CN105118032B - A kind of wide method for dynamically processing of view-based access control model system - Google Patents
A kind of wide method for dynamically processing of view-based access control model system Download PDFInfo
- Publication number
- CN105118032B CN105118032B CN201510510485.3A CN201510510485A CN105118032B CN 105118032 B CN105118032 B CN 105118032B CN 201510510485 A CN201510510485 A CN 201510510485A CN 105118032 B CN105118032 B CN 105118032B
- Authority
- CN
- China
- Prior art keywords
- formula
- image
- wide dynamic
- optical signal
- environmental factor
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Abstract
The invention discloses a kind of wide method for dynamically processing of view-based access control model system.The single image of special scenes is obtained first, then it is transformed into YUV color spaces from RGB color, obtain luminance information Y-component therein, the environmental factor of each pixel is calculated using the neighborhood of pixels information of Y-component, according to human visual system's feature, original RGB image is converted using the wide dynamic mapping formula that environmental factor obtains, obtains wide dynamic RGB image.The wide dynamic approach of view-based access control model system proposed by the present invention preferably retains highlight regions and the detailed information of dark area, and obtained wide dynamic images quality is preferable and resource consumption is few, can be good at meeting the needs of handling in real time.
Description
Technical field：
The invention belongs to image processing fields, refer in particular to a kind of wide method for dynamically processing of view-based access control model system.
Background technology：
Wide dynamic technique is a kind of technology for allowing video camera to see image clearly in the case of brightness change range is very big.When me
With video camera pair the scene including highlight regions and backlight area is found a view simultaneously when, output image highlight regions due to
It is over-exposed to become a piece of white, and in dark area because under-exposure becomes a piece of black, to which these regions can not be seen clearly
Scene so that picture quality is cannot be satisfied the needs of practical application completely.The reason of causing this phenomenon is common camera
Image defects, that is, the dynamic range usually said are insufficient.And wide dynamic technique be due to that can cover broader dynamic range,
The image quality for effectively improving the big scene of light variation range has extensively in video monitoring, remote sensing survey, military surveillance etc.
General application.
Current wide dynamic technique is divided to two classes according to processing mode：One kind is hardware mode, and another kind of is software mode.It is based on
The implementation method of hardware is mainly to improve the structure organization of collecting device, including be based on high dynamic range camera, to sensor core
Piece carries out special transformation etc..But such methods realize that the process is more complicated, and cost is higher, are unfavorable for large-area applications popularization.
Wider processing mode is the method based on software, represents method and exposes fusion method more as tone mapping method and,
All it is that multiple data are acquired with different Exposure modes to Same Scene using general camera, it then will by various algorithm process
Their details is fused into new image namely wide dynamic range image, is finally shown in regular display.This kind of side
Method, which is disadvantageous in that, to be needed repeatedly to shoot scene, is then handled multiple image, real-time is unable to get
Meet.
Invention content：
The present invention proposes a kind of wide method for dynamically processing of view-based access control model system for the deficiency of existing wide dynamic approach,
It only uses a frame image information of current scene, and a two slice width dynamic processors are formulated according to human visual system's feature
System, and use neighborhood of pixels information calculating parameter environmental variance so that treated image can to dark areas also prime information,
Inhibiting strong light to reduce halation bright area influences, is stretched to intermediate region, obtain the dynamic range of bigger.
The technical scheme is that：
A kind of wide method for dynamically processing of view-based access control model system, is as follows：
(1) piece image is shot to given scenario, obtains the RGB color image of the scene；
(2) image is transformed into YUV color spaces from RGB color, obtains luminance information Y-component therein；
Image is transformed into YUV color spaces from RGB color, transformation for mula is as follows：
In YUV color spaces, what luminance information Y and two chrominance informations U, V were completely separate, so this color indicates
Method more meets the visual characteristic of human eye.
(3) the neighborhood of pixels information of Y-component is used to calculate the environmental factor of each pixel；
The value of environmental factor a directly determines output as a result, namely determining the effects of wide dynamic images, environmental factor a
It is calculated using the luma component information Y of YUV color spaces, formula is as follows：
Wherein G is a Gaussian template,It is the mean value of luminance information Y.
(4) according to human visual system's feature, wide dynamic mapping formula is obtained using environmental factor；
Human visual system (Human Visual System, abbreviation HVS) is mainly made of retina, including three work(
Ergosphere：Photosensitive layer, outer plexiform layer.Inner plexiform layer.Photosensitive layer is mainly responsible for the conversion, transmission, compression processing of optical signal.Here of the invention
The optical signal that need to be handled is divided into two parts：First part is incident optical signal, and second part is that outer plexiform layer is fed back based on neighborhood
Optical signal.
First part's incident optical signal is indicated with classics compression formula Naka-Rushton equations：
Wherein x is input, and y is output, and a is environmental factor, and environmental factor a is greater than 0 positive number.
The curve synoptic diagram of formula 1 is as shown in Figure 1, it can be deduced that a rule：When input x belongs to [0,1] section, a
Smaller, y is promoted bigger；A is bigger, and y is promoted smaller.
Second part outer plexiform layer based on the optical signal that neighborhood is fed back is played regulatory role to incident optical signal, this is because people
Eye enters bright place in dark place a adaptation process, or has a adaptation process into dark place in bright place,
This process is exactly a process that gradually feedback adapts to.The present invention indicates this process with following formula：
The curve synoptic diagram of formula 2 is as shown in Figure 2, it can be deduced that a rule：When input x belongs to [0,1] section, a
Smaller, y is promoted smaller；A is bigger, and y is promoted bigger.The changing rule of this changing rule and formula 1 is exactly the opposite, plays one
The effect of adjustment conforms exactly to the visual characteristic of human eye.
Aggregative formula 1 and formula 2 can obtain two layers of vision system of simulation namely proposed by the present invention final
Wide dynamic mapping formula：
From figure 3, it can be seen that by carrying out different adjustment to a, output y is formed with input x by more cluster class gamma curves
The non-linear relation of combination.When input x belongs to [0,1] section, a is smaller, and y is promoted bigger；A is bigger, and y is promoted smaller.
(5) original RGB image is converted using wide dynamic mapping formula, obtains wide dynamic RGB image.
The input x for assigning each pixel as formula 3 when transformation successively in three components (R, G, B) of RGB color, is used
Formula 3 calculates new output (R ', G ', B ').
The wide dynamic approach of view-based access control model system proposed by the present invention preferably retains the thin of highlight regions and dark area
Information is saved, obtained wide dynamic images quality is preferable and resource consumption is few, especially because without passing through repeatedly shooting synthesis one
Panel height dynamic image can be good at meeting the needs of handling in real time.
Description of the drawings：
Fig. 1 is Naka-Rushton equation curve schematic diagrames
Fig. 2 is the equation curve schematic diagram that outer plexiform layer is fed back based on neighborhood；
Fig. 3 is wide dynamic mapping equation curve schematic diagram；
Fig. 4 is the artwork and width dynamic result figure of scene on daytime；
Fig. 5 is the artwork of night-time scene and wide dynamic result figure.
Specific implementation mode：
The present invention is described in further details below in conjunction with the accompanying drawings.
Piece image is shot to given scenario first, obtains the RGB color image of the scene.It is well known that natural
Any type coloured light can all be mixed by R, G, B three primary colours in different ratio additions in boundary, when three primary colours component is all 0
It is mixed into black light；It is mixed into white light when three primary colours component is all 255.RGB color using physics three primary colours indicate,
It is suitble to chromoscope to work, however this representation method and the visual characteristic for not meeting human eye.Thus, we by image from
RGB color transforms to YUV color spaces, and transformation for mula is as follows：
In YUV color spaces, what luminance information Y and two chrominance informations U, V were completely separate, so this color indicates
Method more meets the visual characteristic of human eye.
Human visual system (Human Visual System, abbreviation HVS) is mainly made of retina, including three work(
Ergosphere：Photosensitive layer, outer plexiform layer.Inner plexiform layer.Photosensitive layer is mainly responsible for the conversion, transmission, compression processing of optical signal.Here of the invention
The optical signal that need to be handled is divided into two parts：First part is incident optical signal, and second part is that outer plexiform layer is fed back based on neighborhood
Optical signal.
First part's incident optical signal is indicated with classics compression formula Naka-Rushton equations：
Wherein x is input, and y is output, and a is environmental factor, and environmental factor a is greater than 0 positive number.
The curve synoptic diagram of formula 1 is as shown in Figure 1, it can be deduced that a rule：When input x belongs to [0,1] section, a
Smaller, y is promoted bigger；A is bigger, and y is promoted smaller.
The feedback of second part is played regulatory role to first part, this is because human eye enters bright ground in dark place
There is a adaptation process in side, or has a adaptation process into dark place in bright place, this process is exactly one gradually anti-
Present the process adapted to.The present invention indicates this process with following formula：
The curve synoptic diagram of formula 2 is as shown in Figure 2, it can be deduced that a rule：When input x belongs to [0,1] section, a
Smaller, y is promoted smaller；A is bigger, and y is promoted bigger.The changing rule of this changing rule and formula 1 is exactly the opposite, plays one
The effect of adjustment conforms exactly to the visual characteristic of human eye.
Aggregative formula 1 and formula 2 can obtain two layers of the vision system namely final width proposed by the present invention that one is simulated
Dynamic mapping formula：
From Fig. 3 it will be seen that by carrying out different adjustment to a, output y is formed with input x by more cluster class gammas
The non-linear relation of curve combination.When input x belongs to [0,1] section, a is smaller, and y is promoted bigger；A is bigger, and y is promoted smaller.
The value of environmental factor a directly determines output as a result, namely determining the effects of wide dynamic images.In order to obtain more
Good output effect, the present invention propose a kind of method of the realm information computing environment factor using pixel, and in order to meet people
Eye characteristic, is calculated, formula is as follows using the luma component information Y of YUV color spaces：
Wherein G is a Gaussian template,It is the mean value of luminance information Y.
The environmental factor of each pixel is obtained by the convolution of Y-component and Gaussian template, illustrates to examine when calculating current point
The realm information for having considered pixel, doing so can be to dark areas also prime information；Strong light is inhibited to bright area, reducing halation influences；
Intermediate region is stretched, to obtain the dynamic range of bigger.
A kind of wide method for dynamically processing of view-based access control model system of the present invention, is as follows：
1, piece image is shot to given scenario, obtains the RGB color image of the scene
2, image is transformed into YUV color spaces from RGB color；
3, the environmental factor of each pixel is calculated using pixel neighborhoods information by formula 4；
4, according to human visual system's feature, wide dynamic mapping formula is obtained using environmental factor；
5, three components (R, G, B) each pixel in RGB color regard input successively, are calculated newly with formula 3
Output (R ', G ', B ').
Artwork and the result of wide dynamic images are provided in Fig. 4 and Fig. 5, it can be seen that either scene on daytime or night
Scene, wide dynamic images are substantially better than original image, and more details information, Neng Gouman are remained in bright areas and dark area
The actual application demand of foot.
Claims (1)
1. a kind of wide method for dynamically processing of view-based access control model system, which is characterized in that comprise the following steps：
(1) piece image is shot to given scenario, obtains the RGB color image of the scene；
(2) image is transformed into YUV color spaces from RGB color, obtains luminance information Y-component；
Image is transformed into YUV color spaces from RGB color, transformation for mula is as follows：
In YUV color spaces, what luminance information Y and two chrominance informations U, V were completely separate, so this color showing method
More meet the visual characteristic of human eye；
(3) the neighborhood of pixels information of Y-component is used to calculate the environmental factor of each pixel；
Environmental factor a is calculated using the luma component information Y of YUV color spaces, and formula is as follows：
Wherein G is a Gaussian template,It is the mean value of luminance information Y；
(4) according to human visual system's feature, wide dynamic mapping formula is obtained using environmental factor；
Human visual system is mainly made of retina, including three functional layers：Photosensitive layer, outer plexiform layer and inner plexiform layer；Photosensitive layer
It is mainly responsible for the conversion, transmission, compression processing of optical signal；Here the optical signal that need to be handled is divided into two parts：First part is
Incident optical signal, second part are the optical signal that outer plexiform layer is fed back based on neighborhood；
First part's incident optical signal is indicated with classics compression formula Naka-Rushton equations：
Wherein x is input, and y is output, and a is environmental factor, and environmental factor a is greater than 0 positive number；
One rule can be obtained by formula 1：When input x belongs to [0,1] section, a is smaller, and y is promoted bigger；A is bigger, and y is carried
It rises smaller；
Second part outer plexiform layer based on the optical signal that neighborhood is fed back is played regulatory role to first part's incident optical signal, this be by
Entering bright place in dark place in human eye has a adaptation process, or has a adapted into dark place in bright place
Journey, this process are exactly a process that gradually feedback adapts to；Herein this process is indicated with following formula：
One rule can be obtained by formula 2：When input x belongs to [0,1] section, a is smaller, and y is promoted smaller；A is bigger, and y is carried
It rises bigger；The changing rule of this changing rule and formula 1 is exactly the opposite, plays the role of an adjustment, conforms exactly to
The visual characteristic of human eye；
Aggregative formula 1 and formula 2 can obtain two layers of vision system of a simulation, namely final wide dynamic mapping formula：
(5) original RGB image is converted using wide dynamic mapping formula, obtains wide dynamic RGB image；
The input x for assigning each pixel as formula 3 when transformation successively in three components (R, G, B) of RGB color, uses formula
3 calculate new output (R ', G ', B ').
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510510485.3A CN105118032B (en) | 2015-08-19 | 2015-08-19 | A kind of wide method for dynamically processing of view-based access control model system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510510485.3A CN105118032B (en) | 2015-08-19 | 2015-08-19 | A kind of wide method for dynamically processing of view-based access control model system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105118032A CN105118032A (en) | 2015-12-02 |
| CN105118032B true CN105118032B (en) | 2018-11-06 |
Family
ID=54666007
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510510485.3A Active CN105118032B (en) | 2015-08-19 | 2015-08-19 | A kind of wide method for dynamically processing of view-based access control model system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105118032B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105930854A (en) * | 2016-04-19 | 2016-09-07 | 东华大学 | Manipulator visual system |
| CN106846359B (en) * | 2017-01-17 | 2019-09-20 | 湖南优象科技有限公司 | Moving target rapid detection method based on video sequence |
| CN114051098B (en) * | 2021-11-23 | 2023-05-30 | 河南牧业经济学院 | Intelligent visual image acquisition method and platform |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679157A (en) * | 2013-12-31 | 2014-03-26 | 电子科技大学 | Human face image illumination processing method based on retina model |
| CN103870820A (en) * | 2014-04-04 | 2014-06-18 | 南京工程学院 | Illumination normalization method for extreme illumination face recognition |
-
2015
- 2015-08-19 CN CN201510510485.3A patent/CN105118032B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679157A (en) * | 2013-12-31 | 2014-03-26 | 电子科技大学 | Human face image illumination processing method based on retina model |
| CN103870820A (en) * | 2014-04-04 | 2014-06-18 | 南京工程学院 | Illumination normalization method for extreme illumination face recognition |
Non-Patent Citations (3)
| Title |
|---|
| Model of retinal local adaptation for the tone mapping of color filter array images;Laurence Meylan等;《Optical Society of America》;20070930;第24卷(第9期);第2807-2816页 * |
| YUV空间的彩色图像HDR合成算法;陈军等;《计算机工程》;20120531;第38卷(第9期);第214-216页 * |
| 基于人眼视觉感知的场景明暗适应动态过程模拟;王章野等;《软件学报》;20101231;第21卷;第111-122页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105118032A (en) | 2015-12-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhang et al. | Nighttime haze removal based on a new imaging model | |
| CN106897981A (en) | A kind of enhancement method of low-illumination image based on guiding filtering | |
| CN106886985B (en) | A kind of adaptive enhancement method of low-illumination image reducing colour cast | |
| CN105915909B (en) | A kind of high dynamic range images layered compression method | |
| CN106169182B (en) | A method of synthesizing several different exposure images | |
| CN103034986A (en) | Night vision image enhancement method based on exposure fusion | |
| US20120212477A1 (en) | Fast Haze Removal and Three Dimensional Depth Calculation | |
| CN106204470A (en) | Low-light-level imaging method based on fuzzy theory | |
| CN110706172B (en) | Low-illumination color image enhancement method based on adaptive chaotic particle swarm optimization | |
| CN111970432A (en) | Image processing method and image processing device | |
| CN103632351A (en) | All-weather traffic image enhancement method based on brightness datum drift | |
| CN102340673A (en) | White balance method for video camera aiming at traffic scene | |
| Wang et al. | Variational single nighttime image haze removal with a gray haze-line prior | |
| CN111105359B (en) | Tone mapping method for high dynamic range image | |
| CN105118032B (en) | A kind of wide method for dynamically processing of view-based access control model system | |
| CN108024057A (en) | Background blurring processing method, device and equipment | |
| CN107481214A (en) | A kind of twilight image and infrared image fusion method | |
| CN110163807A (en) | A kind of enhancement method of low-illumination image based on the bright channel of expectation | |
| CN110223262A (en) | A kind of rapid image fusion method based on Pixel-level | |
| CN106550227A (en) | A kind of image saturation method of adjustment and device | |
| CN110175967B (en) | Image defogging processing method, system, computer device and storage medium | |
| CN109427041A (en) | A kind of image white balance method and system, storage medium and terminal device | |
| CN111292279A (en) | Polarization image visualization method based on color image fusion | |
| CN109325905B (en) | Image processing method, image processing device, computer readable storage medium and electronic apparatus | |
| JP4359662B2 (en) | Color image exposure compensation method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |