CN106251358B - A kind of image processing method and device - Google Patents
A kind of image processing method and device Download PDFInfo
- Publication number
- CN106251358B CN106251358B CN201610643274.1A CN201610643274A CN106251358B CN 106251358 B CN106251358 B CN 106251358B CN 201610643274 A CN201610643274 A CN 201610643274A CN 106251358 B CN106251358 B CN 106251358B
- Authority
- CN
- China
- Prior art keywords
- data block
- pixel
- value
- data
- background removal
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Facsimile Image Signal Circuits (AREA)
Abstract
The embodiment of the invention provides a kind of image processing method and devices.On the one hand, the embodiment of the present invention is by being divided at least two data blocks for image to be processed, then, each data block is identified, to obtain the recognition result of each data block, it is content-data block or the data block is block of background data that the recognition result of each data block, which includes the data block, to, according to the recognition result of each data block, obtain the background removal threshold value of each data block, to carry out background removal processing to each data block according to the background removal threshold value of each data block.Therefore, technical solution provided in an embodiment of the present invention, which is able to solve, handles caused excess processes problem due to carrying out unified background removal to whole image in the prior art, and in turn in caused image the problem of content partial loss.
Description
[technical field]
The present invention relates to the processing technology field of image more particularly to a kind of image processing methods and device.
[background technique]
Nowadays, user usually needs that various certificates, document, photo etc. are printed, fax, scan and duplicated
Processing, also, to images such as obtained printout, fax, scanned copy and copies, often picture quality with higher is wanted
It asks.
Due to all including mostly content part and background parts in image, and the background parts in image be it is inessential,
Also, the presence of background parts can interfere identification of the content part in printing, fax, scanning and copying process, cause to obtain
Printout, fax, scanned copy and copy etc. picture quality it is poor.
To solve the above problems, in the prior art, carrying out unified background removal for whole image and handling.That is, needle
To whole image, a fixed background removal threshold value is set, and by this fixed background removal threshold value to entire figure
As carrying out unified background removal processing.
In realizing process of the present invention, at least there are the following problems in the prior art for inventor's discovery:
It in the prior art, for the background parts of whole image and content part, is gone by a fixed background
It except threshold value carries out background removal processing, will cause the excess processes to image, and then lead to the loss of content part in image, carry on the back
The effect of scape removal processing is poor.
[summary of the invention]
In view of this, the embodiment of the invention provides a kind of image processing method and devices, to solve in the prior art
It is interior in caused image due to carrying out excess processes problem caused by unified background removal is handled to whole image, and in turn
The problem of holding partial loss.
On the one hand, the embodiment of the invention provides a kind of image processing methods, comprising:
Image to be processed is divided at least two data blocks;
Each data block is identified, to obtain the recognition result of each data block, each data
It is content-data block or the data block is block of background data that the recognition result of block, which includes the data block,;
According to the recognition result of each data block, the background removal threshold value of each data block is obtained;
According to the background removal threshold value of each data block, background removal processing is carried out to each data block.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, to each described
Data block is identified, to obtain the recognition result of each data block, comprising:
Obtain the average value of the pixel value of pixel in each data block;
According to the pixel value of pixel in the pixel value of pixel in each data block and each data block
Average value, obtain the number of target pixel points in each data block, the target pixel points be within the data block as
Diversity factor in element value and the data block between the average value of the pixel value of pixel is more than or equal to preset difference threshold
The pixel of value;
The number of target pixel points in each data block is compared with preset quantity threshold respectively;
The data block that the number of the target pixel points is less than or equal to the quantity threshold is identified as background data
Block, alternatively, the data block that the number of the target pixel points is greater than the quantity threshold is identified as content-data block.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, to each described
Data block is identified, to obtain the recognition result of each data block, comprising:
Obtain the average value of the pixel value of pixel in each data block;
According to the pixel value of pixel in the pixel value of pixel in each data block and each data block
Average value, obtain the variance of the pixel value of each data block;
The variance of the pixel value of each data block is compared with preset variance threshold values respectively;
The data block that the variance of the pixel value is more than or equal to the variance threshold values is identified as content-data block, or
The data block that the variance of the pixel value is less than the variance threshold values is identified as block of background data by person.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, according to each institute
The recognition result of data block is stated, the background removal threshold value of each data block is obtained, comprising:
If the data block obtains the background removal threshold value by formula as described below for block of background data:
Wherein, CNFor the background removal threshold value of the data block, N is the data block at least two data block
Processing order number, M be processing order number of the data block in block of background data, CM-1For the previous of the data block
The background removal threshold value of block of background data, ANFor the average value of the pixel value of pixel in the data block;
If the data block obtains the background removal threshold value by formula as described below for content-data block:
CN=CN-1+O
Wherein, CNFor the background removal threshold value of the data block, N is the data block at least two data block
Processing order number, CN-1For the background removal threshold value of the previous data block of the data block, O is penalty coefficient.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, according to each institute
The recognition result of data block is stated, the background removal threshold value of each data block is obtained, comprising:
For each data block, a pixel is selected from the data block, using as candidate pixel point;
Obtain pixel number and the candidate picture in the first specified range of the point of candidate pixel described in the data block
The absolute value of difference in second specified range of vegetarian refreshments between pixel number, first specified range and described second refer to
Determine range not to be overlapped completely;
The absolute value is compared with preset absolute value threshold value;
If the absolute value is less than or equal to the absolute value threshold value, determine that the pixel value of the candidate pixel point is institute
State the background removal threshold value of data block.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, according to each institute
The recognition result of data block is stated, the background removal threshold value of each data block is obtained, comprising:
Default first background removal threshold value, the first background removal threshold value are the background removal threshold value of block of background data;
If data block is block of background data, using the first background removal threshold value as the background removal threshold value of the data block;With/
Or,
Default second background removal threshold value, the second background removal threshold value are the background removal threshold value of content-data block;
If data block is content-data block, using the second background removal threshold value as the background removal threshold value of the data block.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, according to each institute
The recognition result of data block is stated, the background removal threshold value of each data block is obtained, comprising:
According to the recognition result of each data block, target data block is determined, it is all before the target data block
Data block is block of background data, is content-data block from all data blocks after the target data block;
Each of before obtaining the target data block in the data block pixel value of pixel average value, using as
The background removal threshold value of each data block in all data blocks before the target data block;And obtain the mesh
The average value for marking the sum of background removal threshold value of all block of background data before data block, as from the target data block
The background removal threshold value of each data block in all data blocks later.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, according to each institute
The background removal threshold value for stating data block carries out background removal processing to each data block, comprising:
According to the background removal threshold value of the pixel value of pixel and each data block in each data block, really
Fixed institute within the data block pixel value greater than the background removal threshold value of the data block pixel, using as each data block
In the first specified pixel point;The pixel value of first specified pixel point in each data block is adjusted to max pixel value;Or
Person,
According to the background removal threshold value of the pixel value of pixel and each data block in each data block, really
Fixed institute within the data block pixel value less than or equal to the background removal threshold value of the data block pixel, using as each institute
State the second specified pixel point in data block;Keep the pixel value of the second specified pixel point in each data block constant.
A technical solution in above-mentioned technical proposal has the following beneficial effects:
In the embodiment of the present invention, by the way that image to be processed is divided at least two data blocks, then, to each number
It is identified according to block, to obtain the recognition result of each data block, the recognition result of each data block includes described
Data block is content-data block or the data block is block of background data, thus, according to the identification knot of each data block
Fruit obtains the background removal threshold value of each data block, in turn, right according to the background removal threshold value of each data block
Each data block carries out background removal processing.In the embodiment of the present invention, by the way that image to be processed is divided at least two
Then lesser data block obtains each data block corresponding one according to the recognition result of the data block after each division
One fixed background removal threshold of middle use is different from the prior art to carry out background removal processing in a background removal threshold value
Value carries out background removal processing to entire image to be processed, and at least two background removal threshold values point have been used in the embodiment of the present invention
The other data block in image to be processed carries out background removal processing, also, the background removal threshold value of each data block is basis
What the recognition result of data block determined, thus the excess processes problem of image is avoided, and then also avoid content partial loss
The problem of, background removal quality can be effectively improved, background removal efficiency is improved, therefore, the embodiment of the present invention can solve
Caused excess processes problem certainly is handled due to carrying out unified background removal to whole image in the prior art, and is led in turn
In the image of cause the problem of content partial loss.
On the other hand, the embodiment of the invention provides a kind of image processing apparatus, comprising:
Division unit, for image to be processed to be divided at least two data blocks;
Recognition unit, for being identified to each data block, to obtain the recognition result of each data block,
It is content-data block or the data block is block of background data that the recognition result of each data block, which includes the data block,;
Acquiring unit, for the recognition result according to each data block, the background for obtaining each data block is gone
Except threshold value;
Processing unit carries on the back each data block for the background removal threshold value according to each data block
Scape removal processing.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the identification are single
Member is specifically used for:
Obtain the average value of the pixel value of pixel in each data block;
According to the pixel value of pixel in the pixel value of pixel in each data block and each data block
Average value, obtain the number of target pixel points in each data block, the target pixel points be within the data block as
Diversity factor in element value and the data block between the average value of the pixel value of pixel is more than or equal to preset difference threshold
The pixel of value;
The number of target pixel points in each data block is compared with preset quantity threshold respectively;
The data block that the number of the target pixel points is less than or equal to the quantity threshold is identified as background data
Block, alternatively, the data block that the number of the target pixel points is greater than the quantity threshold is identified as content-data block.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the identification are single
Member is specifically used for:
Obtain the average value of the pixel value of pixel in each data block;
According to the pixel value of pixel in the pixel value of pixel in each data block and each data block
Average value, obtain the variance of the pixel value of each data block;
The variance of the pixel value of each data block is compared with preset variance threshold values respectively;
The data block that the variance of the pixel value is more than or equal to the variance threshold values is identified as content-data block, or
The data block that the variance of the pixel value is less than the variance threshold values is identified as block of background data by person.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the acquisition are single
Member is specifically used for:
If the data block obtains the background removal threshold value by formula as described below for block of background data:
Wherein, CNFor the background removal threshold value of the data block, N is the data block at least two data block
Processing order number, M be processing order number of the data block in block of background data, CM-1For the previous of the data block
The background removal threshold value of block of background data, ANFor the average value of the pixel value of pixel in the data block;
If the data block obtains the background removal threshold value by formula as described below for content-data block:
CN=CN-1+O
Wherein, CNFor the background removal threshold value of the data block, N is the data block at least two data block
Processing order number, CN-1For the background removal threshold value of the previous data block of the data block, O is penalty coefficient.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the acquisition are single
Member is specifically used for:
For each data block, a pixel is selected from the data block, using as candidate pixel point;
Obtain pixel number and the candidate picture in the first specified range of the point of candidate pixel described in the data block
The absolute value of difference in second specified range of vegetarian refreshments between pixel number, first specified range and described second refer to
Determine range not to be overlapped completely;
The absolute value is compared with preset absolute value threshold value;
If the absolute value is less than or equal to the absolute value threshold value, determine that the pixel value of the candidate pixel point is institute
State the background removal threshold value of data block.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the acquisition are single
Member is specifically used for:
Default first background removal threshold value, the first background removal threshold value are the background removal threshold value of block of background data;
If data block is block of background data, using the first background removal threshold value as the background removal threshold value of the data block;With/
Or,
Default second background removal threshold value, the second background removal threshold value are the background removal threshold value of content-data block;
If data block is content-data block, using the second background removal threshold value as the background removal threshold value of the data block.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the acquisition are single
Member is specifically used for:
According to the recognition result of each data block, target data block is determined, it is all before the target data block
Data block is block of background data, is content-data block from all data blocks after the target data block;
Each of before obtaining the target data block in the data block pixel value of pixel average value, using as
The background removal threshold value of each data block in all data blocks before the target data block;And obtain the mesh
The average value for marking the sum of background removal threshold value of all block of background data before data block, as from the target data block
The background removal threshold value of each data block in all data blocks later.
The aspect and any possible implementation manners as described above, it is further provided a kind of implementation, the processing are single
Member is used for:
According to the background removal threshold value of the pixel value of pixel and each data block in each data block, really
Fixed institute within the data block pixel value greater than the background removal threshold value of the data block pixel, using as each data block
In the first specified pixel point;The pixel value of first specified pixel point in each data block is adjusted to max pixel value;Or
Person,
According to the background removal threshold value of the pixel value of pixel and each data block in each data block, really
Fixed institute within the data block pixel value less than or equal to the background removal threshold value of the data block pixel, using as each institute
State the second specified pixel point in data block;Keep the pixel value of the second specified pixel point in each data block constant.
A technical solution in above-mentioned technical proposal has the following beneficial effects:
In the embodiment of the present invention, image to be processed is divided at least two by the division unit in image processing apparatus
Data block, then, the recognition unit in image processing apparatus identify each data block, to obtain each number
According to the recognition result of block, the recognition result of each data block includes that the data block is content-data block or the data
Block is block of background data, thus, the acquiring unit in image processing apparatus is obtained according to the recognition result of each data block
The background removal threshold value of each data block, in turn, the processing unit in image processing apparatus is according to each data block
Background removal threshold value, background removal processing is carried out to each data block.In the embodiment of the present invention, by by figure to be processed
As being divided at least two lesser data blocks, then, every number is obtained according to the recognition result of the data block after each division
Middle use one is different from the prior art to carry out background removal processing according to the corresponding background removal threshold value of block
Fixed background removal threshold value carries out background removal processing to entire image to be processed, has used at least two in the embodiment of the present invention
A background removal threshold value carries out background removal processing, also, the back of each data block to the data block in image to be processed respectively
Scape removal threshold value is to be determined according to the recognition result of data block, thus avoid the excess processes problem of image, and then also keep away
The problem of having exempted from content partial loss can effectively improve background removal quality, improve background removal efficiency, therefore, this
Inventive embodiments are able to solve in the prior art due to carrying out excessively locating caused by unified background removal processing to whole image
Reason problem, and in turn in caused image the problem of content partial loss.
[Detailed description of the invention]
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be to needed in the embodiment attached
Figure is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for this field
For those of ordinary skill, without any creative labor, it can also be obtained according to these attached drawings other attached
Figure.
Fig. 1 is the flow diagram of the embodiment one of image processing method provided by the embodiment of the present invention;
Fig. 2 is the first pass schematic diagram identified in the embodiment of the present invention to each data block;
Fig. 3 is the second procedure schematic diagram identified in the embodiment of the present invention to each data block;
Fig. 4 is the flow diagram that the background removal threshold value of each data block is obtained in the embodiment of the present invention;
Fig. 5 is the recognition result schematic diagram of one page image to be processed;
Fig. 6 is the recognition result schematic diagram of another page image to be processed;
Fig. 7 is the background removal processing schematic in the embodiment of the present invention to each data block;
Fig. 8 is the flow diagram of the embodiment two of image processing method provided by the embodiment of the present invention;
Fig. 9 is the functional block diagram of image processing apparatus provided by the embodiment of the present invention.
[specific embodiment]
For a better understanding of the technical solution of the present invention, being retouched in detail to the embodiment of the present invention with reference to the accompanying drawing
It states.
It will be appreciated that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Base
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts it is all its
Its embodiment, shall fall within the protection scope of the present invention.
The term used in embodiments of the present invention is only to be not intended to be limiting merely for for the purpose of describing particular embodiments
The present invention.In the embodiment of the present invention and the "an" of singular used in the attached claims, " described " and "the"
It is also intended to including most forms, unless the context clearly indicates other meaning.
It should be appreciated that term "and/or" used herein is only a kind of incidence relation for describing affiliated partner, indicate
There may be three kinds of relationships, for example, A and/or B, can indicate: individualism A, exist simultaneously A and B, individualism B these three
Situation.In addition, character "/" herein, typicallys represent the relationship that forward-backward correlation object is a kind of "or".
It will be appreciated that though specified pixel point may be described in embodiments of the present invention using term first, second etc.,
But these specified pixel points should not necessarily be limited by these terms.These terms are only used to for specified pixel point being distinguished from each other out.For example,
In the case where not departing from range of embodiment of the invention, the first specified pixel point can also be referred to as the second specified pixel point, similar
Ground, the second specified pixel point can also be referred to as the first specified pixel point.
Depending on context, word as used in this " if " can be construed to " ... when " or " when ...
When " or " in response to determination " or " in response to detection ".Similarly, depend on context, phrase " if it is determined that " or " if detection
(condition or event of statement) " can be construed to " when determining " or " in response to determination " or " when the detection (condition of statement
Or event) when " or " in response to detection (condition or event of statement) ".
Embodiment one
The embodiment of the present invention provides a kind of image processing method, referring to FIG. 1, it is figure provided by the embodiment of the present invention
As the flow diagram of the embodiment one of processing method, as shown in Figure 1, method includes the following steps:
Image to be processed is divided at least two data blocks by S101.
It should be noted that image to be processed is to be printed, faxed, scanned and duplicated in the embodiment of the present invention
The image of equal processing, can include but is not limited to various certificates, document, photo, bill etc., the embodiment of the present invention to this without
It is particularly limited to.
Specifically, entire image to be processed is divided at least two data blocks, be in order to each data block respectively into
The processing of row background removal.
During a concrete implementation, the embodiment of the present invention to divide the specific implementation of image to be processed not into
Row is particularly limited to.For example, the division of regular figure can be carried out to image to be processed, the data in regular figure shape are obtained
Block;Alternatively, irregular division can also be carried out to image to be processed, the data block of irregular shape is obtained.
S102 identifies each data block, to obtain the recognition result of each data block.
Specifically, it is content-data block or the data block is background that the recognition result of each data block, which includes the data block,
Data block.
It should be noted that content-data block is the data block for including content-data in the embodiment of the present invention, need
It is bright, it is can wrap in content-data block containing part background data;And block of background data is the number for only including background data
According to block.
For example, when user prints one page image to be processed, in the content portion of the image to be processed
Point, the blank background of certain area is usually contained, that is, includes part background data, the data block obtained at this time is content number
According to block;And there may be the white spaces of certain area on the periphery of the content part in one page image to be processed, for example, document
Margin region, other data are not included in these white spaces, only include background data, at this point, these blank areas
The data block that domain obtains after dividing is block of background data.
S103 obtains the background removal threshold value of each data block according to the recognition result of each data block.
Specifically, needing the recognition result according to each data block after obtaining the recognition result of each data block, obtain every
The background removal threshold value of a data block, in order to be carried out at background removal respectively to each data block in entire image to be processed
Reason.
S104 carries out background removal processing to each data block according to the background removal threshold value of each data block.
It should be noted that the executing subject of S101~S104 can be image processing apparatus, which, which can be located at, is beaten
In the image processing equipments such as print machine, facsimile machine, scanning machine and duplicator.
Currently, there are also a kind of prior art for carrying out background removal processing to image to be processed, the program is by will be whole
A image to be processed divides at least two data blocks, later, obtains a background removal according to first data block therein
Then threshold value carries out unification to all data blocks in entire image to be processed using this obtained background removal threshold value
Background removal processing.Therefore, only image to be processed is placed on the specific region of scanning copy platform, it could be according to be processed
First data block in image gets a relatively reasonable background removal threshold value and goes to carry out background removal processing, so, if to
The placement location of processing image has exceeded the specific region of scanning copy platform, since image processing apparatus scans the presence of blank, this
When, relatively large deviation will appear according to the background removal threshold value that the first data block in image to be processed obtains, it is difficult to to be processed
Image carries out effectively background removal and handles.Also, the program is also simply by a fixed background removal threshold value to entire
Image to be processed carries out background removal processing, and there is also excess processes problems, and in turn caused by image content part lose
The problem of mistake.
In comparison, image processing method provided by the embodiment of the present invention is to divide entire image to be processed,
And all data blocks after division are identified respectively, and obtain the background removal threshold value of each data block, in turn, using every
The background removal threshold value of a data block carries out background removal processing to each data block respectively.Therefore, in the embodiment of the present invention, it is
Background removal processing all is carried out with respective background removal threshold value respectively for all data blocks after division, is dramatically dropped
The influence that scanning blank handles background removal in low image processing apparatus.Therefore, in the embodiment of the present invention, image to be processed
Any position that scanning copy platform can be placed on can carry out effectively background removal processing, without will affect image to be processed
Background removal effect.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two data blocks, then, to each data block
It is identified, to obtain the recognition result of each data block, the recognition result of each data block includes that the data block is content number
It is block of background data according to block or the data block, thus, according to the recognition result of each data block, obtain the back of each data block
Scape removes threshold value, in turn, according to the background removal threshold value of each data block, carries out background removal processing to each data block.This
In inventive embodiments, by the way that image to be processed is divided at least two lesser data blocks, then, after each division
The recognition result of data block obtains each corresponding background removal threshold value of data block, to carry out at background removal
Reason is different from the prior art middle one fixed background removal threshold value of use and carries out at background removal to entire image to be processed
It manages, at least two background removal threshold values has been used to carry out background to the data block in image to be processed respectively in the embodiment of the present invention
Removal processing, also, the background removal threshold value of each data block is to be determined according to the recognition result of data block, thus avoid
The excess processes problem of image, and then the problem of also avoid content partial loss, background removal quality can be effectively improved,
Background removal efficiency is improved, therefore, the embodiment of the present invention is able to solve in the prior art due to carrying out unification to whole image
Background removal handle caused by excess processes problems, and in turn caused by image the problem of content partial loss.
Embodiment two
Image processing method provided by one based on the above embodiment, the embodiment of the present invention in S102 " to each data
Block is identified, to obtain the recognition result of each data block " implementation be specifically described.
Specifically, being identified in the embodiment of the present invention to each data block, to obtain the identification knot of each data block
Fruit can include but is not limited to following two implementation:
Mode 1: obtaining the average value of the pixel value of pixel in each data block, then, according to picture in each data block
The average value of the pixel value of pixel, obtains target pixel points in each data block in the pixel value of vegetarian refreshments and each data block
Number the number of target pixel points in each data block is compared with preset quantity threshold respectively later, in turn,
The data block that the number of target pixel points is less than or equal to quantity threshold is identified as block of background data, alternatively, by target picture
The data block that the number of vegetarian refreshments is greater than quantity threshold is identified as content-data block.
Specifically, target pixel points within the data block in pixel value and data block the pixel value of pixel average value
Between diversity factor be more than or equal to preset discrepancy threshold pixel.
During a concrete implementation, the picture of the pixel value of pixel and pixel in data block in data block is obtained
The diversity factor of the average value of element value, can calculate the picture of pixel in the pixel value and data block of each pixel in data block
The absolute value of the difference of the average value of element value, and using the absolute value as pixel in the pixel value of pixel in data block and data block
The diversity factor of the average value of the pixel value of point.
Specifically, discrepancy threshold and quantity threshold can be preset according to actual needs, the embodiment of the present invention to this not
It is particularly limited.For example, discrepancy threshold can carry out value in the range of [16,64].
Mode 2: obtaining the average value of the pixel value of pixel in each data block, then, according to picture in each data block
The average value of the pixel value of pixel, obtains the side of the pixel value of each data block in the pixel value of vegetarian refreshments and each data block
The variance of the pixel value of each data block is compared with preset variance threshold values, in turn, by pixel value by difference respectively later
Variance be more than or equal to variance threshold values data block be identified as content-data block, alternatively, by the variance of pixel value be less than side
The data block of poor threshold value is identified as block of background data.
It is understood that can include but is not limited at least two pixels in each data block in the embodiment of the present invention
Point, each pixel has respective pixel value, and obtains the average value of the pixel value of pixel in each data block, can first obtain
The pixel value of the point of all pixels included in each data block, then to all pixels point in each data block got
Pixel value carry out mean value calculation.
Specifically, if a data block in image to be processed, the data block is at least two numbers in the embodiment of the present invention
It is N, A according to the processing order number in blockNIt include n in the data block for the average value of the pixel value of pixel in the data block
Pixel, n is the integer greater than 1, and the respective pixel value of n pixel is respectively as follows: P1、P2……Pn, then can be by as follows
The formula (1) obtains the average value A of the pixel value of pixel in the data blockN:
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Embodiment three
Image processing method provided by one based on the above embodiment, the embodiment of the present invention in S103 " according to every number
According to the recognition result of block, obtain the background removal threshold value of each data block " implementation be specifically described.
Specifically, according to the recognition result of each data block, the background for obtaining each data block is gone in the embodiment of the present invention
Except threshold value, following four implementation can include but is not limited to:
Mode A: if data block is block of background data, the background removal threshold of block of background data can be obtained by formula (2)
Value:
Wherein, CNFor the background removal threshold value of data block, N is processing order of the data block at least two data blocks
Number, M is processing order number of the data block in block of background data, CN-1For the background removal threshold of the previous data block of data block
Value, ANFor the average value of the pixel value of pixel in data block.;
If data block is content-data block, the background removal threshold value of content-data block can be obtained by formula (3):
CN=CN-1+O
Wherein, CNFor the background removal threshold value of data block, N is processing order of the data block at least two data blocks
Number, CN-1For the background removal threshold value of the previous data block of data block, O is penalty coefficient.
It should be noted that penalty coefficient O is added value when obtaining each content-data block, if next data block is
Content-data block is obtaining a number then when obtaining the background removal threshold value of next content-data block without cumulative
According to block background removal threshold value when, it is only necessary to add a penalty coefficient O, that is, if during realization, CN-1Included
One penalty coefficient O, then be not further added by penalty coefficient O.Alternatively, if next data block is block of background data, under acquisition
Penalty coefficient O is not considered when the background removal threshold value of one block of background data.
Specifically, the value of penalty coefficient O can background removal degree according to actual needs preset, the present invention is real
Example is applied to this without being particularly limited to.For example, penalty coefficient O can carry out value in the range of [32,80].
Mode B: for each data block, selecting a pixel from data block, using as candidate pixel point, then,
In acquisition data block in the first specified range of candidate pixel point in pixel number and the second specified range of candidate pixel point
The absolute value of difference between pixel number, the first specified range are not overlapped completely with the second specified range, later, will be obtained
Absolute value be compared with preset absolute value threshold value, if obtain absolute value be less than or equal to absolute value threshold value, determine
The pixel value of candidate pixel point is the background removal threshold value of data block.
It is understood that reselecting other pictures in data block if the absolute value obtained is greater than absolute value threshold value
Vegetarian refreshments is as candidate pixel point, until finding to meet when absolute value is less than or equal to the candidate pixel point of absolute value threshold value and being
Only, and using the pixel value of the candidate pixel for the condition that meets point as the background removal threshold value of data block.
During a concrete implementation, the selection of candidate pixel point can be determined according to actual needs, this hair
Bright embodiment is to this without being particularly limited to.
For example, all pixels point in data block can be numbered, the smallest pixel conduct of number can choose
Candidate pixel point;Alternatively, can choose the largest number of pixel as candidate pixel point;Alternatively, can be in all pixels
It is randomly choosed in point.
Alternatively, in another example, it can be selected, can choose most left in data block according to the position of pixel in data block
One pixel on side is as candidate pixel point;Alternatively, can choose a pixel of rightmost in data block as candidate
Pixel;Alternatively, can be randomly choosed in all pixels.
Specifically, when reselecting other pixels in data block as candidate pixel point, it can be according to preset
Rule is selected, and can also be randomly selected.For example, if all pixels point in data block is numbered, at this time
The number for being selected pixel is X, then when reselecting pixel, can choose and number the pixel for being X+Y as new time
Pixel is selected, or also can choose the pixel that number is X-Y is candidate pixel point, wherein Y is the integer greater than 0.
Mode C: default first background removal threshold value, the first background removal threshold value are the background removal threshold of block of background data
Value;If data block is block of background data, using the first background removal threshold value as the background removal threshold value of the data block;And/or in advance
If the second background removal threshold value, the second background removal threshold value is the background removal threshold value of content-data block;If data block is content
Data block, using the second background removal threshold value as the background removal threshold value of the data block.
For example, a fixed background removal threshold value a can be preset for block of background data, it is default for content-data block
One background removal threshold value b, then according to the recognition result of data block, if data block is block of background data, the data block
Background removal threshold value is a;If data block is content-data block, the background removal threshold value of the data block is b.
Mode D: according to the recognition result of each data block, determining target data block, all numbers before target data block
It is block of background data according to block, is content-data block from all data blocks after target data block, then, obtains target data block
The average value of the pixel value of pixel in each data block before, using as every in all data blocks before target data block
The background removal threshold value of a data block;And obtain the background removal threshold value of all block of background data before target data block
The sum of average value, using as the background removal threshold value from each data block in all data blocks after target data block.
For example, if determining that data block 6 is target data block according to recognition result, that is, number in the image to be processed got
It is block of background data according to block 1 to data block 5, is all content-data block from all data blocks after data block 6.Then data block
1 to the background removal threshold value of data block 5 be respectively A1, A2, A3, A4, A5, then from the back of all data blocks after data block 6
Scape removes threshold value
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Example IV
Based on being known otherwise provided by mode 1 in the embodiment of the present invention two to each data block, the present invention is real
It applies example and provides and each data block is identified according to mode 1, it is specific to obtain one kind of recognition result of each data block
Implementation.
Referring to FIG. 2, it is the first pass schematic diagram identified in the embodiment of the present invention to each data block.Such as figure
Shown in 2, each data block is identified, may include following step to obtain the implementation of the recognition result of each data block
It is rapid:
S201 obtains the average value of the pixel value of pixel in each data block.
S202 is calculated in data block between the average value of the pixel value of the pixel value and pixel in the data block of pixel
Diversity factor.
S203, judges whether diversity factor is more than or equal to preset discrepancy threshold;If so, executing S204;If it is not, executing
S205。
The number of S204, target pixel points add one.
S205, the invariable number of target pixel points.
S206 judges whether current pixel point is the last one pixel in the data block;If so, executing S207;If it is not,
Execute S202.
S207 obtains the number of target pixel points.
S208, judges whether the number of target pixel points is less than or equal to preset discrepancy threshold;If so, executing
S209;If it is not, executing S210.
S209 identifies that the data block is block of background data.
S210 identifies that the data block is content-data block.
Each data block is identified specifically, the example above is only to illustrate, to obtain the identification of each data block
As a result a kind of specific implementation, is not intended to limit the invention.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Embodiment five
Based on being known otherwise provided by mode 2 in the embodiment of the present invention two to each data block, the present invention is real
It applies example and provides and each data block is identified according to mode 2, it is specific to obtain one kind of recognition result of each data block
Implementation.
Referring to FIG. 3, it is the second procedure schematic diagram identified in the embodiment of the present invention to each data block.Such as figure
Shown in 3, each data block is identified, may include following step to obtain the implementation of the recognition result of each data block
It is rapid:
S301 obtains the average value of the pixel value of pixel in each data block.
S302, according to the average value of the pixel value of pixel in the pixel value of pixel in data block and the data block, meter
Calculate the variance of the pixel value of data block.
In a concrete implementation, it is illustrated by taking a data block in image to be processed as an example, which exists
Processing order number at least two data blocks is N, QNFor the variance of the pixel value of the pixel of the data block, ANFor the data
The average value of the pixel value of pixel in block includes n pixel in the data block, and n is the integer greater than 1, and n pixel
Respective pixel value is respectively as follows: P1、P2……Pn, then the variance of the pixel value of data block can be calculated by formula (4):
S303, judges whether the variance of pixel value is more than or equal to preset variance threshold values;If so, executing S304;If
It is no, execute S305.
S304 identifies that the data block is content-data block.
S305 identifies that the data block is block of background data.
Each data block is identified specifically, the example above is only to illustrate, to obtain the identification of each data block
As a result a kind of specific implementation, is not intended to limit the invention.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Embodiment six
Based on the mode of the background removal threshold value of each data block of acquisition provided by mode B in the embodiment of the present invention three,
The embodiment of the invention provides a kind of specific implementations for the background removal threshold value that each data block is obtained according to mode B.
Referring to FIG. 4, a process of its background removal threshold value to obtain each data block in the embodiment of the present invention is shown
It is intended to.As shown in figure 4, the implementation for obtaining the background removal threshold value of each data block may include:
S401 selects a unduplicated pixel, as candidate pixel point from each data block.
S402 obtains in each data block pixel number and candidate pixel point in the first specified range of candidate pixel point
The second specified range in difference between pixel number absolute value.
Specifically, the first specified range is not overlapped completely with the second specified range in the embodiment of the present invention.
S403, judges whether the absolute value obtained is less than or equal to preset absolute value threshold value;If so, executing S404;
If it is not, executing S401.
S404 determines that the pixel value of candidate pixel point is the background removal threshold value of data block.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Embodiment seven
Based on the mode of the background removal threshold value of each data block of acquisition provided by mode A in the embodiment of the present invention three,
The embodiment of the invention provides a kind of specific implementations for the background removal threshold value that each data block is obtained according to mode A.
Referring to FIG. 5, it is the recognition result schematic diagram of one page image to be processed.The A in a manner of determines each data block
It is illustrated for background removal threshold value.
As shown in figure 5, the image to be processed divides for multiple data blocks, the arrow direction designation date block in Fig. 5
Therefore processing order according to the direction that arrow indicates, can determine processing order of each data block at least two data blocks
Number.
Below by taking the recognition result of image to be processed shown in fig. 5 as an example, Land use systems A obtains the background of each data block
The process of removal threshold value is specifically described.For convenience of description, by the image to be processed according to processing sequence pair from top to bottom
Each data block is numbered, and the data block of the top is data block 1, is later data block 2, is followed by data block 3 ...
As shown in figure 5, data block 1 is block of background data, the background removal threshold value for obtaining data block 1 can be according to formula
(2) it is calculated, at this point, N is 1, M 1, the background removal threshold value of data block 1 is C1, C1=A1, wherein A1For in data block 1
Pixel pixel value average value.
As shown in figure 5, data block 2 is block of background data, the background removal threshold value for obtaining data block 2 can be according to formula
(2) it is calculated, at this point, N is 2, M 2, the background removal threshold value of data block 2 is C2,
A1For the average value of the pixel value of the pixel in data block 1, A2For the average value of the pixel value of the pixel in data block 2.
As shown in figure 5, data block 3 is block of background data, the background removal threshold value for obtaining data block 3 can be according to formula (2)
It is calculated, at this point, N is 3, M 3, the background removal threshold value of data block 3 is C3,
Wherein, A1For the average value of the pixel value of the pixel in data block 1, A2It is averaged for the pixel value of pixel in data block 2
Value, A3For the average value of the pixel value of the pixel in data block 3.
As shown in figure 5, data block 4 is content-data block, the background removal threshold value for obtaining data block 4 can be according to formula
(3) it is calculated, at this point, N is 4, the background removal threshold value of data block 4 is C4,
Wherein, A1For the average value of the pixel value of the pixel in data block 1, A2It is averaged for the pixel value of pixel in data block 2
Value, A3For the average value of the pixel value of the pixel in data block 3, O is penalty coefficient.
As shown in figure 5, data block 5 is content-data block, the background removal threshold value for obtaining data block 5 can be according to formula
(3) it is calculated, at this point, N is 5, C5=C4+ O, at this point, C4In included a penalty coefficient O, so, at this time no longer
Increase a penalty coefficient O therefore to obtainWherein, A1For in data block 1
Pixel pixel value average value, A2For the average value of the pixel value of the pixel in data block 2, A3For in data block 3
The average value of the pixel value of pixel, O are penalty coefficient.
As shown in figure 5, data block 6 be block of background data, obtain data block 6 background removal threshold value can according to formula (2) into
Row calculates, at this point, N is 6, M 4, the background removal threshold value of data block 6 is C6,
Wherein, A1For the average value of the pixel value of the pixel in data block 1, A2It is averaged for the pixel value of pixel in data block 2
Value, A3For the average value of the pixel value of the pixel in data block 3, A6It is averaged for the pixel value of pixel in data block 6
Value.
In image to be processed as shown in Figure 5, obtain the mode of the background removal threshold value of other untreated data blocks with
Above-mentioned acquisition data block 1 to data block 6 background removal threshold value implementation it is consistent, the embodiment of the present invention no longer carries out this
It repeats.
It is understood that the example above be only to illustrate how the mode A according to provided by S103 in embodiment one into
The acquisition process of the background removal threshold value of each data block of row, is not intended to limit the invention.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Embodiment eight
Based on the mode of the background removal threshold value of each data block of acquisition provided by mode D in the embodiment of the present invention three,
The embodiment of the invention provides a kind of specific implementations for the background removal threshold value that each data block is obtained according to mode D.
Referring to FIG. 6, it is the recognition result schematic diagram of another page image to be processed.Arrow direction indicated number in Fig. 6
According to the processing order of block, therefore, according to the direction that arrow indicates, place of each data block at least two data blocks can be determined
Manage sequence number.For convenience of description, which is numbered each data block according to processing sequence from top to bottom, most
The data block of top is data block 1, is later data block 2, is followed by data block 3 ...
As shown in fig. 6, can determine that data block 4 is target data block according to the recognition result of each data block, that is,
Data block 1, data block 2 and data block 3 before data block 4 are all block of background data, from all data blocks after data block 4
It is all content-data block.
As shown in fig. 6, data block 1, data block 2 and data block 3 are block of background data, then the background removal threshold of data block 1
Value C1For the average value A of the pixel value of pixel in data block 11, that is, C1=A1, then the background removal threshold value C of data block 22For
The average value A of the pixel value of pixel in data block 22, that is, C2=A2, then the background removal threshold value C of data block 33For data
The average value A of the pixel value of pixel in block 33, that is, C3=A3。
As shown in fig. 6, from all data blocks after data block 4 be content-data block.The then background removal threshold of data block 4
It is worth the average value of the sum of background removal threshold value of all block of background data before being data block 4, the i.e. background removal of data block 4
Threshold value is the sum of background removal threshold value, the background removal threshold value of data block 2 and the background removal threshold value of data block 3 of data block 1
Average value, that is,
Specifically, background removal threshold of the background removal threshold value of data block 5 for all block of background data before data block 5
The average value of the sum of value, i.e. the background removal threshold value of data block 5 are that background removal threshold value, the background of data block 2 of data block 1 are gone
Except the average value of the sum of the background removal threshold value of threshold value and data block 3, that is,
That is, from the background removal threshold value of all data blocks after data block 4 be all data block 1 background removal threshold value,
The average value of the sum of the background removal threshold value of the background removal threshold value of data block 2 and data block 3, that is, from target data block it
The back of all block of background data before the background removal threshold value of each data block is target data block in all data blocks afterwards
Scape removes the average value of the sum of threshold value.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Embodiment nine
Image processing method provided by one based on the above embodiment, the embodiment of the present invention in S104 " according to every number
According to the background removal threshold value of block, background removal processing is carried out to each data block " implementation be specifically described.
Specifically, after the background removal threshold value for obtaining each data block, that is, utilizing each data block in present invention implementation
Respective background removal threshold value carries out background removal processing to data block, to realize at the background removal to entire image to be processed
Reason.
It should be noted that in the embodiment of the present invention, after the background removal threshold value that a data block can be obtained, immediately
Background removal processing is carried out to the data block;Alternatively, can also be each in all data blocks for obtaining entire image to be processed
The background removal threshold value of data block and then the background removal threshold value for utilizing acquisition, to every number in entire image to be processed
Background removal processing is carried out respectively according to block.The embodiment of the present invention is to this without being particularly limited to.
During a concrete implementation, according to the background removal threshold value of each data block, each data block is carried out
Background removal processing, can include but is not limited to following two mode:
The first, according to the pixel value of pixel in each data block and the background removal threshold value of each data block, really
Fixed institute within the data block pixel value be greater than data block background removal threshold value pixel, using as the first finger in each data block
Determine pixel, the pixel value of the first specified pixel point in each data block is adjusted to max pixel value.
It is understood that the max pixel value of pixel is related with the binary digit number of pixel, if pixel value is adopted
A pixel value is indicated with i binary digit, then the max pixel value of the pixel is 2i-1。
During a concrete implementation, if pixel indicates a pixel value using 8 binary digits, as
The max pixel value of vegetarian refreshments is 255, and pixel value can be greater than to the pixel value tune of the pixel of the background removal threshold value of data block
Whole is 255.
Second, according to the pixel value of pixel in each data block and the background removal threshold value of each data block, really
Fixed institute within the data block pixel value less than or equal to the background removal threshold value of data block pixel, using as each data block
In the second specified pixel point, keep the pixel value of the second specified pixel point in each data block is constant.
It is understood that being the background removal threshold value according to each data block, to each data in the embodiment of the present invention
Block carries out background removal processing, i.e., all data blocks after dividing to entire image to be processed all carry out background removal processing.
It, can be after carrying out background removal processing to each data block during a concrete implementation, detection should be to
It whether handles in image there is also other untreated data blocks, the embodiment of the present invention is to this without being particularly limited to.
For example, whether the processing order number that can detecte data block is equal to the number of data block in entire image to be processed,
If the processing order number of data block is less than the number of data block in entire image to be processed, it is determined that also deposited in the image to be processed
In other untreated data blocks, does not complete and the background removal of entire image to be processed is handled, continue to untreated data
Block carries out background removal processing;If the processing order number of data block is equal to the number of data block in entire image to be processed, really
Other untreated data blocks are not present in the fixed image to be processed, are completed at the background removal to entire image to be processed
Reason.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Embodiment ten
Based on the two kinds of implementations handled provided by the embodiment of the present invention nine background removal of each data block, originally
Inventive embodiments provide a kind of concrete methods of realizing of background removal processing to each data block.
Referring to FIG. 7, it is the background removal processing schematic in the embodiment of the present invention to each data block, for wait locate
Each data block for managing image, can carry out background according to mode as shown in Figure 7 and go to handle.As shown in fig. 7, this method
Include:
S701, judges whether the pixel value of pixel in data block is greater than the background removal threshold value of the data block;If so, holding
Row S702;If it is not, executing S703.
The pixel value of the pixel is adjusted to max pixel value by S702.
S703 keeps the pixel value of the pixel constant.
It is understood that keeping the pixel value of the pixel constant, that is, without other processing.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Embodiment 11
Based on the mode of the background removal threshold value of each data block of acquisition provided by mode A in the embodiment of the present invention three,
The embodiment of the invention provides a kind of concrete methods of realizing of image procossing.
Referring to FIG. 8, its flow diagram for the embodiment two of image processing method provided by the embodiment of the present invention.
As shown in figure 8, this method comprises:
Image to be processed is divided at least two data blocks by S801.
S802 judges whether each data block is block of background data;If so, executing S803;If it is not, executing S804.
S803 calculates the background removal threshold value of the data block according to formula 2.
Specifically, formula 2 are as follows:
Wherein, CNFor the background removal threshold value of the data block, N is processing of the data block at least two data blocks time
Serial number, M are processing order number of the data block in block of background data, CM-1For the previous block of background data of the data block
Background removal threshold value, ANFor the average value of the pixel value of pixel in the data block
S804 calculates the background removal threshold value of the data block according to formula 3.
Specifically, formula 3 are as follows: CN=CN-1+O。
Wherein, CNFor the background removal threshold value of the data block, N is processing of the data block at least two data blocks time
Serial number, CN-1For the background removal threshold value of the previous data block of the data block, O is penalty coefficient.
S805 carries out background removal processing to each data block according to the background removal threshold value of each data block.
S806 judges whether there are also other untreated data blocks in image to be processed;If so, executing S802;If it is not, knot
Line journey.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, then, according to every
The recognition result of data block after a division obtains each corresponding background removal threshold value of data block, to be carried on the back
Scape removal processing is different from the prior art middle one fixed background removal threshold value of use and carries out background to entire image to be processed
Removal processing, used in the embodiment of the present invention at least two background removal threshold values respectively to the data block in image to be processed into
The processing of row background removal, also, the background removal threshold value of each data block is determined according to the recognition result of data block, thus
The problem of avoiding the excess processes problem of image, and then also avoiding content partial loss can effectively improve background and go
Except quality, improve background removal efficiency, therefore, the embodiment of the present invention be able to solve in the prior art due to whole image into
Excess processes problem caused by the unified background removal of row is handled, and in turn caused by image content partial loss ask
Topic.
Embodiment 12
Image processing method provided by one based on the above embodiment, the embodiment of the present invention, which further provides, realizes above-mentioned side
The Installation practice of each step and method in method embodiment.
Referring to FIG. 9, its functional block diagram for image processing apparatus provided by the embodiment of the present invention.As shown in figure 9,
The device includes:
Division unit 901, for image to be processed to be divided at least two data blocks;
Recognition unit 902, for being identified to each data block, to obtain the recognition result of each data block, each
It is content-data block or the data block is block of background data that the recognition result of data block, which includes the data block,;
Acquiring unit 903 obtains the background removal of each data block for the recognition result according to each data block
Threshold value;
Processing unit 904 carries out background removal to each data block for the background removal threshold value according to each data block
Processing.
Specifically, recognition unit 902 is specifically used in the embodiment of the present invention:
Obtain the average value of the pixel value of pixel in each data block;
According to the average value of the pixel value of pixel in the pixel value of pixel in each data block and each data block,
Obtain the number of target pixel points in each data block, target pixel points are pixel in institute within the data block pixel value and data block
Diversity factor between the average value of the pixel value of point is more than or equal to the pixel of preset discrepancy threshold;
The number of target pixel points in each data block is compared with preset quantity threshold respectively;
The data block that the number of target pixel points is less than or equal to quantity threshold is identified as block of background data, alternatively,
The data block that the number of target pixel points is greater than quantity threshold is identified as content-data block.
Specifically, recognition unit 902 is specifically used in the embodiment of the present invention:
Obtain the average value of the pixel value of pixel in each data block;
According to the average value of the pixel value of pixel in the pixel value of pixel in each data block and each data block,
Obtain the variance of the pixel value of each data block;
The variance of the pixel value of each data block is compared with preset variance threshold values respectively;
The data block that the variance of pixel value is more than or equal to variance threshold values is identified as content-data block, alternatively, by picture
The data block that the variance of plain value is less than variance threshold values is identified as block of background data.
Specifically, acquiring unit 903 is specifically used in the embodiment of the present invention:
If data block obtains the background removal threshold value of the data block by following formula for block of background data:
Wherein, CNFor the background removal threshold value of the data block, N is processing of the data block at least two data blocks time
Serial number, M are processing order number of the data block in block of background data, CM-1For the previous block of background data of the data block
Background removal threshold value, ANFor the average value of the pixel value of pixel in the data block;
If data block obtains the background removal threshold value of the data block by following formula for content-data block:
CN=CN-1+O
Wherein, CNFor the background removal threshold value of the data block, N is processing of the data block at least two data blocks time
Serial number, CN-1For the background removal threshold value of the previous data block of the data block, O is penalty coefficient.
Specifically, acquiring unit 903 is specifically used in the embodiment of the present invention:
For each data block, a pixel is selected from data block, using as candidate pixel point;
Obtain the second finger of pixel number and candidate pixel point in the first specified range of candidate pixel point in data block
Determine the absolute value of the difference in range between pixel number, the first specified range is not overlapped completely with the second specified range;
Absolute value is compared with preset absolute value threshold value;
If absolute value is less than or equal to absolute value threshold value, determine that the pixel value of candidate pixel point goes for the background of data block
Except threshold value.
Specifically, acquiring unit 903 is specifically used in the embodiment of the present invention:
Default first background removal threshold value, the first background removal threshold value are the background removal threshold value of block of background data;If number
It is block of background data according to block, using the first background removal threshold value as the background removal threshold value of the data block;And/or
Default second background removal threshold value, the second background removal threshold value are the background removal threshold value of content-data block;If number
It is content-data block according to block, using the second background removal threshold value as the background removal threshold value of the data block.
During a concrete implementation, acquiring unit 903 is specifically used for:
According to the recognition result of each data block, target data block is determined, all data blocks before target data block are
Block of background data is content-data block from all data blocks after target data block;
The average value for obtaining the pixel value of pixel in each data block before target data block, using as target data
The background removal threshold value of each data block in all data blocks before block;And all back before acquisition target data block
The average value of the sum of the background removal threshold value of scape data block, using as from all data blocks after target data block it is each not
The background removal threshold value of the data block of processing.
Specifically, processing unit 904 is used in the embodiment of the present invention:
According to the pixel value of pixel in each data block and the background removal threshold value of each data block, number where determining
It is greater than the pixel of the background removal threshold value of data block according to pixel value in block, using as the first specified pixel in each data block
Point;The pixel value of first specified pixel point in each data block is adjusted to max pixel value;Alternatively,
According to the pixel value of pixel in each data block and the background removal threshold value of each data block, number where determining
It is less than or equal to the pixel of the background removal threshold value of data block according to pixel value in block, to refer to as in each data block second
Determine pixel;Keep the pixel value of the second specified pixel point in each data block constant.
Method shown in FIG. 1 is able to carry out by each unit in this present embodiment, the part that the present embodiment is not described in detail,
It can refer to the related description to Fig. 1.
A technical solution in the embodiment of the present invention has the advantages that
In the embodiment of the present invention, image to be processed is divided at least two by the division unit in image processing apparatus
Data block, then, the recognition unit in image processing apparatus identify each data block, to obtain the knowledge of each data block
Not as a result, it is content-data block or the data block is block of background data that the recognition result of each data block, which includes the data block,
To which, acquiring unit in image processing apparatus is according to the recognition result of each data block, the background for obtaining each data block is gone
Except threshold value, in turn, processing unit in image processing apparatus is according to the background removal threshold value of each data block, to each data block
Carry out background removal processing.In the embodiment of the present invention, by the way that image to be processed is divided at least two lesser data blocks, so
Afterwards, each corresponding background removal threshold value of data block is obtained according to the recognition result of the data block after each division,
To carry out background removal processing, middle one fixed background removal threshold value of use is different from the prior art to entire figure to be processed
As carrying out background removal processing, use at least two background removal threshold values respectively in image to be processed in the embodiment of the present invention
Data block carry out background removal processing, also, the background removal threshold value of each data block is the recognition result according to data block
It is determining, thus avoid the excess processes problem of image, and then the problem of also avoid content partial loss, it can be effectively
Background removal quality is improved, improves background removal efficiency, therefore, the embodiment of the present invention is able to solve in the prior art due to right
Whole image carries out excess processes problem caused by unified background removal is handled, and content part in caused image in turn
The problem of loss.
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 by the present invention, 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, for example, multiple units or group
Part 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, device or unit it is indirect
Coupling or communication connection 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, the functional units in various embodiments of the present invention may be integrated into 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 hardware adds SFU software functional unit.
The above-mentioned integrated unit being realized in the form of SFU software functional unit can store and computer-readable deposit at one
In storage media.Above-mentioned SFU software functional unit is stored in a storage medium, including some instructions are used so that a computer
It is each that device (can be personal computer, server or network equipment etc.) or processor (Processor) execute the present invention
The part steps of embodiment the method.And storage medium above-mentioned includes: USB flash disk, mobile hard disk, read-only memory (Read-
Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic or disk etc. it is various
It can store the medium of program code.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Within mind and principle, any modification, equivalent substitution, improvement and etc. done be should be included within the scope of the present invention.
Claims (12)
1. a kind of image processing method, which is characterized in that the described method includes:
Image to be processed is divided at least two data blocks;
Each data block is identified, to obtain the recognition result of each data block, each data block
It is content-data block or the data block is block of background data that recognition result, which includes the data block,;
According to the recognition result of each data block, the background removal threshold value of each data block is obtained;
According to the background removal threshold value of each data block, background removal processing is carried out to each data block;
According to the recognition result of each data block, the background removal threshold value of each data block is obtained, comprising:
If the data block obtains the background removal threshold value of the data block by formula as described below for block of background data:
Wherein, CNFor the background removal threshold value of the data block, N is place of the data block at least two data block
Sequence number is managed, M is processing order number of the data block in block of background data, CM-1For the previous background of the data block
The background removal threshold value of data block, ANFor the average value of the pixel value of pixel in the data block;
If the data block obtains the background removal threshold value of the data block by formula as described below for content-data block:
CN=CN-1+O
Wherein, CNFor the background removal threshold value of the data block, N is place of the data block at least two data block
Manage sequence number, CN-1For the background removal threshold value of the previous data block of the data block, O is penalty coefficient.
2. a kind of image processing method, which is characterized in that the described method includes:
Image to be processed is divided at least two data blocks;
Each data block is identified, to obtain the recognition result of each data block, each data block
It is content-data block or the data block is block of background data that recognition result, which includes the data block,;
According to the recognition result of each data block, the background removal threshold value of each data block is obtained;
According to the background removal threshold value of each data block, background removal processing is carried out to each data block;
According to the recognition result of each data block, the background removal threshold value of each data block is obtained, comprising:
For each data block, a pixel is selected from the data block, using as candidate pixel point;
Obtain pixel number and the candidate pixel point in the first specified range of the point of candidate pixel described in the data block
The second specified range in difference between pixel number absolute value, first specified range and the described second specified model
It encloses and is not overlapped completely;
The absolute value is compared with preset absolute value threshold value;
If the absolute value is less than or equal to the absolute value threshold value, determine that the pixel value of the candidate pixel point is the number
According to the background removal threshold value of block.
3. a kind of image processing method, which is characterized in that the described method includes:
Image to be processed is divided at least two data blocks;
Each data block is identified, to obtain the recognition result of each data block, each data block
It is content-data block or the data block is block of background data that recognition result, which includes the data block,;
According to the recognition result of each data block, the background removal threshold value of each data block is obtained;
According to the background removal threshold value of each data block, background removal processing is carried out to each data block;
According to the recognition result of each data block, the background removal threshold value of each data block is obtained, comprising:
According to the recognition result of each data block, target data block, all data before the target data block are determined
Block is block of background data, is content-data block from all data blocks after the target data block;
Each of before obtaining the target data block in the data block pixel value of pixel average value, using as described
The background removal threshold value of each data block in all data blocks before target data block;And obtain the number of targets
According to the average value of the sum of the background removal threshold value of all block of background data before block, as from after the target data block
All data blocks in each data block background removal threshold value.
4. method according to claim 1-3, which is characterized in that each data block is identified, with
Obtain the recognition result of each data block, comprising:
Obtain the average value of the pixel value of pixel in each data block;
According in the pixel value of pixel in each data block and each data block the pixel value of pixel it is flat
Mean value, obtains the number of target pixel points in each data block, the target pixel points be pixel value within the data block
It is more than or equal to preset discrepancy threshold with the diversity factor in the data block between the average value of the pixel value of pixel
Pixel;
The number of target pixel points in each data block is compared with preset quantity threshold respectively;
The data block that the number of the target pixel points is less than or equal to the quantity threshold is identified as block of background data, or
The data block that the number of the target pixel points is greater than the quantity threshold is identified as content-data block by person.
5. method according to claim 1-3, which is characterized in that each data block is identified, with
Obtain the recognition result of each data block, comprising:
Obtain the average value of the pixel value of pixel in each data block;
According in the pixel value of pixel in each data block and each data block the pixel value of pixel it is flat
Mean value obtains the variance of the pixel value of each data block;
The variance of the pixel value of each data block is compared with preset variance threshold values respectively;
The data block that the variance of the pixel value is more than or equal to the variance threshold values is identified as content-data block, alternatively,
The data block that the variance of the pixel value is less than the variance threshold values is identified as block of background data.
6. method according to claim 1-3, which is characterized in that according to the background removal of each data block
Threshold value carries out background removal processing to each data block, comprising:
According to the background removal threshold value of the pixel value of pixel and each data block in each data block, institute is determined
Pixel value is greater than the pixel of the background removal threshold value of the data block within the data block, using as in each data block the
Specified pixel point;The pixel value of first specified pixel point in each data block is adjusted to max pixel value;Alternatively,
According to the background removal threshold value of the pixel value of pixel and each data block in each data block, institute is determined
Pixel value is less than or equal to the pixel of the background removal threshold value of the data block within the data block, using as each number
According to the second specified pixel point in block;Keep the pixel value of the second specified pixel point in each data block constant.
7. a kind of image processing apparatus, which is characterized in that described device includes:
Division unit, for image to be processed to be divided at least two data blocks;
Recognition unit, for being identified to each data block, to obtain the recognition result of each data block, each
It is content-data block or the data block is block of background data that the recognition result of the data block, which includes the data block,;
Acquiring unit obtains the background removal threshold of each data block for the recognition result according to each data block
Value;
Processing unit carries out background to each data block and goes for the background removal threshold value according to each data block
Except processing;
The acquiring unit, is specifically used for:
If the data block obtains the background removal threshold value of the data block by formula as described below for block of background data:
Wherein, CNFor the background removal threshold value of the data block, N is place of the data block at least two data block
Sequence number is managed, M is processing order number of the data block in block of background data, CM-1For the previous background of the data block
The background removal threshold value of data block, ANFor the average value of the pixel value of pixel in the data block;
If the data block obtains the background removal threshold value of the data block by formula as described below for content-data block:
CN=CN-1+O
Wherein, CNFor the background removal threshold value of the data block, N is place of the data block at least two data block
Manage sequence number, CN-1For the background removal threshold value of the previous data block of the data block, O is penalty coefficient.
8. a kind of image processing apparatus, which is characterized in that described device includes:
Division unit, for image to be processed to be divided at least two data blocks;
Recognition unit, for being identified to each data block, to obtain the recognition result of each data block, each
It is content-data block or the data block is block of background data that the recognition result of the data block, which includes the data block,;
Acquiring unit obtains the background removal threshold of each data block for the recognition result according to each data block
Value;
Processing unit carries out background to each data block and goes for the background removal threshold value according to each data block
Except processing;
The acquiring unit, is specifically used for:
For each data block, a pixel is selected from the data block, using as candidate pixel point;
Obtain pixel number and the candidate pixel point in the first specified range of the point of candidate pixel described in the data block
The second specified range in difference between pixel number absolute value, first specified range and the described second specified model
It encloses and is not overlapped completely;
The absolute value is compared with preset absolute value threshold value;
If the absolute value is less than or equal to the absolute value threshold value, determine that the pixel value of the candidate pixel point is the number
According to the background removal threshold value of block.
9. a kind of image processing apparatus, which is characterized in that described device includes:
Division unit, for image to be processed to be divided at least two data blocks;
Recognition unit, for being identified to each data block, to obtain the recognition result of each data block, each
It is content-data block or the data block is block of background data that the recognition result of the data block, which includes the data block,;
Acquiring unit obtains the background removal threshold of each data block for the recognition result according to each data block
Value;
Processing unit carries out background to each data block and goes for the background removal threshold value according to each data block
Except processing;
The acquiring unit, is specifically used for:
According to the recognition result of each data block, target data block, all data before the target data block are determined
Block is block of background data, is content-data block from all data blocks after the target data block;
Each of before obtaining the target data block in the data block pixel value of pixel average value, using as described
The background removal threshold value of each data block in all data blocks before target data block;And obtain the number of targets
According to the average value of the sum of the background removal threshold value of all block of background data before block, as from after the target data block
All data blocks in each data block background removal threshold value.
10. according to the described in any item devices of claim 7-9, which is characterized in that the recognition unit is specifically used for:
Obtain the average value of the pixel value of pixel in each data block;
According in the pixel value of pixel in each data block and each data block the pixel value of pixel it is flat
Mean value, obtains the number of target pixel points in each data block, the target pixel points be pixel value within the data block
It is more than or equal to preset discrepancy threshold with the diversity factor in the data block between the average value of the pixel value of pixel
Pixel;
The number of target pixel points in each data block is compared with preset quantity threshold respectively;
The data block that the number of the target pixel points is less than or equal to the quantity threshold is identified as block of background data, or
The data block that the number of the target pixel points is greater than the quantity threshold is identified as content-data block by person.
11. according to the described in any item devices of claim 7-9, which is characterized in that the recognition unit is specifically used for:
Obtain the average value of the pixel value of pixel in each data block;
According in the pixel value of pixel in each data block and each data block the pixel value of pixel it is flat
Mean value obtains the variance of the pixel value of each data block;
The variance of the pixel value of each data block is compared with preset variance threshold values respectively;
The data block that the variance of the pixel value is more than or equal to the variance threshold values is identified as content-data block, alternatively,
The data block that the variance of the pixel value is less than the variance threshold values is identified as block of background data.
12. according to the described in any item devices of claim 7-9, which is characterized in that the processing unit is specifically used for:
According to the background removal threshold value of the pixel value of pixel and each data block in each data block, institute is determined
Pixel value is greater than the pixel of the background removal threshold value of the data block within the data block, using as in each data block the
Specified pixel point;The pixel value of first specified pixel point in each data block is adjusted to max pixel value;Alternatively,
According to the background removal threshold value of the pixel value of pixel and each data block in each data block, institute is determined
Pixel value is less than or equal to the pixel of the background removal threshold value of the data block within the data block, using as each number
According to the second specified pixel point in block;Keep the pixel value of the second specified pixel point in each data block constant.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610643274.1A CN106251358B (en) | 2016-08-08 | 2016-08-08 | A kind of image processing method and device |
PCT/CN2017/080898 WO2018028234A1 (en) | 2016-08-08 | 2017-04-18 | Image processing method and apparatus |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610643274.1A CN106251358B (en) | 2016-08-08 | 2016-08-08 | A kind of image processing method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106251358A CN106251358A (en) | 2016-12-21 |
CN106251358B true CN106251358B (en) | 2019-05-07 |
Family
ID=58078763
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610643274.1A Active CN106251358B (en) | 2016-08-08 | 2016-08-08 | A kind of image processing method and device |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN106251358B (en) |
WO (1) | WO2018028234A1 (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106251358B (en) * | 2016-08-08 | 2019-05-07 | 珠海赛纳打印科技股份有限公司 | A kind of image processing method and device |
CN109785562B (en) * | 2018-12-29 | 2023-08-15 | 华中光电技术研究所(中国船舶重工集团有限公司第七一七研究所) | Vertical photoelectric ground threat alert system and suspicious target identification method |
CN110443794B (en) * | 2019-05-29 | 2020-12-18 | 腾讯科技(深圳)有限公司 | Pathological image-based image state determination method, device and system |
CN117671497B (en) * | 2023-12-04 | 2024-05-28 | 广东筠诚建筑科技有限公司 | Engineering construction waste classification method and device based on digital images |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7986831B2 (en) * | 2006-08-29 | 2011-07-26 | Sony Corporation | Image processing apparatus, image processing method and computer program |
CN102982545A (en) * | 2012-11-22 | 2013-03-20 | 清华大学深圳研究生院 | Image depth estimation method |
CN103235948A (en) * | 2013-04-22 | 2013-08-07 | 中山大学 | Adaptive threshold binarization method for two-dimensional barcode |
CN104504686A (en) * | 2014-12-04 | 2015-04-08 | 哈尔滨工程大学 | Hyper-spectral image abnormity detection method adopting local self-adaptive threshold segmentation |
CN104657969A (en) * | 2013-11-25 | 2015-05-27 | 方正国际软件(北京)有限公司 | Method and system for removing image background |
CN104917975A (en) * | 2015-06-01 | 2015-09-16 | 北京空间机电研究所 | Adaptive automatic exposure method based on object characteristics |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP1806697B1 (en) * | 2006-01-10 | 2016-08-10 | Microsoft Technology Licensing, LLC | Segmenting image elements |
JP4808267B2 (en) * | 2009-05-27 | 2011-11-02 | シャープ株式会社 | Image processing apparatus, image forming apparatus, image processing method, computer program, and recording medium |
CN106251358B (en) * | 2016-08-08 | 2019-05-07 | 珠海赛纳打印科技股份有限公司 | A kind of image processing method and device |
-
2016
- 2016-08-08 CN CN201610643274.1A patent/CN106251358B/en active Active
-
2017
- 2017-04-18 WO PCT/CN2017/080898 patent/WO2018028234A1/en active Application Filing
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7986831B2 (en) * | 2006-08-29 | 2011-07-26 | Sony Corporation | Image processing apparatus, image processing method and computer program |
CN102982545A (en) * | 2012-11-22 | 2013-03-20 | 清华大学深圳研究生院 | Image depth estimation method |
CN103235948A (en) * | 2013-04-22 | 2013-08-07 | 中山大学 | Adaptive threshold binarization method for two-dimensional barcode |
CN104657969A (en) * | 2013-11-25 | 2015-05-27 | 方正国际软件(北京)有限公司 | Method and system for removing image background |
CN104504686A (en) * | 2014-12-04 | 2015-04-08 | 哈尔滨工程大学 | Hyper-spectral image abnormity detection method adopting local self-adaptive threshold segmentation |
CN104917975A (en) * | 2015-06-01 | 2015-09-16 | 北京空间机电研究所 | Adaptive automatic exposure method based on object characteristics |
Non-Patent Citations (2)
Title |
---|
"Anomaly detection using an adaptive algorithm for cstimating mixtures of background in hyperspectral images";Ariel orfaig等;《2012 IEEE 27th convention of Electrical and Electronics Engineers in Isracl》;20121117;第1-5页 |
"基于改进自动语义标注的图像识别方法";陆伟艳;《科技通报》;20130228;第29卷(第2期);第76-78页 |
Also Published As
Publication number | Publication date |
---|---|
CN106251358A (en) | 2016-12-21 |
WO2018028234A1 (en) | 2018-02-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106251358B (en) | A kind of image processing method and device | |
Li et al. | A new cost function for spatial image steganography | |
JP4358193B2 (en) | Automatic image cropping method | |
WO2013028477A1 (en) | Method for segmenting a composite image | |
CN106991422B (en) | Character cutting method and device, computer readable storage medium and electronic equipment | |
WO2014166377A1 (en) | Image interest point detection method and device | |
CN104580883B (en) | Image processing apparatus, image processing method and recording medium | |
EP2765555B1 (en) | Image evaluation device, image selection device, image evaluation method, recording medium, and program | |
EP2782065B1 (en) | Image-processing device removing encircling lines for identifying sub-regions of image | |
CN102171723A (en) | Method for red-eye detection | |
EP3462378B1 (en) | System and method of training a classifier for determining the category of a document | |
CN106997580B (en) | Picture processing method and device | |
JP5984880B2 (en) | Image processing device | |
EP2966613A1 (en) | Method and apparatus for generating a super-resolved image from an input image | |
CN112785651A (en) | Method and apparatus for determining relative pose parameters | |
CN105100616A (en) | Image processing method and electronic equipment | |
US9811877B2 (en) | Image processing device selecting arrangement method for generating arranged image data | |
CN110738227A (en) | Model training method and device, recognition method, storage medium and electronic equipment | |
JP6365894B2 (en) | Image reading device | |
JP2014239367A (en) | Image processor, image processing method, and program | |
JP6237291B2 (en) | Information processing apparatus, color interpolation method, program, and table generation method | |
JP2012095173A (en) | Image processing device, image formation device, and image processing method | |
JP4967488B2 (en) | Code image processing method, code image processing apparatus, and code image processing program | |
JP2006042367A (en) | Detection and elimination of halftone | |
CN116468611B (en) | Image stitching method, device, equipment and storage medium |
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 | ||
TR01 | Transfer of patent right |
Effective date of registration: 20201125 Address after: 519060 A, building 3, building 01, 3883 Zhuhai Avenue, Zhuhai, Guangdong, China Patentee after: ZHUHAI PANTUM ELECTRONICS Co.,Ltd. Address before: 519060 Area A, 7th Floor, Building 01, No. 3883, Zhuhai Avenue, Xiangzhou District, Zhuhai City, Guangdong Province Patentee before: Zhuhai Saine Printing Technology Co.,Ltd. |
|
TR01 | Transfer of patent right |