CN105809120A - Video image analyzing method and system - Google Patents
Video image analyzing method and system Download PDFInfo
- Publication number
- CN105809120A CN105809120A CN201610121037.9A CN201610121037A CN105809120A CN 105809120 A CN105809120 A CN 105809120A CN 201610121037 A CN201610121037 A CN 201610121037A CN 105809120 A CN105809120 A CN 105809120A
- Authority
- CN
- China
- Prior art keywords
- frame
- image
- group
- border
- space
- 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.)
- Withdrawn
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
- 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
- G06F18/23—Clustering techniques
-
- 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
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
-
- 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
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Compression Or Coding Systems Of Tv Signals (AREA)
Abstract
The invention provides a video image analyzing method and a system. The method comprising the following steps: acquiring a traffic video stream; determining the color features of each video frame of the traffic video stream; calculating the frame distances between neighboring frames according to their color features; calculating the differences of clustered image frame groups in RGB space and YUV space according to the image cluster assessing standards applied to the boundaries among frames; and determining the RGB space and YUV space clustered frame groups formed by the last boundaries of the cluster image frames According to the invention, more stable detection results can be achieved in different environments.
Description
Technical field
The present invention relates to the method for technical field of video processing and a kind of for video traffic information detection equipment.
Background technology
In order to realize automatization, intelligent acquisition of road traffic information, a kind of video vehicle detection method based on image processing techniques is to determine at present to adopt from Traffic Surveillance Video traffic flow.At present, a lot of people set up the video detection of traffic flow product of (including the Video Collection System installed on the road surface in the cities such as Beijing) and come out one after another, but the accuracy of detection that these products are in all weather complex environments is largely, rise and fall indefinite, especially auto lamp is incident upon on road surface or halation to occur at night, with the dust rained at severe weather conditions, smog etc., its testing result can not meet the detection of stability requirement.
It is an object of the invention to: the stability of raising testing result in different environments.For the deficiencies in the prior art, the technical solution adopted for the present invention to solve the technical problems is: a kind of video image analysis method, is used for detecting traffic video information, including: obtain traffic video stream；Determine that the image of each frame of color characteristic is at traffic video stream；According to color characteristic；Interframe distance calculating between consecutive frame calculates the group of the picture frame of cluster and adopts the rgb space of image clustering evaluation criterion with image clustering evaluation criterion in yuv space difference according to interframe distance border；It is gathered in the group of picture frame of rgb space and yuv space cluster with the border determining final image and wherein determines that according to the group of boundary frame the image of each frame of color characteristic includes at traffic video stream: the image of each frame of sampling, to obtain the data sampled；Generate histogrammic R, each two field picture of G, B and Y (brightness) Color Channel carries out the data sampled；And the color characteristic of each two field picture is represented by using four-dimensional rectangular histogram as F={HRi, disintegrating machine, HBi, height is joined }, wherein HRi, disintegrating machine, HBi, height is joined, and the R of histogrammic four Color Channels, G, B and Y are respectively, I=1,2 ..., N, N is the Color Channel of each division of numeral, and described calculating includes case according to the interframe distance between the consecutive frame of color characteristic: calculate each Color Channel at rgb space；Interframe distance between consecutive frame finds R, G and the B of correspondence；The interframe distance crossing of three Color Channels obtain the interframe distance of rgb space be determined by minimum and largest frames spacing three Color Channels R, G and B respectively；And just obtain the interframe distance at yuv space by calculating the interframe distance between Y passage consecutive frame.
The method that the present invention further provides, is in that the acquisition of wherein traffic video stream, including: obtain traffic video stream by being continuously shot.
The method that the present invention further provides, also includes: be previously created the data that each two field picture of histogrammic R, G, B and Y Color Channel carries out sampling, the data de-noising of sampling.
The method that the present invention further provides, be in that wherein to calculate the border of image assemble the interframe distance group of frame adopt image clustering to comprise determining that in the evaluation criterion of rgb space the border of group of picture frame of cluster utilizes the distance of interframe with rgb space and default first object function at rgb space.
The method that the present invention further provides, be in that wherein to calculate the border of image assemble the interframe distance group of frame adopt image clustering to comprise determining that in the evaluation criterion of yuv space the group of picture frame of cluster on border utilizes interframe distance at yuv space and the second object function of presetting at yuv space.
The method that the present invention further provides, it is in that wherein to determine that the last border of the group of the picture frame assembled on the border of image is gathered in the group of rgb space and yuv space framework and includes: find the crossing, border of image and be gathered in rgb space the group of frame and the group of the frame in yuv space is assembled on the border of image, to obtain the ultimate bound of the group of the picture frame of cluster.
A kind of video image analysis system, including processor and internal memory and this, the program code of storage and program code detecting device are by processor execution: acquisition traffic video stream；Determine that the image of each frame of color characteristic is at traffic video stream；Determine color feature；Interframe between consecutive frame adopts the rgb space of image clustering evaluation criterion with image clustering evaluation criterion in yuv space difference apart from the group of the picture frame calculating cluster according to interframe distance border；And determine the ultimate bound of the group of picture frame that the group of the picture frame of cluster clusters at rgb space and yuv space according to border, program code therein is performed by performing processor: sample the image of each frame, to obtain the data sampled；Generate histogrammic R, each two field picture of G, B and Y Color Channel carries out the data sampled；And represent that the image of each frame of color characteristic uses four-dimensional rectangular histogram as F={HRi, disintegrating machine, HBi, height is joined }, wherein HRi, disintegrating machine, HBi, height is joined, and the R of histogrammic four Color Channels, G, B and Y are respectively, I=1,2 ..., N, N is the Color Channel of each division of refuse receptacle of numeral, and program code therein is all by processor execution: calculate each Color Channel at rgb space；Interframe distance between consecutive frame finds R, G and the B of correspondence；The interframe distance crossing of three Color Channels determines that R, G and the B of minimum three color channels with largest frames spacing obtain interframe distance respectively at rgb space；With calculating Y Color Channel at yuv space, to obtain the interframe distance between the consecutive frame of the interframe distance of yuv space.
The system that the present invention further provides, is in that program code therein is all by processor execution: the data of the histogrammic sampling of research and development before denoising, generates the Color Channel of the image of each frame of G, B and Y according to sampled data.
The system that the present invention further provides, it is in that program code therein is all by processor execution: the group of the frame in yuv space is assembled on the group finding the crossing, border of image to be gathered in rgb space frame and the border of image, to obtain the ultimate bound of the group of the picture frame of cluster.
The beneficial effects of the present invention is the method and system by providing in the detection to the traffic video information provided, color and the monochrome information of image are fully used, it is considered as to the relation between time series chart picture frame, and the vehicle of the image clustering reflection motor process of each group；Stronger adaptability is had with method and devices in accordance with embodiments of the present invention, the change to environment, remain to reach higher precision, in the detection that the vile weathers such as rain, snow, mist are stable.
Detailed description of the invention
A kind of video image analysis method, is used for detecting traffic video information, including: obtain traffic video stream；Determine that the image of each frame of color characteristic is at traffic video stream；According to color characteristic；Interframe distance calculating between consecutive frame calculates the group of the picture frame of cluster and adopts the rgb space of image clustering evaluation criterion with image clustering evaluation criterion in yuv space difference according to interframe distance border；It is gathered in the group of picture frame of rgb space and yuv space cluster with the border determining final image and wherein determines that according to the group of boundary frame the image of each frame of color characteristic includes at traffic video stream: the image of each frame of sampling, to obtain the data sampled；Generate histogrammic R, each two field picture of G, B and Y (brightness) Color Channel carries out the data sampled；And the color characteristic of each two field picture is represented by using four-dimensional rectangular histogram as F={HRi, disintegrating machine, HBi, height is joined }, wherein HRi, disintegrating machine, HBi, height is joined, and the R of histogrammic four Color Channels, G, B and Y are respectively, I=1,2 ..., N, N is the Color Channel of each division of numeral, and described calculating includes case according to the interframe distance between the consecutive frame of color characteristic: calculate each Color Channel at rgb space；Interframe distance between consecutive frame finds R, G and the B of correspondence；The interframe distance crossing of three Color Channels obtain the interframe distance of rgb space be determined by minimum and largest frames spacing three Color Channels R, G and B respectively；And just obtain the interframe distance at yuv space by calculating the interframe distance between Y passage consecutive frame.
The wherein acquisition of traffic video stream, including: obtain traffic video stream by being continuously shot.
The method that the present invention further provides, also includes: be previously created the data that each two field picture of histogrammic R, G, B and Y Color Channel carries out sampling, the data de-noising of sampling.
Wherein calculate the border of image assemble the interframe distance group of frame adopt image clustering to comprise determining that in the evaluation criterion of rgb space the border of group of picture frame of cluster utilizes the distance of interframe with rgb space and default first object function at rgb space.
Wherein calculate the border of image assemble the interframe distance group of frame adopt image clustering to comprise determining that in the evaluation criterion of yuv space the group of picture frame of cluster on border utilizes interframe distance at yuv space and the second object function of presetting at yuv space.
Wherein determine that the last border of group of the picture frame assembled on the border of image is gathered in the group of rgb space and yuv space framework and includes: find the crossing, border of image and be gathered in rgb space the group of frame and the group of the frame in yuv space is assembled on the border of image, to obtain the ultimate bound of the group of the picture frame of cluster.
A kind of video image analysis system, including processor and internal memory and this, the program code of storage and program code detecting device are by processor execution: acquisition traffic video stream；Determine that the image of each frame of color characteristic is at traffic video stream；Determine color feature；Interframe between consecutive frame adopts the rgb space of image clustering evaluation criterion with image clustering evaluation criterion in yuv space difference apart from the group of the picture frame calculating cluster according to interframe distance border；And determine the ultimate bound of the group of picture frame that the group of the picture frame of cluster clusters at rgb space and yuv space according to border, program code therein is performed by performing processor: sample the image of each frame, to obtain the data sampled；Generate histogrammic R, each two field picture of G, B and Y Color Channel carries out the data sampled；And represent that the image of each frame of color characteristic uses four-dimensional rectangular histogram as F={HRi, disintegrating machine, HBi, height is joined }, wherein HRi, disintegrating machine, HBi, height is joined, and the R of histogrammic four Color Channels, G, B and Y are respectively, I=1,2 ..., N, N is the Color Channel of each division of refuse receptacle of numeral, and program code therein is all by processor execution: calculate each Color Channel at rgb space；Interframe distance between consecutive frame finds R, G and the B of correspondence；The interframe distance crossing of three Color Channels determines that R, G and the B of minimum three color channels with largest frames spacing obtain interframe distance respectively at rgb space；With calculating Y Color Channel at yuv space, to obtain the interframe distance between the consecutive frame of the interframe distance of yuv space.
Program code therein is all by processor execution: the data of the histogrammic sampling of research and development before denoising, generates the Color Channel of the image of each frame of G, B and Y according to sampled data.
Program code therein is all by processor execution: find the border of group that the crossing, border of image is gathered in rgb space frame and image to assemble the group of the frame in yuv space, to obtain the ultimate bound of the group of the picture frame of cluster.
Visible, this instrument is the embodiment traffic video information provided by the invention in detection, makes full use of color and the monochrome information of image, and is considered as the relation between time series chart picture frame, and the group of the cluster of each image reflects the motor process of vehicle.Method and devices in accordance with embodiments of the present invention have stronger adaptability, the change to environment, and remain to reach higher precision, in the detection that the vile weathers such as rain, snow, mist are stable.
Each incremental model being embodied in described in this specification, the same or similar part of embodiment is likely to refer to the other side, and the main content described in each example is all differences from other embodiments.Particularly, the embodiment to this system, because it is the method for substantially similarity, embodies it and be described relatively easy, and correlated parts, refer to the embodiment of the method for the description of part.
Obviously, it should appreciate that the step of those above-mentioned modules or the present invention of being skillful in art is likely to adopt conventional computing equipment.Module or step, can be integrated in a single computing equipment or be distributed in the network of multiple computing equipment composition.Module or step can by performing the executable computing equipment of program code, therefore, module or step may be stored in storage device and the computing equipment that performs or make integrated circuit modules respectively, or, multiple modules or step are made a single integrated circuit modules and are performed.This mode, the present invention is not limited to any specific combination of hardware.
Described above is the simply first-selected embodiment present invention, rather than the restriction present invention.Being skillful in art for those, the present invention is likely to have various carrying out to transform and reconstruct.The principle making any amendment, equivalent substitution, improvement or the class Sihe present invention in spirit should fall into protection scope of the present invention.
Claims (9)
1. a video image analysis method, is used for detecting traffic video information, it is characterised in that including: obtain traffic video stream；Determine that the image of each frame of color characteristic is at traffic video stream；According to color characteristic；Interframe distance calculating between consecutive frame calculates the group of the picture frame of cluster and adopts the rgb space of image clustering evaluation criterion with image clustering evaluation criterion in yuv space difference according to interframe distance border；It is gathered in the group of picture frame of rgb space and yuv space cluster with the border determining final image and wherein determines that according to the group of boundary frame the image of each frame of color characteristic includes at traffic video stream: the image of each frame of sampling, to obtain the data sampled；Generate histogrammic R, each two field picture of G, B and Y (brightness) Color Channel carries out the data sampled；And the color characteristic of each two field picture is represented by using four-dimensional rectangular histogram as F={HRi, disintegrating machine, HBi, height is joined }, wherein HRi, disintegrating machine, HBi, height is joined, and the R of histogrammic four Color Channels, G, B and Y are respectively, I=1,2 ..., N, N is the Color Channel of each division of numeral, and described calculating includes case according to the interframe distance between the consecutive frame of color characteristic: calculate each Color Channel at rgb space；Interframe distance between consecutive frame finds R, G and the B of correspondence；The interframe distance crossing of three Color Channels obtain the interframe distance of rgb space be determined by minimum and largest frames spacing three Color Channels R, G and B respectively；And just obtain the interframe distance at yuv space by calculating the interframe distance between Y passage consecutive frame.
2. method according to claim 1, it is characterised in that the wherein acquisition of traffic video stream, including: obtain traffic video stream by being continuously shot.
3. method according to claim 1, it is characterised in that also include: be previously created the data that each two field picture of histogrammic R, G, B and Y Color Channel carries out sampling, the data de-noising of sampling.
4. method according to claim 1, it is characterized in that, wherein calculate the border of image assemble the interframe distance group of frame adopt image clustering to comprise determining that in the evaluation criterion of rgb space the border of group of picture frame of cluster utilizes the distance of interframe with rgb space and default first object function at rgb space.
5. method according to claim 1, it is characterized in that, wherein calculate the border of image assemble the interframe distance group of frame adopt image clustering to comprise determining that in the evaluation criterion of yuv space the group of picture frame of cluster on border utilizes interframe distance at yuv space and the second object function of presetting at yuv space.
6. method according to claim 5, it is characterized in that, wherein determine that the last border of group of the picture frame assembled on the border of image is gathered in the group of rgb space and yuv space framework and includes: find the crossing, border of image and be gathered in rgb space the group of frame and the group of the frame in yuv space is assembled on the border of image, to obtain the ultimate bound of the group of the picture frame of cluster.
7. a video image analysis system, it is characterised in that include processor and internal memory and this, the program code of storage and program code detecting device are by processor execution: obtain traffic video stream；Determine that the image of each frame of color characteristic is at traffic video stream；Determine color feature；Interframe between consecutive frame adopts the rgb space of image clustering evaluation criterion with image clustering evaluation criterion in yuv space difference apart from the group of the picture frame calculating cluster according to interframe distance border；And determine the ultimate bound of the group of picture frame that the group of the picture frame of cluster clusters at rgb space and yuv space according to border, program code therein is performed by performing processor: sample the image of each frame, to obtain the data sampled；Generate histogrammic R, each two field picture of G, B and Y Color Channel carries out the data sampled；And represent that the image of each frame of color characteristic uses four-dimensional rectangular histogram as F={HRi, disintegrating machine, HBi, height is joined }, wherein HRi, disintegrating machine, HBi, height is joined, and the R of histogrammic four Color Channels, G, B and Y are respectively, I=1,2 ..., N, N is the Color Channel of each division of refuse receptacle of numeral, and program code therein is all by processor execution: calculate each Color Channel at rgb space；Interframe distance between consecutive frame finds R, G and the B of correspondence；The interframe distance crossing of three Color Channels determines that R, G and the B of minimum three color channels with largest frames spacing obtain interframe distance respectively at rgb space；With calculating Y Color Channel at yuv space, to obtain the interframe distance between the consecutive frame of the interframe distance of yuv space.
8. system according to claim 7, it is characterised in that program code therein is all by processor execution: the data of the histogrammic sampling of research and development before denoising, generates the Color Channel of the image of each frame of G, B and Y according to sampled data.
9. system according to claim 7, it is characterized in that, program code therein is all by processor execution: find the border of group that the crossing, border of image is gathered in rgb space frame and image to assemble the group of the frame in yuv space, to obtain the ultimate bound of the group of the picture frame of cluster.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610121037.9A CN105809120A (en) | 2016-03-03 | 2016-03-03 | Video image analyzing method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610121037.9A CN105809120A (en) | 2016-03-03 | 2016-03-03 | Video image analyzing method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN105809120A true CN105809120A (en) | 2016-07-27 |
Family
ID=56466093
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610121037.9A Withdrawn CN105809120A (en) | 2016-03-03 | 2016-03-03 | Video image analyzing method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105809120A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108573223A (en) * | 2018-04-03 | 2018-09-25 | 同济大学 | A kind of EMU operating environment cognitive method based on bow net video |
| CN111488772A (en) * | 2019-01-29 | 2020-08-04 | 杭州海康威视数字技术股份有限公司 | Method and apparatus for smoke detection |
-
2016
- 2016-03-03 CN CN201610121037.9A patent/CN105809120A/en not_active Withdrawn
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108573223A (en) * | 2018-04-03 | 2018-09-25 | 同济大学 | A kind of EMU operating environment cognitive method based on bow net video |
| CN108573223B (en) * | 2018-04-03 | 2021-11-23 | 同济大学 | Motor train unit operation environment sensing method based on pantograph-catenary video |
| CN111488772A (en) * | 2019-01-29 | 2020-08-04 | 杭州海康威视数字技术股份有限公司 | Method and apparatus for smoke detection |
| CN111488772B (en) * | 2019-01-29 | 2023-09-22 | 杭州海康威视数字技术股份有限公司 | Method and device for detecting smoke |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Boult et al. | Into the woods: Visual surveillance of noncooperative and camouflaged targets in complex outdoor settings | |
| CN104303193B (en) | Target classification based on cluster | |
| CN102307274B (en) | Motion detection method based on edge detection and frame difference | |
| CN105046218B (en) | A kind of multiple features traffic video smog detection method based on serial parallel processing | |
| CN103646257B (en) | A kind of pedestrian detection and method of counting based on video monitoring image | |
| WO2022027931A1 (en) | Video image-based foreground detection method for vehicle in motion | |
| CN104050481B (en) | Multi-template infrared image real-time pedestrian detection method combining contour feature and gray level | |
| CN106128121B (en) | Vehicle queue length fast algorithm of detecting based on Local Features Analysis | |
| CN105160691A (en) | Color histogram based vehicle body color identification method | |
| WO2012100522A1 (en) | Ptz video visibility detection method based on luminance characteristic | |
| CN102881160B (en) | Outdoor traffic sign identification method under low-illumination scene | |
| CN109446917A (en) | A kind of vanishing Point Detection Method method based on cascade Hough transform | |
| CN105160649A (en) | Multi-target tracking method and system based on kernel function unsupervised clustering | |
| CN111291587A (en) | Pedestrian detection method based on dense crowd, storage medium and processor | |
| CN103679704A (en) | Video motion shadow detecting method based on lighting compensation | |
| CN110032946A (en) | A kind of aluminium based on machine vision/aluminium blister package tablet identification and localization method | |
| CN110659546A (en) | Illegal booth detection method and device | |
| CN110276318A (en) | Nighttime road rains recognition methods, device, computer equipment and storage medium | |
| Nizar et al. | Multi-object tracking and detection system based on feature detection of the intelligent transportation system | |
| CN104778723A (en) | Method for performing motion detection on infrared image with three-frame difference method | |
| CN105809120A (en) | Video image analyzing method and system | |
| Li et al. | Intelligent transportation video tracking technology based on computer and image processing technology | |
| CN112115778B (en) | Intelligent lane line identification method under ring simulation condition | |
| CN113947744A (en) | Fire image detection method, system, equipment and storage medium based on video | |
| CN116189116B (en) | Traffic state sensing method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20160727 |