of the pixel of relevant position in accident image includes；
When the pixel value that the prospect counts the pixel of relevant position in image is more than or equal to first threshold, will be described
The pixel value of the pixel of relevant position is set to the second pixel value in accident image.
Note 15, the traffic accident detection method as described in note 13, wherein, the traffic accident detection method is also
Including：
In the accident image of the identification after the Morphological scale-space, by continuously having second pixel value
The region that pixel is constituted.
Note 16, the traffic accident detection method as described in note 11, wherein, form is carried out to the accident image
Process includes：
The pixel with second pixel value in the accident image carries out delete processing and/or expansion process.
Note 17, the traffic accident detection method as described in note 13, wherein, according to pixel in the foreground image
Pixel value, setting prospect counts the pixel value of pixel of relevant position in image also to be included：
For the pixel in foreground image described in each frame, when the pixel has four pixel values, the auxiliary meter is made
The pixel value of the pixel of relevant position increases in number image；
When the pixel value of the pixel of the relevant position described in the auxiliary counting image is more than or equal to Second Threshold, will
The pixel value of the pixel of the relevant position is set to the 3rd pixel value in the auxiliary counting image, and the prospect is counted
The pixel value of the pixel of relevant position is set to the 4th pixel value in image,
Wherein, the auxiliary counting image counts image and is of the same size with the prospect.
Note 18, the traffic accident detection method as described in note 17, wherein, counted in image according to the prospect
The pixel value of the pixel of relevant position, arranging the pixel value of the pixel of relevant position in accident image also includes：
When the pixel value of the pixel of the relevant position described in the auxiliary counting image is more than or equal to Second Threshold, will
The pixel value of the pixel of relevant position is set to the 4th pixel value in the accident image.
Note 19, the traffic accident detection method as described in note 17, wherein, the traffic accident detection method is also
Including：
When the pixel in the foreground image has the first pixel value, by relevant position in the auxiliary counting image
Pixel is set to have the 3rd pixel value.
Claims (10)
1. a kind of traffic accident detection means, for detecting traffic monitoring image in there is the position of traffic accident, bag
Include：
First detector unit, it is used to detect the foreground image of traffic monitoring image；
First setting unit,, according to the pixel value of pixel in the foreground image, the prospect that arranges counts corresponding in image for it
The pixel value of the pixel of position；
Second setting unit, it counts the pixel value of the pixel of relevant position in image according to the prospect, arranges accident
The pixel value of the pixel of relevant position in image；
First processing units, it is used to carry out Morphological scale-space to the accident image,
Wherein, the foreground image, the prospect count image and the accident image is all of the same size,
And the foreground image and the accident image are bianry image.
2. traffic accident detection means as claimed in claim 1, wherein,
First setting unit, for the pixel in foreground image described in each frame, when the pixel has the first pixel
During value, increase the pixel value that the prospect counts the pixel of relevant position in image.
3. traffic accident detection means as claimed in claim 2, wherein,
Second setting unit, the pixel value of the pixel of relevant position is more than or equal in the prospect counts image
During first threshold, the pixel value of the pixel of relevant position in the accident image is set to into the second pixel value.
4. traffic accident detection means as claimed in claim 3, wherein, the traffic accident detection means is also wrapped
Include：
Recognition unit, it is used to recognize in the accident image after the Morphological scale-space, by continuously having
The region that the pixel of second pixel value is constituted.
5. traffic accident detection means as claimed in claim 3, wherein,
The first processing units to the accident image in the pixel with second pixel value carry out at deletion
Reason and/or expansion process.
6. traffic accident detection means as claimed in claim 2, wherein,
First setting unit, for the pixel in foreground image described in each frame, when the pixel has the 4th pixel
During value, also making the pixel value of the pixel of relevant position in auxiliary counting image increases,
Also, the pixel value of the pixel of the relevant position described in the auxiliary counting image is more than or equal to Second Threshold
When, the pixel value of the pixel of the relevant position in the auxiliary counting image is set to into the 3rd pixel value, before described
Scape counts the pixel value of the pixel of relevant position in image and is set to the 4th pixel value,
Wherein, the auxiliary counting image counts image and is of the same size with the prospect.
7. traffic accident detection means as claimed in claim 6, wherein,
Second setting unit, the pixel value of the pixel of the relevant position described in the auxiliary counting image be more than or
During equal to Second Threshold, also the pixel value of the pixel of relevant position in the accident image is set to into the 4th pixel
Value.
8. traffic accident detection means as claimed in claim 6, wherein,
First setting unit, when the pixel in the foreground image has the first pixel value, also by the auxiliary
The pixel for counting relevant position in image is set to the 3rd pixel value.
9. a kind of electronic equipment, the electronic equipment includes the traffic accident as any one of claim 1-8
Detection means.
10. a kind of traffic accident detection method, for detecting traffic monitoring image in there is the position of traffic accident, bag
Include：
The foreground image of detection traffic monitoring image；
According to the pixel value of pixel in the foreground image, the pixel that prospect counts the pixel of relevant position in image is set
Value；
The pixel value of the pixel of relevant position in image is counted according to the prospect, relevant position in accident image is set
The pixel value of pixel；
Morphological scale-space is carried out to the accident image,
Wherein, the foreground image, the prospect count image and the accident image is all of the same size,
And the foreground image and the accident image are bianry image.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510755728.XA CN106682566A (en) | 2015-11-09 | 2015-11-09 | Traffic accident detection method, traffic accident detection device and electronic device |
| JP2016217213A JP2017091530A (en) | 2015-11-09 | 2016-11-07 | Method for detecting traffic accident, traffic accident detector, and electronic apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510755728.XA CN106682566A (en) | 2015-11-09 | 2015-11-09 | Traffic accident detection method, traffic accident detection device and electronic device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106682566A true CN106682566A (en) | 2017-05-17 |
Family
ID=58770720
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510755728.XA Pending CN106682566A (en) | 2015-11-09 | 2015-11-09 | Traffic accident detection method, traffic accident detection device and electronic device |
Country Status (2)
| Country | Link |
|---|---|
| JP (1) | JP2017091530A (en) |
| CN (1) | CN106682566A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738842A (en) * | 2018-07-18 | 2020-01-31 | 阿里巴巴集团控股有限公司 | Accident responsibility division and behavior analysis method, device, equipment and storage medium |
| CN111369807A (en) * | 2020-03-24 | 2020-07-03 | 北京百度网讯科技有限公司 | Traffic accident detection method, device, equipment and medium |
| CN114005074A (en) * | 2021-12-30 | 2022-02-01 | 以萨技术股份有限公司 | Traffic accident determination method and device and electronic equipment |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101231696A (en) * | 2008-01-30 | 2008-07-30 | 安防科技（中国）有限公司 | Method and system for detection of hangover |
| CN102073851A (en) * | 2011-01-13 | 2011-05-25 | 北京科技大学 | Method and system for automatically identifying urban traffic accident |
| CN102081844A (en) * | 2011-01-25 | 2011-06-01 | 华中科技大学 | Traffic video behavior analyzing and alarming server |
| CN102467800A (en) * | 2010-11-05 | 2012-05-23 | 无锡市美网网络信息技术有限公司 | Intrusion detection alarm system |
| CN102163361B (en) * | 2011-05-16 | 2012-10-17 | 公安部沈阳消防研究所 | Image-type fire detection method based on cumulative prospect image |
| CN102902960A (en) * | 2012-09-25 | 2013-01-30 | 武汉烽火众智数字技术有限责任公司 | Leave-behind object detection method based on Gaussian modelling and target contour |
| CN103400113A (en) * | 2013-07-10 | 2013-11-20 | 重庆大学 | Method for detecting pedestrian on expressway or in tunnel based on image processing |
-
2015
- 2015-11-09 CN CN201510755728.XA patent/CN106682566A/en active Pending
-
2016
- 2016-11-07 JP JP2016217213A patent/JP2017091530A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101231696A (en) * | 2008-01-30 | 2008-07-30 | 安防科技（中国）有限公司 | Method and system for detection of hangover |
| CN102467800A (en) * | 2010-11-05 | 2012-05-23 | 无锡市美网网络信息技术有限公司 | Intrusion detection alarm system |
| CN102073851A (en) * | 2011-01-13 | 2011-05-25 | 北京科技大学 | Method and system for automatically identifying urban traffic accident |
| CN102081844A (en) * | 2011-01-25 | 2011-06-01 | 华中科技大学 | Traffic video behavior analyzing and alarming server |
| CN102163361B (en) * | 2011-05-16 | 2012-10-17 | 公安部沈阳消防研究所 | Image-type fire detection method based on cumulative prospect image |
| CN102902960A (en) * | 2012-09-25 | 2013-01-30 | 武汉烽火众智数字技术有限责任公司 | Leave-behind object detection method based on Gaussian modelling and target contour |
| CN103400113A (en) * | 2013-07-10 | 2013-11-20 | 重庆大学 | Method for detecting pedestrian on expressway or in tunnel based on image processing |
Non-Patent Citations (1)
| Title |
|---|
| 张具琴等: "智能视频监控中的遗留物检测技术", 《计算机与现代化》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110738842A (en) * | 2018-07-18 | 2020-01-31 | 阿里巴巴集团控股有限公司 | Accident responsibility division and behavior analysis method, device, equipment and storage medium |
| CN111369807A (en) * | 2020-03-24 | 2020-07-03 | 北京百度网讯科技有限公司 | Traffic accident detection method, device, equipment and medium |
| CN111369807B (en) * | 2020-03-24 | 2022-04-12 | 北京百度网讯科技有限公司 | Traffic accident detection method, device, equipment and medium |
| CN114005074A (en) * | 2021-12-30 | 2022-02-01 | 以萨技术股份有限公司 | Traffic accident determination method and device and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| JP2017091530A (en) | 2017-05-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Wang et al. | Asphalt pavement pothole detection and segmentation based on wavelet energy field | |
| US10068343B2 (en) | Method and apparatus for recognizing moving target | |
| US20190188528A1 (en) | Text detection method and apparatus, and storage medium | |
| WO2016127736A1 (en) | Computing method for area of fingerprint overlapping area and electronic apparatus | |
| CN106920233B (en) | Scratch detection method, apparatus and electronic equipment based on image procossing | |
| US20180129883A1 (en) | Detection method and apparatus of a status of a parking lot and electronic equipment | |
| EP3032462A1 (en) | Method and apparatus for tracking object, and non-transitory computer-readable recording medium | |
| US8811750B2 (en) | Apparatus and method for extracting edge in image | |
| CN103886598A (en) | Tunnel smoke detecting device and method based on video image processing | |
| JP2016058085A (en) | Method and device for detecting shielding of object | |
| CN111127358B (en) | Image processing method, device and storage medium | |
| CN106682566A (en) | Traffic accident detection method, traffic accident detection device and electronic device | |
| Zhang et al. | Counting vehicles in urban traffic scenes using foreground time‐spatial images | |
| CN114627526A (en) | Fusion duplicate removal method and device based on multi-camera snapshot image and readable medium | |
| CN104268595A (en) | General object detecting method and system | |
| CN113762220B (en) | Object recognition method, electronic device, and computer-readable storage medium | |
| CN110796129A (en) | Text line region detection method and device | |
| CN111160274B (en) | Pedestrian detection method based on binaryzation fast RCNN (radar cross-correlation neural network) | |
| CN112069899A (en) | Road shoulder detection method and device and storage medium | |
| CN111932515B (en) | Short circuit detection method and system for product residual defects and defect classification system | |
| CN106991684B (en) | Foreground extracting method and device | |
| CN112950961B (en) | Traffic flow statistical method, device, equipment and storage medium | |
| CN112133100B (en) | Vehicle detection method based on R-CNN | |
| CN115719472A (en) | Complementary optimization method, system, equipment and storage medium based on detection and tracking | |
| CN106682565A (en) | Vehicle counting method, vehicle counting device and electronic device |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170517 |