CN109214996A - A kind of image processing method and device - Google Patents
A kind of image processing method and device Download PDFInfo
- Publication number
- CN109214996A CN109214996A CN201810994081.XA CN201810994081A CN109214996A CN 109214996 A CN109214996 A CN 109214996A CN 201810994081 A CN201810994081 A CN 201810994081A CN 109214996 A CN109214996 A CN 109214996A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- gray
- gray level
- level image
- 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.)
- Granted
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 61
- 238000001914 filtration Methods 0.000 claims abstract description 47
- 238000004590 computer program Methods 0.000 claims description 23
- 238000012545 processing Methods 0.000 claims description 23
- 230000015654 memory Effects 0.000 claims description 17
- 238000003860 storage Methods 0.000 claims description 14
- 230000016776 visual perception Effects 0.000 claims description 9
- 238000009826 distribution Methods 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 2
- 230000011218 segmentation Effects 0.000 abstract description 5
- 238000009434 installation Methods 0.000 description 9
- 238000005520 cutting process Methods 0.000 description 8
- 230000006870 function Effects 0.000 description 7
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 238000003706 image smoothing Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 230000000644 propagated effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Classifications
-
- G06T5/77—
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/20172—Image enhancement details
Abstract
The embodiment of the invention discloses a kind of image processing method and devices, and foreground image and background image in image are gone out for Fast Segmentation.The embodiment of the present application method includes: acquisition original image, judges whether the original image is single channel image；If it is not, the original image is converted to single channel image, to obtain the gray level image of the original image；High frequency filter is carried out to the gray value of each pixel in the gray level image, to obtain the first gray level image；According to the gray level image and first gray level image, Kalman filtering is carried out to the gray value of pixel each in gray level image, to obtain the template image of the gray level image；Calculate the gray value of each pixel norm at a distance from the gray value difference of respective pixel in first gray level image in the template image, if described be greater than preset threshold apart from norm, current pixel is then defined as foreground image, otherwise, current pixel is defined as background image.
Description
Technical field
This application involves technical field of image processing more particularly to a kind of image processing methods and device.
Background technique
With the development of computer technology, more and more information are propagated in the form of digital picture, and are being calculated
In machine image procossing, foreground segmentation and extraction are a fundamental operations, and so-called foreground segmentation refers to that computer is allowed to scheme from a pair
Judge which is foreground object in piece, which is background object, and from from being partitioned into interested prospect critical object.
In natural scene, image background is complicated, resolution ratio is low, and image diversification, and distribution is random, and in traditional figure
As being mainly directed towards the file and picture of high quality, needing first to image denoising, increase, distortion correction, scaling etc. to image in identification
It is pre-processed, it is horizontal that very high identification can be reached in the case where meeting the requirements.Because of good image preprocessing process,
It is the committed step for influencing the identification of later image prospect.
And traditional image preprocessing process, it is illuminated by the light and image shadow effect is larger, can not fast implement quickly to figure
The pretreatment of picture, with the foreground picture and Background being partitioned into image.
Summary of the invention
The embodiment of the present application provides a kind of image processing method and device, for quickly carrying out High frequency filter to image,
And Kalman filtering is carried out again to the image after High frequency filter, foreground image and background in present image are gone out with Fast Segmentation
Image.
The embodiment of the present application first aspect provides a kind of image processing method, comprising:
Original image is obtained, judges whether the original image is single channel image；
If it is not, the original image is converted to single channel image, to obtain the gray level image of the original image；
High frequency filter is carried out to the gray value of each pixel in the gray level image, to obtain the first gray level image；
According to the gray level image and first gray level image, to the gray value card of pixel each in gray level image
Kalman Filtering, to obtain the template image of the gray level image；
Calculate the gray scale of the gray value of each pixel and respective pixel in first gray level image in the template image
Current pixel is defined as foreground image if described be greater than preset threshold apart from norm apart from norm by value difference value, otherwise,
Current pixel is defined as background image.
Preferably, described according to the gray level image and first gray level image, to pixel each in gray level image
Gray value carries out Kalman filtering, to obtain the template image of the gray level image, comprising:
The background gray scale of the gray level image is estimated, to obtain the background characteristics gray value of the gray level image；
It in the gray level image, chooses centered on G (i, j), m*n is the region of size, to each in the region
The gray value of pixel usesHigh frequency filter is carried out, to obtain first gray level image in the region；
(1) and formula (2) calculate the gray scale predicted value of each pixel in gray level image according to the following formula:
W1+w2+w3=1； (2)
If any pixel in B (i-1, j-1), B (i-1, j) or B (i, j-1) exceeds the boundary of B, then enables and exceed the side
The pixel value on boundary is the background characteristics gray value；
It is modified according to gray scale predicted value of the formula (3) to each pixel:
To obtain the gray value of each pixel in template image.
Preferably, the gray value for calculating each pixel in the template image is corresponding with first gray level image
Current pixel is defined as prospect if described be greater than preset threshold apart from norm apart from norm by the gray value difference of pixel
Otherwise current pixel is defined as background image by image, comprising:
The gray value for calculating each pixel in the template image according to formula (4) is corresponding with first gray level image
The gray value difference of pixel apart from norm:
Determine that current pixel is foreground image or background image according to formula (5), ε is visual perception gray threshold:
If described be greater than the ε apart from norm, current pixel is defined as foreground image, if described little apart from norm
In the ε, then current pixel is defined as background image.
Preferably, the algorithm estimated the background gray scale of the gray level image includes:
One of background gray scale mode method, background gray average method and background gray scale fitted Gaussian distribution averaging method are more
Kind.
Preferably, the method to the gray value progress High frequency filter of each pixel in the gray level image includes:
Mean filter, gaussian filtering or Gauss-Laplace are carried out to the gray value of each pixel in the gray level image
Filtering.
Preferably, the method also includes:
Output is carried out to the foreground image to show.
The embodiment of the present application also provides a kind of image processing apparatus, comprising:
Acquiring unit judges whether the original image is single channel image for obtaining original image；
Converting unit, for when the original image is not single channel image, the original image to be converted to single-pass
Road image, to obtain the gray level image of the original image；
High frequency filter unit carries out High frequency filter for the gray value to each pixel in the gray level image, to obtain
First gray level image；
Kalman filtering unit is used for according to the gray level image and first gray level image, to every in gray level image
The gray value of a pixel carries out Kalman filtering, to obtain the template image of the gray level image；
Determination unit, for calculate in the template image gray value of each pixel with it is right in first gray level image
Answer the gray value difference of pixel apart from norm, if described be greater than preset threshold apart from norm, before current pixel is defined as
Otherwise current pixel is defined as background image by scape image.
Preferably, the Kalman filtering unit, comprising:
Background gray scale estimation module is estimated for the background gray scale to the gray level image, to obtain the gray scale
The background characteristics gray value of image；
High frequency filter module, for choosing centered on G (i, j) in the gray level image, m*n is the region of size,
The gray value of each pixel in the region is usedHigh frequency filter is carried out, to obtain the of the region
One gray level image；
Gray scale prediction module calculates the gray scale of each pixel in gray level image for (1) according to the following formula and formula (2)
Predicted value:
W1+w2+w3=1； (2)
If any pixel in B (i-1, j-1), B (i-1, j) or B (i, j-1) exceeds the boundary of B, then enables and exceed the side
The pixel value on boundary is the background characteristics gray value；
Correction module, for being modified according to gray scale predicted value of the formula (3) to each pixel:
To obtain the gray value of each pixel in template image.
Preferably, the determination unit, comprising:
Computing module, for calculating the gray value and described first of each pixel in the template image according to formula (4)
In gray level image the gray value difference of respective pixel apart from norm:
Determining module, for determining that current pixel is foreground image or background image according to formula (5), ε is visual perception
Gray threshold:
If described be greater than the ε apart from norm, current pixel is defined as foreground image, if described little apart from norm
In the ε, then current pixel is defined as background image.
Preferably, the algorithm packet background gray scale of the gray level image estimated in the background gray scale estimation module
It includes:
One of background gray scale mode method, background gray average method and background gray scale fitted Gaussian distribution averaging method are more
Kind.
Preferably, High frequency filter is carried out to the gray value of each pixel in the gray level image in the High frequency filter unit
Method include:
Mean filter, gaussian filtering or Gauss-Laplace are carried out to the gray value of each pixel in the gray level image
Filtering.
Preferably, described image processing unit further includes output module, is shown for carrying out output to the foreground image.
The embodiment of the present application also provides a kind of readable storage medium storing program for executing, are stored thereon with computer program, the computer journey
When sequence is executed by processor, for realizing the image processing method of the application first aspect offer.
As can be seen from the above technical solutions, the embodiment of the present application has the advantage that
In the embodiment of the present application, original image is first obtained, original image is converted into gray level image, gray level image is executed
High frequency filter generates the first gray level image, then according to gray level image and the first gray level image, executes Kalman to gray level image
Filtering generates template image then according to template image and gray level image and calculates the distance of the gray value difference of corresponding pixel points
Norm, and when this is greater than preset threshold apart from norm, which is defined as foreground image, and it is little in the distance range
When preset threshold, which is defined as background image, to simplify the cutting procedure of foreground image, and at the image
Reason method has the characteristics that very high anti-light photograph and anti-shade.
Detailed description of the invention
Fig. 1 is a kind of one embodiment schematic diagram of image processing method in the embodiment of the present application；
Fig. 2 is the refinement step of step 104 in embodiment described in Fig. 1；
Fig. 3 is the refinement step of step 105 in embodiment described in Fig. 1；
Fig. 4 is a kind of another embodiment schematic diagram of image processing method in the embodiment of the present application；
Fig. 5 is a kind of one embodiment schematic diagram of image processing apparatus in the embodiment of the present application；
Fig. 6 is a kind of another embodiment signal of image processing apparatus in the embodiment of the present application.
Specific embodiment
The embodiment of the present application provides a kind of image processing method and device, for quickly carrying out High frequency filter to image,
And Kalman filtering is carried out again to the image after High frequency filter, foreground image and background in present image are gone out with Fast Segmentation
Image.
In order to make those skilled in the art more fully understand application scheme, below in conjunction in the embodiment of the present application
Attached drawing, the technical scheme in the embodiment of the application is clearly and completely described, it is clear that described embodiment is only
The embodiment of the application a part, instead of all the embodiments.Based on the embodiment in the application, ordinary skill people
Member's every other embodiment obtained without making creative work, all should belong to the model of the application protection
It encloses.
The description and claims of this application and term " first ", " second ", " third ", " in above-mentioned attached drawing
Four " etc. be to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should be understood that using in this way
Data be interchangeable under appropriate circumstances, so that the embodiments described herein can be in addition to illustrating herein or describing
Sequence other than appearance is implemented.In addition, term " includes " and " having " and their any deformation, it is intended that covering is non-exclusive
Include, for example, the process, method, system, product or equipment for containing a series of steps or units are not necessarily limited to clearly arrange
Those of out step or unit, but may include be not clearly listed or it is solid for these process, methods, product or equipment
The other step or units having.
For convenience of understanding, the image processing method in the embodiment of the present application is described below, referring to Fig. 1, this Shen
Please in embodiment image processing method one embodiment, comprising:
101, original image is obtained, judges whether the original image is single channel image, if it is not, 102 are thened follow the steps,
If so then execute step 103；
Before to image procossing, it is necessary first to obtain processed original image, the original image in the application can be from
It is read out in video camera, computer, camera or other image storage apparatus, and original image can be jpeg, flash
Any one of pix, Tiff, gif or mpeg are not particularly limited herein.
After getting original image, judge whether the original image is single channel image, i.e. gray level image, if the original graph
As this as single channel image (gray level image), then directly to the original image execute step 103, if the original image sheet as
Color image, i.e., non-single channel image then execute step 102 to the original image.
102, the original image is converted into single channel image, to obtain the gray level image of the original image；
If original image is non-single channel image, original image is handled according to following formula, by original graph
As being converted to single channel image:
G (i, j)=0.299rA(i,j)+0.587·gA(i,j)+0.114·bA(i,j)
Wherein, A (i, j) is the pixel in original image, and rA(i,j),gA(i, j) and bA(i, j) is respectively original graph
As the channel r of A, the channel g and the channel b, G (i, j) are single channel image, i.e. pixel in gray level image.
103, High frequency filter is carried out to the gray value of each pixel in the gray level image, to obtain the first gray level image；
After obtaining the gray level image of original image, High frequency filter is carried out to each pixel G (i, j) in gray level image,
To obtain filtered first gray level image.
Often there is the noise jamming of certain procedures during actual image acquisition, transmission and processing image, it should
The noise penalty quality of image has flooded feature so that image is fuzzy, brings difficulty to picture analyzing, and High frequency filter,
It is a logical Image Smoothing Skill, the noise in Image Acquisition, transmission and treatment process can be eliminated.
Specifically, High frequency filter can be executed to gray level image by a variety of methods, it is such as high to eliminate the noise in image
This filtering, mean filter, Gauss-Laplace filtering etc., can choose size in practical applications is m*n, and different type
Template typeGray level image is filtered, whereinIt can be Gauss operator, mean operator or Gauss-Laplace are calculated
Son.
Below by taking m*n is the mean operator of 3*3 as an example, the filtering of gray level image is illustrated:
Assuming that Filtering Template is as described in Table 1:
Table 1
Wherein, mean filter is to object pixel to a template, which includes surrounding adjacent pixels (with mesh
8 pixels around centered on pixel are marked, a Filtering Template is constituted, that is, removes object pixel itself), then with complete in template
The average value of volumetric pixel replaces original pixel value.
Target pixel value then after mean filter is as shown in table 2:
According to the definition of mean filter it is found that the mean value pixel of object pixel are as follows:
(5+3+6+2+1+9+8+4+7)/9=45/9=5
And the process of gaussian filtering and Laplacian filtering, have specific descriptions in the prior art, herein no longer
It repeats.
104, according to the gray level image and first gray level image, to the gray value of pixel each in gray level image into
Row Kalman filtering, to obtain the template image of the gray level image；
High frequency filter is being carried out to gray level image, after obtaining the first gray level image, further according to gray level image and the first gray scale
Image carries out Kalman filtering to the gray value of pixel each in gray level image, to obtain the template image of gray level image.
Specifically, Kalman filtering is first to be carried out according to pixel value of the optimization algorithm to pixel each in gray level image
Estimation, is then modified estimated value using the pixel value of actual measurement, to obtain the pixel value closer to true value, and it is right
The process of specific Kalman filtering is described in detail in the following embodiments in this present embodiment, and details are not described herein again.
105, the gray value of each pixel and respective pixel in first gray level image in the calculating template image
Current pixel is defined as foreground image if described be greater than preset threshold apart from norm apart from norm by gray value difference, no
Then, current pixel is defined as background image.
After obtaining template image in step 104, the gray value of each pixel and the first ash in template image are further calculated
Spend image (gray level image i.e. after High frequency filter) in respective pixel gray value difference apart from norm, if obtain apart from model
Number is greater than preset threshold, then current pixel is defined as foreground pixel, otherwise, current pixel is defined as background pixel.
Specifically, the realization process of step 105 is described in detail in the following embodiments.
In the embodiment of the present application, original image is first obtained, original image is converted into gray level image, gray level image is executed
High frequency filter generates the first gray level image, then according to gray level image and the first gray level image, executes Kalman to gray level image
Filtering generates template image then according to template image and gray level image and calculates the distance of the gray value difference of corresponding pixel points
Norm, and when this is greater than preset threshold apart from norm, which is defined as foreground image, and it is little in the distance range
When preset threshold, which is defined as background image, to simplify the cutting procedure of foreground image, and at the image
Reason method has the characteristics that very high anti-light photograph and anti-shade.
Based on Fig. 1 the embodiment described, the step 104 of Fig. 1 is described below in detail, referring to Fig. 2, Fig. 2 is step in Fig. 1
104 refinement step:
1041, the background gray scale of gray level image is estimated, to obtain the background characteristics gray value of the gray level image；
After obtaining gray level image in a step 102, the background gray scale of the gray level image is estimated, specific estimation is calculated
Method can be gray level image mode, and mean value or the mean value of fitted Gaussian distribution etc. are not particularly limited herein.
After executing the estimation of background gray scale using the above method to gray level image, the background characteristics ash of available gray level image
Angle value K.
1042, it in gray level image, chooses centered on G (i, j), m*n is the region of size, to each in the region
The gray value of pixel usesHigh frequency filter is carried out, to obtain the first gray level image；
The step is similar to step 103, is a specific implementation of step 103, in gray level image G, chooses
Centered on G (i, j), m*n is the region of size, to the gray value of each pixel in the region with High frequency filter algorithm into
Row filtering, to obtain first gray level image in the region.
1043, according to formula (1) and formula (2), the gray scale predicted value of each pixel in gray level image is calculated, and works as B (i-
1, j-1), any pixel in B (i-1, j) or B (i, j-1) exceeds the boundary of B, then enabling the pixel value beyond the boundary is institute
State background characteristics gray value；
W1+w2+w3=1； (2)
Specifically, formula (1) and formula (2), for the process that the gray value to pixel each in gray level image is predicted,
Wherein formula (1) is the prediction according to the pixel value of three pixels in the target pixel points upper left corner to target pixel points pixel value
Process,
Such as
And B (0,0), B (0,1), B (1,0), all have exceeded the boundary of B, therefore enable B (0,0), B (0,1), B (1,0), all for
The background gray feature value of gray level image, i.e. k in step 1041, thenAnd for other
The pixel value of location point is then calculated using similar recurrence method, and details are not described herein again.
1044, gray scale predicted value is modified according to formula (3), to obtain the gray scale of each pixel in template image
Value.
In step 1043, the predicted value to pixel each in gray level image has been obtainedAfterwards, it is further filtered according to high frequency
The pixel value measured after wave is modified predicted value, so that revised pixel value B (i, j) is closer to true value, and
Revised grey scale pixel value B (i, j) is the gray value of each pixel in template image.
How the embodiment of the present application is described in detail according to gray level image and the first gray level image, is carried out to gray level image
Kalman filtering improves the exploitativeness of the application to obtain the process of template image.
Based on Fig. 1 the embodiment described, the step 105 in embodiment described in Fig. 1 is described below in detail, referring to Fig. 3, figure
3 be the refinement step of step 105 in Fig. 1:
1051, the gray value and first gray level image of each pixel in the template image are calculated according to formula (4)
The gray value difference of middle respective pixel apart from norm:
It obtains in template image after the gray value of each pixel, further according to each in formula (4) calculation template image
The gray value of pixel norm at a distance from the difference of respective pixel gray value in the first gray level image, wherein B (i, j) is Prototype drawing
The gray value of each pixel as in,For the gray value of each pixel in the first gray level image, and C (i, j) is
The gray value of each pixel and norm at a distance from respective pixel gray value difference in the first gray level image in template image.
It should be noted that the norm in the present embodiment can be L1 norm, L2 norm, L- ∞ norm, it can be according to reality
Border demand, progress is customized, is not particularly limited herein.
1052, determine that current pixel is foreground image or background image according to formula (5), ε is visual perception gray threshold:
If described be greater than the ε apart from norm, current pixel is defined as foreground image, if described little apart from norm
In the ε, then current pixel is defined as background image.
In specific image processing process, obtain in template image the gray value of each pixel with it is right in the first gray level image
Answer the gray value difference of pixel after norm C (i, j), this is compared apart from norm and visual perception gray threshold ε,
If this is greater than ε apart from norm, illustrate that the contrast of current pixel is obvious, as foreground image, if this is not more than ε apart from norm,
Then illustrate that current pixel contrast is unobvious, as background image.
The embodiment of the present application is described in detail the cutting procedure of foreground image and background image in gray level image, is improved
The exploitativeness of the application.
Below with reference to Fig. 1, Fig. 2 and Fig. 3 the embodiment described, the method that the application image procossing is described in detail is please referred to
Fig. 4, another embodiment of image processing method in the application, comprising:
401, original image is obtained, judges whether the original image is single channel image, if it is not, 402 are thened follow the steps,
If so then execute step 403；
402, the original image is converted into single channel image, to obtain the gray level image of the original image；
403, High frequency filter is carried out to the gray value of each pixel in the gray level image, to obtain the first gray level image；
404, the background gray scale of gray level image is estimated, to obtain the background characteristics gray value of the gray level image；
405, it in gray level image, chooses centered on G (i, j), m*n is the region of size, to each in the region
The gray value of pixel usesHigh frequency filter is carried out, to obtain first gray level image in the region；
406, according to formula (1) and formula (2), the gray scale predicted value of each pixel in gray level image is calculated, and works as B (i-
1, j-1), any pixel in B (i-1, j) or B (i, j-1) exceeds the boundary of B, then enabling the pixel value beyond the boundary is institute
State background characteristics gray value；
W1+w2+w3=1； (2)
407, pixel predictors are modified according to formula (3), to obtain the gray value of each pixel in template image；
408, it is calculated in the template image in the gray value and first gray level image of each pixel according to formula (4)
The gray value difference of respective pixel apart from norm:
409, determine that current pixel is foreground image or background image according to formula (5), ε is visual perception gray threshold:
If described be greater than the ε apart from norm, current pixel is defined as foreground image, if described little apart from norm
In the ε, then current pixel is defined as background image.
It should be noted that step 401 in the present embodiment is to 409 and the step in embodiment described in Fig. 1, Fig. 2 and Fig. 3
Similar, details are not described herein again.
410, output is carried out to foreground image to show.
It, can be further to foreground image location point after being partitioned into the foreground image in gray level image in step 409
Gray value is shown, to obtain the foreground image in gray level image, gray scale that can also further to background image location point
Value is shown, to obtain the background image in gray level image.
In the embodiment of the present application, original image is first obtained, original image is converted into gray level image, gray level image is executed
High frequency filter generates the first gray level image, then according to gray level image and the first gray level image, executes Kalman to gray level image
Filtering generates template image then according to template image and gray level image and calculates the distance of the gray value difference of corresponding pixel points
Norm, and when this is greater than preset threshold apart from norm, which is defined as foreground image, and it is little in the distance range
When preset threshold, which is defined as background image, to simplify the cutting procedure of foreground image, and at the image
Reason method has the characteristics that very high anti-light photograph and anti-shade.
Secondly, the embodiment of the present application is described in detail how according to gray level image and the first gray level image, to gray level image
Kalman filtering is carried out to be also well described according to template image to obtain the process of template image to prospect in gray level image
The cutting procedure of image and background image improves the exploitativeness of the application.
Described above is the image processing methods in the embodiment of the present application, below to the image procossing in the embodiment of the present application
Device is described, referring to Fig. 5, in the embodiment of the present application image processing apparatus one embodiment, comprising:
Acquiring unit 501 judges whether the original image is single channel image for obtaining original image；
Converting unit 502, for when the original image is not single channel image, the original image to be converted to list
Channel image, to obtain the gray level image of the original image；
High frequency filter unit 503 carries out High frequency filter for the gray value to each pixel in the gray level image, with
To the first gray level image；
Kalman filtering unit 504 is used for according to the gray level image and first gray level image, in gray level image
The gray value of each pixel carries out Kalman filtering, to obtain the template image of the gray level image；
Determination unit 505, for calculating the gray value of each pixel and first gray level image in the template image
The gray value difference of middle respective pixel defines current pixel if described be greater than preset threshold apart from norm apart from norm
Current pixel is otherwise defined as background image for foreground image.
It should be noted that the effect of each unit is similar with the description in embodiment described in Fig. 1 in the embodiment of the present application, this
Place repeats no more.
In the embodiment of the present application, original image is obtained by acquiring unit 501, and judge whether original image is single channel
Original image is converted to gray level image by converting unit 502, is held by High frequency filter unit 503 to gray level image by image
Row High frequency filter generates the first gray level image, then according to gray level image and the first gray level image, executes karr to gray level image
Graceful filtering generates template image, then according to template image and gray level image, calculate the gray value difference of corresponding pixel points away from
From norm, and when this is greater than preset threshold apart from norm, which is defined as foreground image, and the distance range not
When greater than preset threshold, which is defined as background image, to simplify the cutting procedure of foreground image, and the image
Processing method has the characteristics that very high anti-light photograph and anti-shade.
Based on Fig. 5 the embodiment described, the image processing apparatus in the embodiment of the present application is described below in detail, please refers to figure
6, another embodiment of image processing apparatus in the embodiment of the present application, comprising:
Acquiring unit 601 judges whether the original image is single channel image for obtaining original image；
Converting unit 602, for when the original image is not single channel image, the original image to be converted to list
Channel image, to obtain the gray level image of the original image；
High frequency filter unit 603 carries out High frequency filter for the gray value to each pixel in the gray level image, with
To the first gray level image；
Kalman filtering unit 604 is used for according to the gray level image and first gray level image, in gray level image
The gray value of each pixel carries out Kalman filtering, to obtain the template image of the gray level image；
Determination unit 605, for calculating the gray value of each pixel and first gray level image in the template image
The gray value difference of middle respective pixel defines current pixel if described be greater than preset threshold apart from norm apart from norm
Current pixel is otherwise defined as background image for foreground image.
Preferably, the Kalman filtering unit 604, comprising:
Background gray scale estimation module 6041 is estimated for the background gray scale to the gray level image, described to obtain
The background characteristics gray value of gray level image；
High frequency filter module 6042, for choosing centered on G (i, j) in the gray level image, m*n is size
Region uses the gray value of each pixel in the regionHigh frequency filter is carried out, to obtain the region
The first gray level image；
Gray value prediction module 6043 calculates each pixel in gray level image for (1) according to the following formula and formula (2)
Gray scale predicted value:
W1+w2+w3=1； (2)
If any pixel in B (i-1, j-1), B (i-1, j) or B (i, j-1) exceeds the boundary of B, then enables and exceed the side
The pixel value on boundary is the background characteristics gray value；
Correction module 6044, for being modified according to gray scale predicted value of the formula (3) to each pixel:
To obtain the gray value of each pixel in template image.
Preferably, the determination unit 605, comprising:
Computing module 6051, for according to formula (4) calculate in the template image gray value of each pixel with it is described
In first gray level image the gray value difference of respective pixel apart from norm:
Determining module 6052, for determining that current pixel is foreground image or background image according to formula (5), ε is vision
Perceive gray threshold:
If described be greater than the ε apart from norm, current pixel is defined as foreground image, if described little apart from norm
In the ε, then current pixel is defined as background image.
It should be noted that retouching in embodiment described in the effect of each unit and each module and Fig. 4 in the embodiment of the present application
State similar, details are not described herein again.
In the embodiment of the present application, original image is obtained by acquiring unit 601, and judge whether original image is single channel
Original image is converted to gray level image by converting unit 602, is held by High frequency filter unit 603 to gray level image by image
Row High frequency filter generates the first gray level image, then according to gray level image and the first gray level image, executes karr to gray level image
Graceful filtering generates template image, then according to template image and gray level image, calculate the gray value difference of corresponding pixel points away from
From norm, and when this is greater than preset threshold apart from norm, which is defined as foreground image, and the distance range not
When greater than preset threshold, which is defined as background image, to simplify the cutting procedure of foreground image, and the image
Processing method has the characteristics that very high anti-light photograph and anti-shade.
Secondly, the embodiment of the present application be described in detail how according to Kalman filtering unit 604 to gray level image card
Kalman Filtering has been also well described according to determination unit 605 with obtaining the process of template image to foreground image in gray level image
With the cutting procedure of background image, the exploitativeness of the application is improved.
The image processing apparatus in the embodiment of the present application is described from the angle of modular functionality entity above, under
Face is described the image processing apparatus in the embodiment of the present application from the angle of hardware handles:
Image processing apparatus one embodiment includes: in the embodiment of the present application
Processor and memory；
Memory can when processor is used to execute the computer program stored in memory for storing computer program
To realize following steps:
Original image is obtained, judges whether the original image is single channel image；
If it is not, the original image is converted to single channel image, to obtain the gray level image of the original image；
High frequency filter is carried out to the gray value of each pixel in the gray level image, to obtain the first gray level image；
According to the gray level image and first gray level image, to the gray value card of pixel each in gray level image
Kalman Filtering, to obtain the template image of the gray level image；
Calculate the gray scale of the gray value of each pixel and respective pixel in first gray level image in the template image
Current pixel is defined as foreground image if described be greater than preset threshold apart from norm apart from norm by value difference value, otherwise,
Current pixel is defined as background image.
In some embodiments of the present application, processor can be also used for realizing following steps:
The background gray scale of the gray level image is estimated, to obtain the background characteristics gray value of the gray level image；
It in the gray level image, chooses centered on G (i, j), m*n is the region of size, to each in the region
The gray value of pixel usesHigh frequency filter is carried out, to obtain first gray level image in the region；
(1) and formula (2) calculate the gray scale predicted value of each pixel in gray level image according to the following formula:
W1+w2+w3=1； (2)
If any pixel in B (i-1, j-1), B (i-1, j) or B (i, j-1) exceeds the boundary of B, then enables and exceed the side
The pixel value on boundary is the background characteristics gray value；
It is modified according to gray scale predicted value of the formula (3) to each pixel:
To obtain the gray value of each pixel in template image.
In some embodiments of the present application, processor can be also used for realizing following steps:
The gray value for calculating each pixel in the template image according to formula (4) is corresponding with first gray level image
The gray value difference of pixel apart from norm:
Determine that current pixel is foreground image or background image according to formula (5), ε is visual perception gray threshold:
If described be greater than the ε apart from norm, current pixel is defined as foreground image, if described little apart from norm
In the ε, then current pixel is defined as background image.
In some embodiments of the present application, processor can be also used for realizing following steps:
Output is carried out to the foreground image to show.
It is understood that when the processor in above explained image processing apparatus executes the computer program,
The function of each unit in above-mentioned corresponding each Installation practice may be implemented, details are not described herein again.Illustratively, the computer
Program can be divided into one or more module/units, and one or more of module/units are stored in the storage
It in device, and is executed by the processor, to complete the application.One or more of module/units, which can be, can complete spy
Determine the series of computation machine program instruction section of function, which fills for describing the computer program in described image processing
Implementation procedure in setting.For example, the computer program can be divided into each unit in above-mentioned image processing apparatus, each list
The concrete function as described in above-mentioned respective image processing unit may be implemented in member.
The computer installation can be the calculating such as desktop PC, notebook, palm PC and cloud server and set
It is standby.The computer installation may include, but are not limited to processor, memory.It will be understood by those skilled in the art that processor,
Memory is only the example of computer installation, does not constitute the restriction to computer installation, may include more or fewer
Component perhaps combines certain components or different components, such as the computer installation can also be set including input and output
Standby, network access equipment, bus etc..
The processor can be central processing unit (Central Processing Unit, CPU), can also be it
His general processor, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable GateArray, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng the processor is the control centre of the computer installation, utilizes various interfaces and the entire computer installation of connection
Various pieces.
The memory can be used for storing the computer program and/or module, and the processor is by operation or executes
Computer program in the memory and/or module are stored, and calls the data being stored in memory, described in realization
The various functions of computer installation.The memory can mainly include storing program area and storage data area, wherein storage program
It area can application program needed for storage program area, at least one function etc.；Storage data area can store the use according to terminal
The data etc. created.In addition, memory may include high-speed random access memory, it can also include non-volatile memories
Device, such as hard disk, memory, plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure
Digital, SD) card, flash card (Flash Card), at least one disk memory, flush memory device or other volatibility are solid
State memory device.
Present invention also provides a kind of computer readable storage mediums, and the computer readable storage medium is for realizing image
The function of processing unit is stored thereon with computer program, and when computer program is executed by processor, processor can be used for
Execute following steps:
Original image is obtained, judges whether the original image is single channel image；
If it is not, the original image is converted to single channel image, to obtain the gray level image of the original image；
High frequency filter is carried out to the gray value of each pixel in the gray level image, to obtain the first gray level image；
According to the gray level image and first gray level image, to the gray value card of pixel each in gray level image
Kalman Filtering, to obtain the template image of the gray level image；
Calculate the gray scale of the gray value of each pixel and respective pixel in first gray level image in the template image
Current pixel is defined as foreground image if described be greater than preset threshold apart from norm apart from norm by value difference value, otherwise,
Current pixel is defined as background image.
In some embodiments of the present application, the computer program of computer-readable recording medium storage is executed by processor
When, processor can be specifically used for executing following steps:
The background gray scale of the gray level image is estimated, to obtain the background characteristics gray value of the gray level image；
It in the gray level image, chooses centered on G (i, j), m*n is the region of size, to each in the region
The gray value of pixel usesHigh frequency filter is carried out, to obtain first gray level image in the region；
(1) and formula (2) calculate the gray scale predicted value of each pixel in gray level image according to the following formula:
W1+w2+w3=1； (2)
If any pixel in B (i-1, j-1), B (i-1, j) or B (i, j-1) exceeds the boundary of B, then enables and exceed the side
The pixel value on boundary is the background characteristics gray value；
It is modified according to gray scale predicted value of the formula (3) to each pixel:
To obtain the gray value of each pixel in template image.
In some embodiments of the present application, the computer program of computer-readable recording medium storage is executed by processor
When, processor can be specifically used for executing following steps:
The gray value for calculating each pixel in the template image according to formula (4) is corresponding with first gray level image
The gray value difference of pixel apart from norm:
Determine that current pixel is foreground image or background image according to formula (5), ε is visual perception gray threshold:
If described be greater than the ε apart from norm, current pixel is defined as foreground image, if described little apart from norm
In the ε, then current pixel is defined as background image.
In some embodiments of the present application, the computer program of computer-readable recording medium storage is executed by processor
When, processor can be specifically used for executing following steps:
Output is carried out to the foreground image to show.
It is understood that if the integrated unit is realized in the form of SFU software functional unit and as independent production
Product when selling or using, can store in a corresponding computer-readable storage medium.Based on this understanding, this Shen
It please realize all or part of the process in above-mentioned corresponding embodiment method, can also be instructed by computer program relevant
Hardware is completed, and the computer program can be stored in a computer readable storage medium, which is being located
It manages when device executes, it can be achieved that the step of above-mentioned each embodiment of the method.Wherein, the computer program includes computer program generation
Code, the computer program code can be source code form, object identification code form, executable file or certain intermediate forms
Deng.The computer-readable medium may include: any entity or device, record that can carry the computer program code
Medium, USB flash disk, mobile hard disk, magnetic disk, CD, computer storage, read-only memory (ROM, Read-Only Memory), with
Machine access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..
It should be noted that the content that the computer-readable medium includes can be according to legislation and patent practice in jurisdiction
It is required that carrying out increase and decrease appropriate, such as in certain jurisdictions, do not wrapped according to legislation and patent practice, computer-readable medium
Include electric carrier signal and telecommunication signal.
It is apparent to those skilled in the art that for convenience and simplicity of description, the system of foregoing description,
The specific work process of device and unit, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In several embodiments provided herein, it should be understood that disclosed system, device and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of logical function partition, there may be another division manner in actual implementation, such as multiple units or components
It can be combined or can be integrated into another system, or some features can be ignored or not executed.Another point, it is shown or
The mutual coupling, direct-coupling or communication connection discussed can be through some interfaces, the indirect coupling of device or unit
It closes or communicates to connect, can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, each functional unit in each embodiment of the application can integrate in one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
The above, above embodiments are only to illustrate the technical solution of the application, rather than its limitations；Although referring to before
Embodiment is stated the application is described in detail, those skilled in the art should understand that: it still can be to preceding
Technical solution documented by each embodiment is stated to modify or equivalent replacement of some of the technical features；And these
It modifies or replaces, the spirit and scope of each embodiment technical solution of the application that it does not separate the essence of the corresponding technical solution.
Claims (10)
1. a kind of image processing method characterized by comprising
Original image is obtained, judges whether the original image is single channel image；
If it is not, the original image is converted to single channel image, to obtain the gray level image of the original image；
High frequency filter is carried out to the gray value of each pixel in the gray level image, to obtain the first gray level image；
According to the gray level image and first gray level image, Kalman is carried out to the gray value of pixel each in gray level image
Filtering, to obtain the template image of the gray level image；
Calculate the gray value differences of the gray value of each pixel and respective pixel in first gray level image in the template image
Current pixel is defined as foreground image, otherwise, will worked as by value apart from norm if described be greater than preset threshold apart from norm
Preceding pixel is defined as background image.
2. the method according to claim 1, wherein described according to the gray level image and first grayscale image
Picture carries out Kalman filtering to the gray value of pixel each in gray level image, to obtain the template image of the gray level image, packet
It includes:
The background gray scale of the gray level image is estimated, to obtain the background characteristics gray value of the gray level image；
It in the gray level image, chooses centered on G (i, j), m*n is the region of size, to each pixel in the region
Gray value useHigh frequency filter is carried out, to obtain first gray level image in the region；
(1) and formula (2) calculate the gray scale predicted value of each pixel in gray level image according to the following formula:
W1+w2+w3=1； (2)
If any pixel in B (i-1, j-1), B (i-1, j) or B (i, j-1) exceeds the boundary of B, then enable beyond the boundary
Pixel value is the background characteristics gray value；
It is modified according to gray scale predicted value of the formula (3) to each pixel:
To obtain the gray value of each pixel in template image.
3. according to the method described in claim 2, it is characterized in that, the gray scale for calculating each pixel in the template image
Value norm at a distance from the gray value difference of respective pixel in first gray level image, if described be greater than default threshold apart from norm
Value, then be defined as foreground image for current pixel, otherwise, current pixel be defined as background image, comprising:
According to the gray value of each pixel and respective pixel in first gray level image in formula (4) calculating template image
Gray value difference apart from norm:
Determine that current pixel is foreground image or background image according to formula (5), ε is visual perception gray threshold:
If described be greater than the ε apart from norm, current pixel is defined as foreground image, if described be not more than institute apart from norm
ε is stated, then current pixel is defined as background image.
4. according to the method described in claim 2, it is characterized in that, the calculation estimated the background gray scale of the gray level image
Method includes:
One of background gray scale mode method, background gray average method and background gray scale fitted Gaussian distribution averaging method are a variety of.
5. the method according to claim 1, wherein the gray value to each pixel in the gray level image
Carry out High frequency filter method include:
Mean filter, gaussian filtering or Gauss-Laplace filtering are carried out to the gray value of each pixel in the gray level image.
6. a kind of image processing apparatus characterized by comprising
Acquiring unit judges whether the original image is single channel image for obtaining original image；
Converting unit, for when the original image is not single channel image, the original image to be converted to single channel figure
Picture, to obtain the gray level image of the original image；
High frequency filter unit carries out High frequency filter for the gray value to each pixel in the gray level image, to obtain first
Gray level image；
Kalman filtering unit is used for according to the gray level image and first gray level image, to picture each in gray level image
The gray value of element carries out Kalman filtering, to obtain the template image of the gray level image；
Determination unit corresponds to picture in the gray value of each pixel and first gray level image for calculating in the template image
Current pixel is defined as foreground picture if described be greater than preset threshold apart from norm apart from norm by the gray value difference of element
Otherwise current pixel is defined as background image by picture.
7. device according to claim 6, which is characterized in that the Kalman filtering unit, comprising:
Background gray scale estimation module is estimated for the background gray scale to the gray level image, to obtain the gray level image
Background characteristics gray value；
High frequency filter module, for choosing centered on G (i, j) in the gray level image, m*n is the region of size, to institute
The gray value for stating each pixel in region usesHigh frequency filter is carried out, to obtain first ash in the region
Spend image；
Gray scale prediction module calculates the gray scale prediction of each pixel in gray level image for (1) according to the following formula and formula (2)
Value:
W1+w2+w3=1； (2)
If any pixel in B (i-1, j-1), B (i-1, j) or B (i, j-1) exceeds the boundary of B, then enable beyond the boundary
Pixel value is the background characteristics gray value；
Correction module, for being modified according to gray scale predicted value of the formula (3) to each pixel:
To obtain the gray value of each pixel in template image.
8. device according to claim 7, which is characterized in that the determination unit, comprising:
Computing module, for calculating the gray value of each pixel and first gray scale in the template image according to formula (4)
In image the gray value difference of respective pixel apart from norm:
Determining module, for determining that current pixel is foreground image or background image according to formula (5), ε is visual perception gray scale
Threshold value:
If described be greater than the ε apart from norm, current pixel is defined as foreground image, if described be not more than institute apart from norm
ε is stated, then current pixel is defined as background image.
9. a kind of image processing apparatus, including processor, which is characterized in that the processor is stored on memory in execution
When computer program, for realizing the image processing method as described in any one of claims 1 to 5.
10. a kind of readable storage medium storing program for executing, is stored thereon with computer program, which is characterized in that the computer program is by processor
When execution, for realizing the image processing method as described in any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810994081.XA CN109214996B (en) | 2018-08-29 | 2018-08-29 | Image processing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810994081.XA CN109214996B (en) | 2018-08-29 | 2018-08-29 | Image processing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109214996A true CN109214996A (en) | 2019-01-15 |
| CN109214996B CN109214996B (en) | 2021-11-12 |
Family
ID=64985561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810994081.XA Active CN109214996B (en) | 2018-08-29 | 2018-08-29 | Image processing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109214996B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135519A (en) * | 2019-05-27 | 2019-08-16 | 广东工业大学 | A kind of image classification method and device |
| CN111105428A (en) * | 2019-11-08 | 2020-05-05 | 上海航天控制技术研究所 | Star sensor forward filtering hardware image processing method |
| CN111798389A (en) * | 2020-06-30 | 2020-10-20 | 中国工商银行股份有限公司 | Self-adaptive image enhancement method and device |
| CN113298812A (en) * | 2021-04-22 | 2021-08-24 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Image segmentation method, device, system, electronic equipment and readable storage medium |
| CN114255185A (en) * | 2021-12-16 | 2022-03-29 | 武汉高德智感科技有限公司 | Image processing method, device, terminal and storage medium |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040239762A1 (en) * | 2003-05-21 | 2004-12-02 | Porikli Fatih M. | Adaptive background image updating |
| CN101727672A (en) * | 2008-10-24 | 2010-06-09 | 云南正卓信息技术有限公司 | Method for detecting, tracking and identifying object abandoning/stealing event |
| CN101916448A (en) * | 2010-08-09 | 2010-12-15 | 云南清眸科技有限公司 | Moving object detecting method based on Bayesian frame and LBP (Local Binary Pattern) |
| CN102034247A (en) * | 2010-12-23 | 2011-04-27 | 中国科学院自动化研究所 | Motion capture method for binocular vision image based on background modeling |
| CN102750535A (en) * | 2012-04-01 | 2012-10-24 | 北京京东世纪贸易有限公司 | Method and system for automatically extracting image foreground |
| CN102819841A (en) * | 2012-07-30 | 2012-12-12 | 中国科学院自动化研究所 | Global threshold partitioning method for partitioning target image |
| US20140133746A1 (en) * | 2011-10-24 | 2014-05-15 | International Business Machines Corporation | Background understanding in video data |
| CN103873743A (en) * | 2014-03-24 | 2014-06-18 | 中国人民解放军国防科学技术大学 | Video de-noising method based on structure tensor and Kalman filtering |
| CN104166841A (en) * | 2014-07-24 | 2014-11-26 | 浙江大学 | Rapid detection identification method for specified pedestrian or vehicle in video monitoring network |
| CN104599271A (en) * | 2015-01-20 | 2015-05-06 | 中国科学院半导体研究所 | CIE Lab color space based gray threshold segmentation method |
| CN105761261A (en) * | 2016-02-17 | 2016-07-13 | 南京工程学院 | Method for detecting artificial malicious damage to camera |
-
2018
- 2018-08-29 CN CN201810994081.XA patent/CN109214996B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20040239762A1 (en) * | 2003-05-21 | 2004-12-02 | Porikli Fatih M. | Adaptive background image updating |
| CN101727672A (en) * | 2008-10-24 | 2010-06-09 | 云南正卓信息技术有限公司 | Method for detecting, tracking and identifying object abandoning/stealing event |
| CN101916448A (en) * | 2010-08-09 | 2010-12-15 | 云南清眸科技有限公司 | Moving object detecting method based on Bayesian frame and LBP (Local Binary Pattern) |
| CN102034247A (en) * | 2010-12-23 | 2011-04-27 | 中国科学院自动化研究所 | Motion capture method for binocular vision image based on background modeling |
| US20140133746A1 (en) * | 2011-10-24 | 2014-05-15 | International Business Machines Corporation | Background understanding in video data |
| CN102750535A (en) * | 2012-04-01 | 2012-10-24 | 北京京东世纪贸易有限公司 | Method and system for automatically extracting image foreground |
| CN102819841A (en) * | 2012-07-30 | 2012-12-12 | 中国科学院自动化研究所 | Global threshold partitioning method for partitioning target image |
| CN103873743A (en) * | 2014-03-24 | 2014-06-18 | 中国人民解放军国防科学技术大学 | Video de-noising method based on structure tensor and Kalman filtering |
| CN104166841A (en) * | 2014-07-24 | 2014-11-26 | 浙江大学 | Rapid detection identification method for specified pedestrian or vehicle in video monitoring network |
| CN104599271A (en) * | 2015-01-20 | 2015-05-06 | 中国科学院半导体研究所 | CIE Lab color space based gray threshold segmentation method |
| CN105761261A (en) * | 2016-02-17 | 2016-07-13 | 南京工程学院 | Method for detecting artificial malicious damage to camera |
Non-Patent Citations (4)
| Title |
|---|
| CHRISTOF RIDDER等: "Adaptive Background Estimation and Foreground Detection using Kalman-Filtering", 《COMPUTER SCIENCE》 * |
| 任典元等: "基于颜色和局部二值相似模式的背景减除", 《计算机科学》 * |
| 李文光等: "一种改进的卡尔曼滤波背景减除方法", 《信号处理》 * |
| 罗松飞等: "基于灰度区间统计的背景自适应更新算法", 《科技资讯》 * |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110135519A (en) * | 2019-05-27 | 2019-08-16 | 广东工业大学 | A kind of image classification method and device |
| CN110135519B (en) * | 2019-05-27 | 2022-10-21 | 广东工业大学 | Image classification method and device |
| CN111105428A (en) * | 2019-11-08 | 2020-05-05 | 上海航天控制技术研究所 | Star sensor forward filtering hardware image processing method |
| CN111105428B (en) * | 2019-11-08 | 2023-11-14 | 上海航天控制技术研究所 | Star sensor forward filtering hardware image processing method |
| CN111798389A (en) * | 2020-06-30 | 2020-10-20 | 中国工商银行股份有限公司 | Self-adaptive image enhancement method and device |
| CN111798389B (en) * | 2020-06-30 | 2023-08-15 | 中国工商银行股份有限公司 | Adaptive image enhancement method and device |
| CN113298812A (en) * | 2021-04-22 | 2021-08-24 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Image segmentation method, device, system, electronic equipment and readable storage medium |
| CN113298812B (en) * | 2021-04-22 | 2023-11-03 | 中国电子产品可靠性与环境试验研究所（（工业和信息化部电子第五研究所）（中国赛宝实验室）） | Image segmentation method, device, system, electronic equipment and readable storage medium |
| CN114255185A (en) * | 2021-12-16 | 2022-03-29 | 武汉高德智感科技有限公司 | Image processing method, device, terminal and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109214996B (en) | 2021-11-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Bui et al. | Single image dehazing using color ellipsoid prior | |
| US11526995B2 (en) | Robust use of semantic segmentation for depth and disparity estimation | |
| CN109214996A (en) | A kind of image processing method and device | |
| CN107025660B (en) | Method and device for determining image parallax of binocular dynamic vision sensor | |
| CN111402170B (en) | Image enhancement method, device, terminal and computer readable storage medium | |
| GB2526838A (en) | Relightable texture for use in rendering an image | |
| CN112150371B (en) | Image noise reduction method, device, equipment and storage medium | |
| CN108431751B (en) | Background removal | |
| US10728446B2 (en) | Method and apparatus for performing processing in a camera | |
| CN111192226A (en) | Image fusion denoising method, device and system | |
| CN110675334A (en) | Image enhancement method and device | |
| CN113039576A (en) | Image enhancement system and method | |
| CN114627034A (en) | Image enhancement method, training method of image enhancement model and related equipment | |
| CN111563517A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
| Wang et al. | An efficient method for image dehazing | |
| CN110717864B (en) | Image enhancement method, device, terminal equipment and computer readable medium | |
| Chen et al. | Improve transmission by designing filters for image dehazing | |
| CN111447428A (en) | Method and device for converting plane image into three-dimensional image, computer readable storage medium and equipment | |
| WO2015175907A1 (en) | Three dimensional moving pictures with a single imager and microfluidic lens | |
| JP5914046B2 (en) | Image processing apparatus and image processing method | |
| Chang et al. | A self-adaptive single underwater image restoration algorithm for improving graphic quality | |
| CN113888509A (en) | Method, device and equipment for evaluating image definition and storage medium | |
| CN113052923A (en) | Tone mapping method, tone mapping apparatus, electronic device, and storage medium | |
| CN108805838A (en) | A kind of image processing method, mobile terminal and computer readable storage medium | |
| CN109255311A (en) | A kind of information identifying method and system based on image |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |