CN106407895A - Vehicle shadow detection algorithm based on image gray and Lab color space - Google Patents
Vehicle shadow detection algorithm based on image gray and Lab color space Download PDFInfo
- Publication number
- CN106407895A CN106407895A CN201610750367.4A CN201610750367A CN106407895A CN 106407895 A CN106407895 A CN 106407895A CN 201610750367 A CN201610750367 A CN 201610750367A CN 106407895 A CN106407895 A CN 106407895A
- Authority
- CN
- China
- Prior art keywords
- image
- shadow
- detection
- algorithm
- carried out
- 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
- G06V20/50—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
- 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
- G06V10/60—Extraction of image or video features relating to illumination properties, e.g. using a reflectance or lighting model
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a vehicle shadow detection algorithm based on image gray and Lab color space. The algorithm comprises steps that (i), external image input is carried out; (ii), Gauss filtering noise reduction is carried out; (iii), Canny edge detection and edge projection integration are carried out; (iv), RGB2Lab color space transformation is carried out; (v) Otsu image segmentation is carried out; and (vi), images after shadow distinguishing are outputted. Through the algorithm, influence of brightness and colors of a vehicle body on shadow detection is eliminated, gray of shadow and non-shadow areas and color characteristics are well utilized, staggered type shadows can be relatively precisely and accurately detected, and the algorithm is applicable to wide scenes and especially for occasions with relatively large color temperature change.
Description
Technical field
The invention belongs to a kind of image detection and recognizer are and in particular to a kind of be based on gradation of image and Lab color sky
Between vehicle shadow detection algorithm.
Background technology
Detection for moving vehicle is a very important project in intelligent transportation system at present, then ubiquitous
Vehicle shadow motion detection can be caused with ratio large effect, in order to preferably carry out vehicle object positioning, tracking, right
Detection in shade is particularly important with exclusion.
Content of the invention
The present invention, in order to overcome shortcoming present in prior art to propose, its objective is to provide one kind to be based on image
The vehicle shadow detection algorithm of gray scale and Lab color space.
The technical scheme is that：
A kind of vehicle shadow detection algorithm based on gradation of image and Lab color space, described algorithm comprises the following steps：
（ⅰ）External image inputs；
（ⅱ）Gaussian filtering noise reduction
Remove the noise that image causes because of circuit or other image processing algorithms；
（ⅲ）Canny rim detection and edge projection integration
Rim detection binaryzation are carried out to whole detection zone, then obtains projecting integral's value G_ of detection zone horizontal direction
Add, when certain row integration G_add meets：G_add>Then it is assumed that this is classified as vehicle body column during G_Th；
（ⅳ）RGB2Lab colour space transformation
In CIE 1976 L*a*b, b value represents the blue change to yellow, in the image of 8, the scope selection of b value [255,
0]；
（ⅴ）Otsu Da-Jin algorithm image segmentation
RGB image is asked with its gray-scale map, does the Otsu segmentation of gray scale, on image after singulation, make worthwhile for b gray value and do again
Otsu segmentation, the bright part of the image being partitioned into is non-hatched area, and dark part is shadow region；
（ⅵ）Output area second movie queen image.
The invention has the beneficial effects as follows：
Present invention eliminates the brightness of vehicle itself and color are for the impact of shadow Detection；Make use of shade and non-the moon well
The gray scale in shadow zone domain, the feature of color, being capable of the more careful shade accurately detecting staggered type；Applicable scene is extensive,
Especially prominent under the scene that colour temperature changes greatly.
Brief description
Fig. 1 is the schematic flow sheet based on the vehicle shadow detection algorithm of gradation of image and Lab color space for the present invention.
Specific embodiment
With reference to Figure of description and embodiment to the present invention vehicle shadow based on gradation of image and Lab color space
Detection algorithm is described in detail：
As shown in figure 1, a kind of vehicle shadow detection algorithm based on gradation of image and Lab color space, described algorithm include with
Lower step：
（ⅰ）External image inputs；
（ⅱ）Gaussian filtering noise reduction
Remove the noise that image causes because of circuit or other image processing algorithms；Prepare for doing Canny rim detection below, protect
The accuracy at card detection edge.
（ⅲ）Canny rim detection and edge projection integration
Being Canny is to detect vehicle body position in object detection area.For in traffic scene, the side of vehicle itself
Edge information is abundanter, and for pavement particles point, vehicle body edge gradient is much larger than ground.Using this feature, to whole
Detection zone carries out rim detection binaryzation, then obtains projecting integral's value G_add of detection zone horizontal direction, when certain row is long-pending
G_add is divided to meet：G_add>Then it is assumed that this is classified as vehicle body column during G_Th.Profit finds out vehicle body position in this way, so that
The exclusion vehicle body brightness and color interference to shadow Detection.
（ⅳ）RGB2Lab colour space transformation
In CIE 1976 L*a*b, b value represents the blue change to yellow, in the image of 8, the scope selection of b value [127 ,-
128], [255,0] are typically got for convenience of calculating.In real road light environment, shade and non-hatched area are often in b value
Size on have certain difference, certain L value difference is different to be also apparent from.
（ⅴ）Otsu Da-Jin algorithm image segmentation based on gradation of image and Lab color space b value
During this, it will RGB image is asked with its gray-scale map.Then done to eliminating vehicle body using image segmentation algorithm Otsu
The gray-scale map disturbed carries out the segmentation of foreground and background.The bright part of the image that is partitioned into is non-hatched area, and dark part is the moon
Shadow zone domain.But because Otsu is based only on gray scale using the algorithm of intensity slicing image, do not utilize shadow character, often go out
The effect come is unsatisfactory, and improved method is here：First do the Otsu segmentation of gray scale, on image after singulation, b is worthwhile
Try again Otsu segmentation, in the image so obtaining, shade and non-shadow that for leaf, such scenery is formed as gray value
Interlaced area can have good differentiation.
（ⅵ）Output area second movie queen image.
Otsu is the foreground and background partitioning algorithm for gray level image, and it can utilize the half-tone information automatic seeking of image
Find suitable segmentation threshold, for image binaryzation.In this algorithm, shade and non-shadow are just suitable for, they are in gray scale
On differ greatly, be well suited for this automatic algorithms.The present invention has done 2 points of process for the existing defect of Otsu：
（ⅰ）Shade be accompanied by object generation, during the shadow Detection of vehicle, mostly all can by vehicle itself enrich color with
Brightness is disturbed, and causes to be difficult to by Shadow segmentation out in single gray scale, so this algorithm Canny rim detection, using car
Edge gradient enriches this information, vehicle is screened, preferably to carry out shadow Detection.
（ⅱ）Some shades are mottled staggered, and this shade and non-shadow are very gentle in transition interval, often
It also is difficult to make a distinction according to gray scale, the image being partitioned into all can be with a point hot spot, and not including of segmentation is thorough.So this algorithm
Make use of shade and unshaded color characteristic, under Lab color space, under outdoor scene, shade and non-hatched area are in b value
Can there is difference, and this difference exactly we are utilized, do Otsu segmentation using b value and can be very good leaf, tree
The staggered Shadow segmentation that this scenery of branch is formed comes.
Present invention eliminates the brightness of vehicle itself and color are for the impact of shadow Detection；Make use of well shade and
The gray scale of non-hatched area, the feature of color, being capable of the more careful shade accurately detecting staggered type；Applicable scene is wide
General, especially prominent under the scene that colour temperature changes greatly.
Claims (1)
1. a kind of vehicle shadow detection algorithm based on gradation of image and Lab color space it is characterised in that：Described algorithm includes
Following steps：
（ⅰ）External image inputs；
（ⅱ）Gaussian filtering noise reduction
Remove the noise that image causes because of circuit or other image processing algorithms；
（ⅲ）Canny rim detection and edge projection integration
Rim detection binaryzation are carried out to whole detection zone, then obtains projecting integral's value G_ of detection zone horizontal direction
Add, when certain row integration G_add meets：G_add>Then it is assumed that this is classified as vehicle body column during G_Th；
（ⅳ）RGB2Lab colour space transformation
In CIE 1976 L*a*b, b value represents the blue change to yellow, in the image of 8, the scope selection of b value [255,
0]；
（ⅴ）Otsu Da-Jin algorithm image segmentation
RGB image is asked with its gray-scale map, does the Otsu segmentation of gray scale, on image after singulation, make worthwhile for b gray value and do again
Otsu segmentation, the bright part of the image being partitioned into is non-hatched area, and dark part is shadow region；
（ⅵ）Output area second movie queen image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610750367.4A CN106407895A (en) | 2016-08-30 | 2016-08-30 | Vehicle shadow detection algorithm based on image gray and Lab color space |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610750367.4A CN106407895A (en) | 2016-08-30 | 2016-08-30 | Vehicle shadow detection algorithm based on image gray and Lab color space |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106407895A true CN106407895A (en) | 2017-02-15 |
Family
ID=58002496
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610750367.4A Pending CN106407895A (en) | 2016-08-30 | 2016-08-30 | Vehicle shadow detection algorithm based on image gray and Lab color space |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106407895A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107220981A (en) * | 2017-06-22 | 2017-09-29 | 深圳怡化电脑股份有限公司 | Character segmentation method, device, equipment and storage medium |
| CN108510450A (en) * | 2018-02-07 | 2018-09-07 | 北京农业信息技术研究中心 | A kind of photo-irradiation treatment method and device of crop leaf image |
| CN109544508A (en) * | 2018-10-22 | 2019-03-29 | 塔特工业科技(珠海)有限公司 | A kind of inspiration piece appearance detecting method |
| CN110569840A (en) * | 2019-08-13 | 2019-12-13 | 浙江大华技术股份有限公司 | Target detection method and related device |
| CN113763410A (en) * | 2021-09-30 | 2021-12-07 | 江苏天汇空间信息研究院有限公司 | Image shadow detection method based on HIS combined with spectral feature detection condition |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509098A (en) * | 2011-10-08 | 2012-06-20 | 天津大学 | Fisheye image vehicle identification method |
| CN102622763A (en) * | 2012-02-21 | 2012-08-01 | 芮挺 | Method for detecting and eliminating shadow |
| CN105389556A (en) * | 2015-11-10 | 2016-03-09 | 中南大学 | High-resolution-remote-sensing-image vehicle detection method considering shadow region |
| CN105701844A (en) * | 2016-01-15 | 2016-06-22 | 苏州大学 | Method for detecting obstacle or shadow on the basis of color characteristics |
-
2016
- 2016-08-30 CN CN201610750367.4A patent/CN106407895A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102509098A (en) * | 2011-10-08 | 2012-06-20 | 天津大学 | Fisheye image vehicle identification method |
| CN102622763A (en) * | 2012-02-21 | 2012-08-01 | 芮挺 | Method for detecting and eliminating shadow |
| CN105389556A (en) * | 2015-11-10 | 2016-03-09 | 中南大学 | High-resolution-remote-sensing-image vehicle detection method considering shadow region |
| CN105701844A (en) * | 2016-01-15 | 2016-06-22 | 苏州大学 | Method for detecting obstacle or shadow on the basis of color characteristics |
Non-Patent Citations (3)
| Title |
|---|
| LIANG XU.ETC: "Adaptive Threshold Shadow Detection Based On Image Block Statistics", 《2008 9TH INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING》 * |
| 刘琼等: "基于Lab和YUV颜色空间的农田图像分割方法", 《国外电子测量技术》 * |
| 孙庆: "道路视频车辆检测研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107220981A (en) * | 2017-06-22 | 2017-09-29 | 深圳怡化电脑股份有限公司 | Character segmentation method, device, equipment and storage medium |
| CN108510450A (en) * | 2018-02-07 | 2018-09-07 | 北京农业信息技术研究中心 | A kind of photo-irradiation treatment method and device of crop leaf image |
| CN108510450B (en) * | 2018-02-07 | 2020-06-09 | 北京农业信息技术研究中心 | Illumination processing method and device for crop leaf image |
| CN109544508A (en) * | 2018-10-22 | 2019-03-29 | 塔特工业科技(珠海)有限公司 | A kind of inspiration piece appearance detecting method |
| CN110569840A (en) * | 2019-08-13 | 2019-12-13 | 浙江大华技术股份有限公司 | Target detection method and related device |
| CN113763410A (en) * | 2021-09-30 | 2021-12-07 | 江苏天汇空间信息研究院有限公司 | Image shadow detection method based on HIS combined with spectral feature detection condition |
| CN113763410B (en) * | 2021-09-30 | 2022-08-02 | 江苏天汇空间信息研究院有限公司 | Image shadow detection method based on HIS combined with spectral feature detection condition |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106407895A (en) | Vehicle shadow detection algorithm based on image gray and Lab color space | |
| O'Malley et al. | Rear-lamp vehicle detection and tracking in low-exposure color video for night conditions | |
| US8036427B2 (en) | Vehicle and road sign recognition device | |
| CN102385753B (en) | Illumination-classification-based adaptive image segmentation method | |
| CN112036254B (en) | Moving vehicle foreground detection method based on video image | |
| CN104899554A (en) | Vehicle ranging method based on monocular vision | |
| CN104715252A (en) | License plate character segmentation method with combination of dynamic template and pixel points | |
| Halmaoui et al. | Contrast restoration of road images taken in foggy weather | |
| CN103500451B (en) | A kind of independent floating ice extracting method for satellite data | |
| CN106815587A (en) | Image processing method and device | |
| Jau et al. | A comparison of RGB and HSI color segmentation in real-time video images: A preliminary study on road sign detection | |
| KR101026778B1 (en) | Vehicle image detection apparatus | |
| JP2002083301A (en) | Traffic monitoring device | |
| Alami et al. | Local fog detection based on saturation and RGB-correlation | |
| Fleyeh | Road and traffic sign color detection and segmentation-a fuzzy approach | |
| CN107016343A (en) | A kind of traffic lights method for quickly identifying based on Bel's format-pattern | |
| CN103065145A (en) | Vehicle movement shadow eliminating method | |
| US10417518B2 (en) | Vehicle camera system | |
| CN107066929B (en) | Hierarchical recognition method for parking events of expressway tunnel integrating multiple characteristics | |
| CN109800693B (en) | Night vehicle detection method based on color channel mixing characteristics | |
| Nagumo et al. | Extraction of forward vehicles by front-mounted camera using brightness information | |
| Kawarabuki et al. | Snowfall detection in a foggy scene | |
| Xiang et al. | Research on target detection method based on HSV fusion Gaussian mixture model | |
| CN113537147B (en) | Night lane line detection method based on illumination compensation | |
| Ha et al. | Robust reflection detection and removal in rainy conditions using LAB and HSV color spaces |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170215 |
|
| RJ01 | Rejection of invention patent application after publication |