CN107169984A - A kind of underbody shadow detection method - Google Patents
A kind of underbody shadow detection method Download PDFInfo
- Publication number
- CN107169984A CN107169984A CN201710327852.5A CN201710327852A CN107169984A CN 107169984 A CN107169984 A CN 107169984A CN 201710327852 A CN201710327852 A CN 201710327852A CN 107169984 A CN107169984 A CN 107169984A
- Authority
- CN
- China
- Prior art keywords
- underbody
- shadow
- composition
- detection method
- value
- 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- Withdrawn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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
-
- 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
- G06T2207/30248—Vehicle exterior or interior
- G06T2207/30252—Vehicle exterior; Vicinity of vehicle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of underbody shadow detection method, comprise the following steps：1）Area-of-interest is extracted to the original-gray image collected；2）Underbody shadow thresholds are obtained using the method for clustering, and then obtain Shadow segmentation image；3）By the image zooming-out and the intersection on road surface after Shadow segmentation；4）Vehicle candidate region is efficiently generated using the intersection position of extraction.The present invention proposes a kind of moving vehicle underbody shadow detection method based on clustering, and this method can effectively exclude the interference of different periods on daytime and periphery varying environment shade, underbody shade is accurately detected in real time, orient the position of vehicle.
Description
Technical field
Present invention relates particularly to a kind of underbody shadow detection method.
Background technology
Shadow Detection technology is broadly divided into method based on model and the method for feature based.Method based on model needs
It is strict it is assumed that the change of various environment can not be advantageously used, but be easy to model when interesting target, project have it is different
Effect well can be reached during orientation.The method of feature based quickly can realize and run, but to noise-sensitive, it is low full
It is not high with the lower verification and measurement ratio of degree.
The content of the invention
The technical problem to be solved in the present invention is to provide a kind of underbody shadow detection method.
A kind of underbody shadow detection method, comprises the following steps：
1）Area-of-interest is extracted to the original-gray image collected；
2）Underbody shadow thresholds are obtained using the method for clustering, and then obtain Shadow segmentation image；
3）By the image zooming-out and the intersection on road surface after Shadow segmentation；
4）Vehicle candidate region is efficiently generated using the intersection position of extraction.
Further, step 2）It is middle to assume to comprise only four class targets in extraction area-of-interest：Road surface, lane line, vehicle
With the shade of underbody；Using mixed Gauss model to each composition（That is image object）Clustered, utilize improved EM algorithms
Maximum likelihood value is solved, shadow thresholds are determined by the parameter of shadow model.
Further, step 2）Mixed Gauss model it is as follows：
IfFor stochastic variable X n random sample value, then the probability of mixed Gauss model
Density function is expressed as：
,
In formula, K is expressed as in fraction, value K=4, i.e. image contained target sum,, represent in gauss hybrid models
The proportionality coefficient of each composition, and meet,WithThe average and standard deviation of respectively the i-th composition；
If,Be respectively the probability of each composition for missing data, i.e. random sample value, then institute
There is sample data Joint probability density function be expressed as：
,
In formula,Value is 0 or 1, when, represent sampleBelong to k-th of composition, be not otherwise；
Above-mentioned variable is calculated using maximum likelihood function, expression formula is as follows：
。
Further, step 2）Underbody shadow thresholds computational methods it is as follows：
Image intensity value is clustered, the average and variance of each composition is calculated, the average after sequence is expressed as；
；
The average of underbody shadow region is represented,For its corresponding standard deviation.
The beneficial effects of the invention are as follows：
The present invention proposes a kind of moving vehicle underbody shadow detection method based on clustering, and this method can be excluded effectively
Daytime different periods and periphery varying environment shade interference, underbody shade is accurately detected in real time, the position of vehicle is oriented
Put.
Embodiment
The present invention is further elaborated for specific examples below, but not as a limitation of the invention.
A kind of underbody shadow detection method, comprises the following steps：
1）Area-of-interest is extracted to the original-gray image collected；
2）Underbody shadow thresholds are obtained using the method for clustering, and then obtain Shadow segmentation image；
3）By the image zooming-out and the intersection on road surface after Shadow segmentation；
4）Vehicle candidate region is efficiently generated using the intersection position of extraction.
Step 2）It is middle to assume to comprise only four class targets in extraction area-of-interest：Road surface, lane line, the moon of vehicle and underbody
Shadow；Using mixed Gauss model to each composition（That is image object）Clustered, it is maximum seemingly using improved EM Algorithm for Solving
So value, shadow thresholds are determined by the parameter of shadow model.
Step 2）Mixed Gauss model it is as follows：
IfFor stochastic variable X n random sample value, then the probability of mixed Gauss model
Density function is expressed as：
,
In formula, K is expressed as in fraction, value K=4, i.e. image contained target sum,, represent in gauss hybrid models
The proportionality coefficient of each composition, and meet,WithThe average and standard deviation of respectively the i-th composition；
If,Be respectively the probability of each composition for missing data, i.e. random sample value, then institute
There is sample data Joint probability density function be expressed as：
,
In formula,Value is 0 or 1, when, represent sampleBelong to k-th of composition, be not otherwise；
Above-mentioned variable is calculated using maximum likelihood function, expression formula is as follows：
。
Step 2）Underbody shadow thresholds computational methods it is as follows：
Image intensity value is clustered, the average and variance of each composition is calculated, the average after sequence is expressed as；
；
The average of underbody shadow region is represented,For its corresponding standard deviation.
Claims (4)
1. a kind of underbody shadow detection method, it is characterised in that comprise the following steps：
1）Area-of-interest is extracted to the original-gray image collected；
2）Underbody shadow thresholds are obtained using the method for clustering, and then obtain Shadow segmentation image；
3）By the image zooming-out and the intersection on road surface after Shadow segmentation；
4）Vehicle candidate region is efficiently generated using the intersection position of extraction.
2. underbody shadow detection method according to claim 1, it is characterised in that step 2）It is middle to assume to extract region of interest
Four class targets are comprised only in domain：Road surface, lane line, the shade of vehicle and underbody；Using mixed Gauss model to each composition（I.e.
Image object）Clustered, using improved EM Algorithm for Solving maximum likelihood value, shade threshold is determined by the parameter of shadow model
Value.
3. underbody shadow detection method according to claim 1, it is characterised in that step 2）Mixed Gauss model such as
Under：
IfFor stochastic variable X n random sample value, then the probability of mixed Gauss model is close
Spending function representation is：
,
In formula, K is expressed as in fraction, value K=4, i.e. image contained target sum,, represent in gauss hybrid models
The proportionality coefficient of each composition, and meet,WithThe average and standard deviation of respectively the i-th composition；
If,It is respectively the probability of each composition for missing data, i.e. random sample value, then owns
Sample data Joint probability density function be expressed as：
,
In formula,Value is 0 or 1, when, represent sampleBelong to k-th of composition, be not otherwise；
Above-mentioned variable is calculated using maximum likelihood function, expression formula is as follows：
。
4. underbody shadow detection method according to claim 1, it is characterised in that step 2）Underbody shadow thresholds meter
Calculation method is as follows：
Image intensity value is clustered, the average and variance of each composition is calculated, the average after sequence is expressed as；
；
The average of underbody shadow region is represented,For its corresponding standard deviation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710327852.5A CN107169984A (en) | 2017-05-11 | 2017-05-11 | A kind of underbody shadow detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710327852.5A CN107169984A (en) | 2017-05-11 | 2017-05-11 | A kind of underbody shadow detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107169984A true CN107169984A (en) | 2017-09-15 |
Family
ID=59814853
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710327852.5A Withdrawn CN107169984A (en) | 2017-05-11 | 2017-05-11 | A kind of underbody shadow detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107169984A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110321828A (en) * | 2019-06-27 | 2019-10-11 | 四川大学 | A kind of front vehicles detection method based on binocular camera and vehicle bottom shade |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103544487A (en) * | 2013-11-01 | 2014-01-29 | 扬州瑞控汽车电子有限公司 | Front car identification method based on monocular vision |
| CN106548135A (en) * | 2016-10-17 | 2017-03-29 | 北海益生源农贸有限责任公司 | A kind of road barrier detection method |
-
2017
- 2017-05-11 CN CN201710327852.5A patent/CN107169984A/en not_active Withdrawn
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103544487A (en) * | 2013-11-01 | 2014-01-29 | 扬州瑞控汽车电子有限公司 | Front car identification method based on monocular vision |
| CN106548135A (en) * | 2016-10-17 | 2017-03-29 | 北海益生源农贸有限责任公司 | A kind of road barrier detection method |
Non-Patent Citations (1)
| Title |
|---|
| 任薇 等: ""面向车辆防撞的车底阴影检测方法"", 《计算机工程与设计》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110321828A (en) * | 2019-06-27 | 2019-10-11 | 四川大学 | A kind of front vehicles detection method based on binocular camera and vehicle bottom shade |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106875424B (en) | A kind of urban environment driving vehicle Activity recognition method based on machine vision | |
| CN104658011B (en) | A kind of intelligent transportation moving object detection tracking | |
| Sun et al. | A novel supervised level set method for non-rigid object tracking | |
| CN107742306B (en) | Moving target tracking algorithm in intelligent vision | |
| CN108805149A (en) | A kind of winding detection method and device of visual synchronization positioning and map structuring | |
| CN107315998A (en) | Vehicle class division method and system based on lane line | |
| Klein et al. | Boosting scalable gradient features for adaptive real-time tracking | |
| Yiyang et al. | Individual pig object detection algorithm based on Gaussian mixture model | |
| Kim et al. | Smoke detection using GMM and adaboost | |
| CN103985139B (en) | Particle filter target tracking method based on color model and prediction vector cluster model information fusion | |
| Widyantara et al. | Gamma correction-based image enhancement and canny edge detection for shoreline extraction from coastal imagery | |
| Zeng et al. | Aircraft recognition based on improved iterative threshold selection and skeleton Zernike moment | |
| CN107169984A (en) | A kind of underbody shadow detection method | |
| CN104008374B (en) | Miner's detection method based on condition random field in a kind of mine image | |
| CN107133598A (en) | A kind of detection method of unstructured road | |
| Elashry et al. | Feature matching enhancement using the graph neural network (gnn-ransac) | |
| CN104778443A (en) | Natural road sign recognition based on real-time local characteristic description | |
| Shu et al. | Real time target tracking scale adaptive based on LBP operator and nonlinear meanshift | |
| CN110660081B (en) | Target tracking method based on self-adaptive feature selection | |
| Tan et al. | An improved ORB-GMS image feature extraction and matching algorithm | |
| CN106815862B (en) | Target tracking method based on convolution contour features | |
| Zhai et al. | A moving target detection algorithm based on combination of GMM and LBP texture pattern | |
| Li et al. | Research on two-dimensional entropy threshold method based on improved genetic algorithm | |
| Moreira et al. | Tracking targets in sea surface with the WiSARD weightless neural network | |
| Wang et al. | Combining color histogram and ORB features for robust visual tracking |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20170915 |