CN109255784A - Image processing method and device, electronic equipment and storage medium - Google Patents
Image processing method and device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN109255784A CN109255784A CN201811068237.8A CN201811068237A CN109255784A CN 109255784 A CN109255784 A CN 109255784A CN 201811068237 A CN201811068237 A CN 201811068237A CN 109255784 A CN109255784 A CN 109255784A
- Authority
- CN
- China
- Prior art keywords
- pixel
- image
- connection
- value
- multiple pixels
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Abstract
This disclosure relates to a kind of image processing method and device, electronic equipment and storage medium, which comprises pre-processed to the pixel of image to be processed, obtain the first image;First image is inputted into first network, obtains pixel connection figure;According to pixel connection figure, the target area in the first image is determined.Image processing method according to an embodiment of the present disclosure, by being pre-processed to pixel, improve treatment effeciency, target area is determined by pixel connection figure, reduce the demand to training sample, also, the parameters such as the first probability value and the first connection value in pixel connection figure reduce the complexity of processing, improve the accuracy of identification of target area.
Description
Technical field
This disclosure relates to field of computer technology more particularly to a kind of image processing method and device, electronic equipment and deposit
Storage media.
Background technique
The target area in image is obtained, and then the information in target area can be analyzed.For example, in cell image
In equal medical images, the region where nucleus need to be determined, and analyze the information in nuclear area, to judge lesion
Degree, for treatment foundation is provided.In the related art, more due to attributes such as the shape of the target areas such as nucleus and colors
Sample determines the ineffective of target area using traditional image analysis method, and due to the complexity of cell image, institute
The training sample needed is more, and the speed of service is slower.
Summary of the invention
The present disclosure proposes a kind of image processing method and devices, electronic equipment and storage medium.
According to the one side of the disclosure, a kind of image processing method is provided, comprising:
Multiple pixels of image to be processed are pre-processed, the first image is obtained;
The first image input first network is handled, the pixel of multiple pixels of the first image is obtained
Connection figure;
According to the pixel connection figure of multiple pixels of the first image, one or more in the first image is determined
A target area.
Image processing method according to an embodiment of the present disclosure improves processing effect by pre-processing to pixel
Rate determines target area by pixel connection figure, reduces the demand to training sample, also, in pixel connection figure
The parameters such as one probability value and the first connection value reduce the complexity of processing, improve the accuracy of identification of target area.
In one possible implementation, the pixel connection figure of first pixel includes first pixel
Position, first pixel belong to the first probability value of target area and first pixel is directed to and first picture
First connection value of the adjacent multiple pixels of vegetarian refreshments, wherein first pixel is multiple pixels of the first image
Point in any one.
In one possible implementation, multiple pixels of image to be processed are pre-processed, obtains the first figure
Picture, comprising:
Multiple pixels of the image to be processed are normalized, the first image is obtained.
In this way, the diversity of image to be processed can be reduced, treatment effeciency is improved.
In one possible implementation, the pixel connection figure of the second pixel includes the position of second pixel
It sets, second pixel belongs to the second probability value of target area and second pixel is directed to and second pixel
Second connection value of the adjacent multiple pixels of point,
Wherein, second pixel is any one in multiple pixels adjacent with first pixel, with
The adjacent multiple pixels of second pixel include first pixel,
Wherein, according to the pixel connection figure of multiple pixels of the first image, one in the first image is determined
A or multiple target areas, comprising:
According to first probability value, second probability value, first pixel for second pixel
First connection value and/or second pixel are directed to the second connection value of first pixel, determine first pixel
The classification of point;
According to the classification of multiple pixels of the first image, the boundary of one or more target area is determined;
According to the boundary of one or more of target areas, one or more of target areas are determined.
In this way, it can be connected to value by the first probability value, the second probability value, the first connection value and second and determine
The classification of one pixel, and determine the boundary of target area, the complexity on identification object region boundary is reduced, identification is improved
The treatment effeciency of target area.
In one possible implementation, according to first probability value, second probability value, first pixel
Point connects for the first connection value of second pixel and/or second pixel for the second of first pixel
Logical value, determines the classification of first pixel, comprising:
First probability value be greater than or equal to probability threshold value, and second probability value be less than probability threshold value the case where
Under, the classification of first pixel is determined as to the boundary pixel point of target area.
In this way, the classification that the first pixel can be determined by the first probability value and the second probability value, reduces
The complexity for determining the classification of the first pixel, improves treatment effeciency.
In one possible implementation, according to first probability value, second probability value, first pixel
Point connects for the first connection value of second pixel and/or second pixel for the second of first pixel
Logical value, determines the classification of first pixel, comprising:
In the case where the first connection value with described second is connected to value and indicates disconnected situation, first pixel is determined
Point is not connected to second pixel;
It is greater than or equal to the probability threshold value, and first pixel and second pixel in first probability value
In the disconnected situation of point, the classification of first pixel is determined as to the boundary pixel point of target area.
In this way, value can be connected to second by the first probability value, the first connection value determine the first pixel
Classification reduces the complexity for determining the classification of the first pixel, improves treatment effeciency.
In one possible implementation, the method also includes:
One or more target areas addition number mark respectively in the first image, obtains the first image
Regional prediction figure.
In one possible implementation, the method also includes:
According to the number mark of the regional prediction figure, pixel distribution in respectively one or more target areas with
The number identifies corresponding rgb value, obtains the second colored image.
It in this way, can be by distributing rgb value for target area, to improve the differentiation between different target region
Degree improves the effect of visualization of the second image.
In one possible implementation, multiple pixels of the image to be processed are normalized, are obtained
Obtain the first image, comprising:
At least one of the rgb values of multiple pixels of image to be processed, sum of the grayscale values contrast are normalized
Processing obtains the first image.
In one possible implementation, the method also includes:
The first network is trained by the training set being made of multiple sample images after normalized.
According to another aspect of the present disclosure, a kind of image processing apparatus is provided, comprising:
Preprocessing module is pre-processed for multiple pixels to image to be processed, obtains the first image;
Pixel connection figure obtains module, for handling the first image input first network, obtains described the
The pixel connection figure of multiple pixels of one image;
Target area determining module determines institute for the pixel connection figure according to multiple pixels of the first image
State one or more target areas in the first image.
In one possible implementation, the pixel connection figure of first pixel includes first pixel
Position, first pixel belong to the first probability value of target area and first pixel is directed to and first picture
First connection value of the adjacent multiple pixels of vegetarian refreshments, wherein first pixel is multiple pixels of the first image
Point in any one.
In one possible implementation, the preprocessing module is further used for:
Multiple pixels of the image to be processed are normalized, the first image is obtained.
In one possible implementation, the pixel connection figure of the second pixel includes the position of second pixel
It sets, second pixel belongs to the second probability value of target area and second pixel is directed to and second pixel
Second connection value of the adjacent multiple pixels of point,
Wherein, second pixel is any one in multiple pixels adjacent with first pixel, with
The adjacent multiple pixels of second pixel include first pixel,
Wherein, the target area determining module is further used for:
According to first probability value, second probability value, first pixel for second pixel
First connection value and/or second pixel are directed to the second connection value of first pixel, determine first pixel
The classification of point;
According to the classification of multiple pixels of the first image, the boundary of one or more target area is determined;
According to the boundary of one or more of target areas, one or more of target areas are determined.
In one possible implementation, the target area determining module is further used for:
First probability value be greater than or equal to probability threshold value, and second probability value be less than probability threshold value the case where
Under, the classification of first pixel is determined as to the boundary pixel point of target area.
In one possible implementation, the target area determining module is further used for:
In the case where the first connection value with described second is connected to value and indicates disconnected situation, first pixel is determined
Point is not connected to second pixel;
It is greater than or equal to the probability threshold value, and first pixel and second pixel in first probability value
In the disconnected situation of point, the classification of first pixel is determined as to the boundary pixel point of target area.
In one possible implementation, described device further include:
Adding module is obtained for being respectively one or more target areas addition number mark in the first image
Obtain the regional prediction figure of the first image.
In one possible implementation, described device further include:
Distribution module, for being identified according to the number of the regional prediction figure, in respectively one or more target areas
Pixel distribution rgb value corresponding with the digital mark, obtain the second image of colour.
In one possible implementation, the preprocessing module is further used for:
At least one of the rgb values of multiple pixels of image to be processed, sum of the grayscale values contrast are normalized
Processing obtains the first image.
In one possible implementation, described device further include:
Training module, for training described by the training set being made of multiple sample images after normalized
One network.
According to another aspect of the present disclosure, a kind of electronic equipment is provided, comprising:
Processor;
Memory for storage processor executable instruction;
Wherein, the processor is configured to: execute above-mentioned image processing method.
According to another aspect of the present disclosure, a kind of computer readable storage medium is provided, computer journey is stored thereon with
Sequence instruction, the computer program instructions realize above-mentioned image processing method when being executed by processor.
It should be understood that above general description and following detailed description is only exemplary and explanatory, rather than
Limit the disclosure.
According to below with reference to the accompanying drawings to detailed description of illustrative embodiments, the other feature and aspect of the disclosure will become
It is clear.
Detailed description of the invention
The drawings herein are incorporated into the specification and forms part of this specification, and those figures show meet this public affairs
The embodiment opened, and together with specification it is used to illustrate the technical solution of the disclosure.
Fig. 1 shows the flow chart of the image processing method according to the embodiment of the present disclosure;
Fig. 2 shows the flow charts according to the image processing method of the embodiment of the present disclosure;
Fig. 3 shows the schematic diagram of the regional prediction figure according to the image processing method of the embodiment of the present disclosure;
Fig. 4 shows the flow chart of the image processing method according to the embodiment of the present disclosure;
Fig. 5 shows the flow chart of the image processing method according to the embodiment of the present disclosure;
Fig. 6 shows the application schematic diagram of the image processing method according to the embodiment of the present disclosure;
Fig. 7 shows the block diagram of the image processing apparatus according to the embodiment of the present disclosure;
Fig. 8 shows the block diagram of the image processing apparatus according to the embodiment of the present disclosure;
Fig. 9 shows the block diagram of the electronic equipment according to the embodiment of the present disclosure;
Figure 10 shows the block diagram of the electronic equipment according to the embodiment of the present disclosure.
Specific embodiment
Various exemplary embodiments, feature and the aspect of the disclosure are described in detail below with reference to attached drawing.It is identical in attached drawing
Appended drawing reference indicate element functionally identical or similar.Although the various aspects of embodiment are shown in the attached drawings, remove
It non-specifically points out, it is not necessary to attached drawing drawn to scale.
Dedicated word " exemplary " means " being used as example, embodiment or illustrative " herein.Here as " exemplary "
Illustrated any embodiment should not necessarily be construed as preferred or advantageous over other embodiments.
The terms "and/or", only a kind of incidence relation for describing affiliated partner, indicates that there may be three kinds of passes
System, for example, A and/or B, can indicate: individualism A exists simultaneously A and B, these three situations of individualism B.In addition, herein
Middle term "at least one" indicate a variety of in any one or more at least two any combination, it may for example comprise A,
B, at least one of C can indicate to include any one or more elements selected from the set that A, B and C are constituted.
In addition, giving numerous details in specific embodiment below to better illustrate the disclosure.
It will be appreciated by those skilled in the art that without certain details, the disclosure equally be can be implemented.In some instances, for
Method, means, element and circuit well known to those skilled in the art are not described in detail, in order to highlight the purport of the disclosure.
Fig. 1 shows the flow chart of the image processing method according to the embodiment of the present disclosure, as shown in Figure 1, which comprises
In step s 11, multiple pixels of image to be processed are pre-processed, obtains the first image;
In step s 12, the first image input first network is handled, obtains the multiple of the first image
The pixel connection figure of pixel, wherein the pixel connection figure of the first pixel includes the position of first pixel, described
The first probability value and first pixel that one pixel belongs to target area are for adjacent with first pixel
First connection value of multiple pixels, wherein first pixel is any in multiple pixels of the first image
One;
In step s 13, according to the pixel connection figure of multiple pixels of the first image, the first image is determined
In one or more target areas.
Image processing method according to an embodiment of the present disclosure improves processing effect by pre-processing to pixel
Rate determines target area by pixel connection figure, reduces the demand to training sample, also, in pixel connection figure
The parameters such as one probability value and the first connection value reduce the complexity of processing, improve the accuracy of identification of target area.
In one possible implementation, described image processing method can be executed by terminal device, and terminal device can
Think user equipment (User Equipment, UE), mobile device, user terminal, terminal, cellular phone, wireless phone, individual
Digital processing (Personal Digital Assistant, PDA), calculates equipment, mobile unit, wearable sets handheld device
Standby etc., the method can be realized in such a way that processor calls the computer-readable instruction stored in memory.Alternatively,
Image to be processed can be obtained by terminal device or image capture device (such as camera etc.), and image to be processed is sent to service
Device, to execute the method by server.
In one possible implementation, image to be processed may include image acquiring device (such as camera, X-ray machine or
Medical imaging devices etc.) image that gets, for example, image data, medical image data or cell image etc..
In one possible implementation, in step s 11, the attributes such as the shape of image to be processed and color can have
Therefore diversity can pre-process image to be processed, to reduce the diversity of image to be processed, improve treatment effeciency.
In one possible implementation, multiple pixels of image to be processed are pre-processed, obtains the first figure
Picture, comprising: multiple pixels of the image to be processed are normalized, the first image is obtained.In this example,
Multiple pixels of the image to be processed are normalized, the first image is obtained, it may include: to figure to be processed
At least one of the rgb values of multiple pixels of picture, sum of the grayscale values contrast are normalized, and obtain the first image.
In this example, the rgb value of multiple pixels in image to be processed can be normalized, for example, the rgb value of certain pixel point
Not Wei [50,120,30], after the rgb value of the pixel is normalized, the rgb value of the pixel be respectively [0.25,
0.6,0.15]。
In this way, the diversity of image to be processed can be reduced, treatment effeciency is improved.
In one possible implementation, in step s 12, the first image first network can be inputted to handle,
Obtain the pixel connection figure of multiple pixels of the first image.Wherein, the pixel connection figure of the first pixel includes described
The position of first pixel, first pixel belong to the first probability value of target area and first pixel is directed to
First connection value of the multiple pixels adjacent with first pixel, wherein first pixel is first figure
Any one in multiple pixels of picture.
In this example, the first network can be the connection of the pixel between multiple pixels of the first image for identification
The identification network of relationship, the first network can be BP neural network, convolutional neural networks, Recognition with Recurrent Neural Network and recurrence mind
Through neural networks such as networks, the disclosure to the type of first network with no restrictions.The first network can be with multi-layer
The deep learning neural network of structure (that is, with multiple hidden layers), the input layer of the first network, multiple hidden layers and defeated
It can be connected entirely between each neuron of layer out or the tree-shaped such as non-full connection connect.The input layer of the first network can input
Image after the normalizeds such as the first image, after the processing of hidden layer, the exportable institute of the output layer of the first network
The pixel connection figure of each pixel of image after stating normalized, in the pixel connection figure, the pixel of pixel connects
Map interlinking may include that the position of the pixel, the pixel belong to the first probability value of target area (that is, the pixel belongs to target
The Semantic judgement branch in region) and the pixel and 8 adjacent pixels (for example, the adjacent picture of the pixel lower left
The neighbor pixel of neighbor pixel, the pixel lower right below vegetarian refreshments, the pixel, the pixel right adjacent picture
Vegetarian refreshments, the neighbor pixel in the pixel upper right side, the neighbor pixel above the pixel, the pixel are upper left adjacent
Pixel and the neighbor pixel of the pixel left) the first connection value (that is, the connection of the pixel and neighbor pixel
Branch).The recognition capability to input picture can be improved using the first network of multi-layer, improve the precision of output result.
In this example, the first image can be inputted first network to handle, obtains the pixel connection figure of multiple pixels,
First pixel of the first image has the rgb value after the position of the first pixel of description and the normalized of the first pixel
Data, for example, the first pixel in the first image can have the data of [100,50,0.25,0.6,0.15], wherein
[100,50] position coordinates for indicating the first pixel, after [0.25,0.6,0.15] indicates the normalized of the first pixel
Rgb value.
In this example, in the pixel connection figure of the first pixel, there can be the data of the position of the first pixel of description
Belong to the first probability value of target area with the first pixel and first pixel is directed to and the first pixel phase
First connection value of adjacent multiple pixels.For example, in addition to the first pixel position coordinates (such as coordinate value (x, y)) it
Outside, the pixel connection figure of the first pixel can also have 18 data, wherein the 1st data can indicate the first pixel category
The first probability value in target area (such as belonging to nuclear area), the 2nd data can indicate that the first pixel is not belonging to mesh
The probability value of region (for example, belonging to background area) is marked, the first pixel belongs to the first probability value and the first picture of target area
It is 1 that vegetarian refreshments, which is not belonging to the sum of probability value of target area,.Probability threshold value, such as 0.5 can be set, if the first pixel belongs to mesh
First probability value in mark region is greater than or equal to 0.5, and (in this case, the first pixel is not belonging to the probability value of target area
Less than or equal to 0.5), it is believed that the first pixel belongs to target area, otherwise, it is believed that the first pixel is not belonging to target area
Domain.
In this example, the 3rd data of the pixel connection figure of the first pixel to the 18th data (totally 16 data) can
It is divided into 8 groups, every group of data there can be 2 data, and respectively the first pixel is directed to 8 pixels adjacent with the first pixel
(for example, neighbor pixel, the first pixel bottom right below the neighbor pixel of the first pixel lower left, the first pixel
Neighbor pixel, the neighbor pixel of the first pixel right, the neighbor pixel in the first pixel upper right side, the first picture of side
The adjacent pixel of neighbor pixel, the upper left neighbor pixel of the first pixel and the first pixel left above vegetarian refreshments
Point) the first connection value, for example, the 3rd and the 4th data indicate the first pixel for the adjacent of the first pixel lower left
First connection value of pixel, the 5th and the 6th data indicate the first pixel for the adjacent pixel below the first pixel
First connection value of point, the 7th and the 8th data indicate the first pixel for the neighbor pixel of the first pixel lower right
The first connection value, the 9th and the 10th data indicate the first pixel for the neighbor pixel of the first pixel right
First connection value, the 11st and 12-bit data indicate the first pixel for the neighbor pixel in the first pixel upper right side
First connection value, the 13rd and the 14th data indicate the first pixel for the of the neighbor pixel above the first pixel
One connection value, the 15th and the 16th data indicate the first pixel for the of the upper left neighbor pixel of the first pixel
One connection value, the 17th and the 18th data indicate the first pixel for the first pixel left neighbor pixel first
Connection value.That is, each first connection value may include two bits, in this example, two bits may each be 0 or 1, in double figures
When according to being 0, the first connection value indicates that the first pixel is not connected to for its neighbor pixel, is 0 and 1,1 and in two bits
When 0 or 1 and 1, the first connection value indicates that the first pixel is connected to for its neighbor pixel, for example, in the 5th and the 6th
When data are 0, then it represents that the first pixel is not connected to for neighbor pixel below, in the 5th and the 6th data
At least a data when not being 0, then it represents that the first pixel is for neighbor pixel connection below.It should be appreciated that with
On be only the disclosure a kind of possible example, those skilled in the art can arbitrarily set the data in pixel connection figure, this
It the digits of the data such as open the first probability value, the first connection value in the pixel connection figure of the first pixel, sequence and takes
Value mode etc. is with no restriction.
In one possible implementation, for any one in multiple pixels adjacent with first pixel
A pixel (the second pixel), the pixel connection figure of the second pixel may include the position of second pixel, described
The second probability value and second pixel that two pixels belong to target area are for adjacent with second pixel
Second connection value of multiple pixels.In this case, the multiple pixels adjacent with second pixel include described
One pixel.
In step s 13, according to the pixel connection figure of multiple pixels of the first image, the first image is determined
In one or more target areas, it may include:
According to first probability value, second probability value, first pixel for second pixel
First connection value and/or second pixel are directed to the second connection value of first pixel, determine first pixel
The classification of point;According to the classification of multiple pixels of the first image, the boundary of one or more target area is determined;According to
The boundary of one or more of target areas determines one or more of target areas.Wherein, described in the classification expression
First pixel whether be target area borderline pixel.
In this way, it can be connected to value by the first probability value, the second probability value, the first connection value and second and determine
The classification of one pixel, and determine the boundary of target area, the complexity on identification object region boundary is reduced, identification is improved
The treatment effeciency of target area.
In one possible implementation, the first pixel and the second pixel are two adjacent pixels, first
The pixel connection figure of pixel includes the first connection value for the second pixel, and the first pixel can be indicated for the second pixel
Whether point is connected to.The pixel connection figure of second pixel includes the second connection value for the first pixel, can indicate the second picture
Whether vegetarian refreshments is connected to for the first pixel.In this example, the first connection value or the second connection value can indicate the first pixel and
Whether the second pixel belongs to the same area (for example, belong to target area simultaneously or belong to background area simultaneously), in the first picture
First probability value of vegetarian refreshments and the second probability value of the second pixel are all larger than or are equal to probability threshold value, and the first pixel is directed to
First connection value of the second pixel is connected at least one of value for the second of the first pixel with the second pixel and indicates
In the case where connection, it is believed that the first pixel and the connection of the second pixel, that is, the first pixel and the second pixel belong to together
One target area.
In this example, image to be processed is cell image, and target area is the region where nucleus, a pixel and the
Two pixels are two adjacent pixels, and first probability value in the region that the first pixel belongs to where nucleus is greater than or waits
In 0.5, for example, the first probability value is 0.7, then it is believed that the first pixel belongs to the region where nucleus, the second pixel
Belong to the region where nucleus the second probability value be greater than or equal to 0.5, for example, the second probability value be 0.75, then it is believed that
Second pixel belongs to the region where nucleus, also, the first pixel is for the first connection value of the second pixel and the
Two pixels indicate connection at least one of second connection value of the first pixel, it is believed that the first pixel and second
Pixel belongs to the region where same nucleus.
In this example, the second pixel is the neighbor pixel of the first pixel right, and the first pixel is the second pixel
The neighbor pixel of point left, in the pixel connection figure of the first pixel, the 9th and the 10th data the first pixel of expression
For the first connection value of the neighbor pixel (that is, second pixel) of the first pixel right, in the pixel of the first pixel
When the 9th of connection figure and the 10th data are not 0 simultaneously, the first pixel is indicated for the first connection value of the second pixel
Connection.In the pixel connection figure of the second pixel, the 17th and the 18th data indicate that the second pixel is directed to the second pixel
Second connection value of the neighbor pixel (that is, first pixel) of point left, the 17th of the pixel connection figure of the second pixel the
When position and the 18th data are not 0 simultaneously, the second pixel indicates connection for the first connection value of the first pixel.In example
In, the 9th of the pixel connection figure of the first pixel and the 10th digit accordingly and the pixel connection figure of the second pixel
When 17 and the 18th data are not 0 simultaneously, it is believed that the first pixel and the second pixel belong to the same area, in this feelings
Under condition, if the first probability value and the second probability value are all larger than or are equal to 0.5, it is believed that the first pixel and the second pixel
Belong to the region where same nucleus.
In one possible implementation, according to first probability value, second probability value, first pixel
Point connects for the first connection value of second pixel and/or second pixel for the second of first pixel
Logical value, determines the classification of first pixel, comprising:
First probability value be greater than or equal to probability threshold value, and second probability value be less than probability threshold value the case where
Under, the classification of first pixel is determined as to the boundary pixel point of target area.
In this example, the first pixel and the second pixel are two adjacent pixels, and the first pixel belongs to target
First probability value in region is greater than or equal to probability threshold value, then it is believed that the first pixel belongs to target area, the second pixel
Belong to the second probability of target area be less than probability threshold value therefore, can then it is believed that the second pixel is not belonging to target area
Think that the first pixel is on the boundary of target area, that is, the first pixel is the borderline pixel of target area.
In this example, image to be processed be cell image, target area be nucleus where region, the first pixel and
Second pixel be two adjacent pixels, first probability value in the region that the first pixel belongs to where nucleus be greater than or
Equal to 0.5, for example, the first probability value is 0.7, then it is believed that the first pixel belongs to the region where nucleus.Second pixel
Point belong to the region where nucleus the second probability less than 0.5, for example, the second probability value be 0.2, then it is believed that the second picture
Vegetarian refreshments is not belonging to the region where nucleus.Thus, it is believed that the first pixel is on the boundary in the region where nucleus,
That is, the first pixel is the borderline pixel in the region where nucleus.
In this way, the classification that the first pixel can be determined by the first probability value and the second probability value, reduces
The complexity for determining the classification of the first pixel, improves treatment effeciency.
In one possible implementation, according to first probability value, second probability value, first pixel
Point connects for the first connection value of second pixel and/or second pixel for the second of first pixel
Logical value, determines the classification of first pixel, comprising: being connected to value with described second in the first connection value indicates not connecting
In the case where logical, determine that first pixel is not connected to second pixel;It is greater than or waits in first probability value
In the probability threshold value, and in first pixel and the disconnected situation of the second pixel, by first pixel
The classification of point is determined as the boundary pixel point of target area.
In one possible implementation, the first pixel is directed to the first connection value and the second pixel of the second pixel
In the case that point indicates connection at least one of second connection value of the first pixel, it is believed that the first pixel and the
Two pixels connection, that is, the first pixel and the second pixel belong to the same area (for example, belonging to target area or belonging to
In background area), the first pixel is directed to for the first connection value of the second pixel and the second pixel in the first pixel
The second connection value indicate in disconnected situation, it is believed that the first pixel and the second pixel are not connected to, that is, the first picture
Vegetarian refreshments and the second pixel belong to different regions (for example, one in the first pixel and the second pixel belongs to target area
Domain, another belongs to background area or the first pixel and the second pixel belongs to different target areas).
In this example, image to be processed be cell image, target area be nucleus where region, the first pixel and
Second pixel is two adjacent pixels, and the second pixel is the neighbor pixel of the first pixel right, the first pixel
Point is the neighbor pixel of the second pixel left, in the pixel connection figure of the first pixel, the 9th and the 10th tables of data
Show the first pixel for the first connection value of the neighbor pixel (that is, second pixel) of the first pixel right, second
In the pixel connection figure of pixel, the 17th and the 18th data indicate the second pixel for the adjacent of the second pixel left
Second connection value of pixel (that is, first pixel), in the 9th and the 10th data of the pixel connection figure of the first pixel
And second the 17th of pixel connection figure of pixel and the 18th data are when being simultaneously 0, it is believed that the first pixel and the
Two pixels are not connected to, that is, the first pixel and the second pixel are not belonging to the same area.
In one possible implementation, it is greater than or equal to probability threshold value in the first probability value of the first pixel, and
Under first pixel and the disconnected situation of the second pixel, the classification of the first pixel is determined as to the boundary picture of target area
Vegetarian refreshments.First pixel and the second pixel are two adjacent pixels, and the first probability value of the first pixel is greater than or waits
In probability threshold value, then it is believed that the first pixel belongs to target area, the first pixel and the second pixel are not connected to, it is believed that
First pixel and the second pixel are not belonging to the same area, and therefore, the first pixel belongs to target area, and the second pixel
Belong to other regions (for example, other target areas or background area) other than the target area where the first pixel, and
It is believed that the first pixel is on the boundary of target area, that is, the first pixel is the borderline pixel of target area.
In this example, if the second probability value of the second pixel is greater than or equal to probability threshold value, it is believed that the second pixel belongs to
On the boundary of other target areas, if the second probability value of the second pixel is less than probability threshold value, then it is assumed that the second pixel
Belong to background area.
In this example, the first probability value that the first pixel belongs to the region where nucleus is greater than or equal to 0.5, example
Such as, the first probability value is 0.7, then it is believed that the first pixel belongs to the region where nucleus, also, the picture of the first pixel
The 9th of plain connection figure and the 10th digit accordingly and the 17th of the pixel connection figure of the second pixel and the 18th data simultaneously
When being 0, it is believed that the first pixel and the second pixel are not connected to, that is, the first pixel and the second pixel are not belonging to same
Region where nucleus, it is believed that the first pixel belongs on the boundary in the region where nucleus, that is, the first pixel is
The borderline pixel in the region where nucleus.In this case, if the second probability value of the second pixel is greater than
Or it is equal to 0.5, for example, the second probability value is 0.75, then it is believed that the second pixel belongs to the region where another nucleus
On boundary, if the second probability value of the second pixel less than 0.5, for example, the second probability value be 0.2, then it is assumed that the second pixel
Point belongs to background area.
In this way, value can be connected to second by the first probability value, the first connection value determine the first pixel
Classification reduces the complexity for determining the classification of the first pixel, improves treatment effeciency.
In one possible implementation, one or more target area can be determined by the classification of multiple pixels
Boundary, that is, can determine all borderline pixels in target area, these pixels according to the classification of pixel
Constitute the boundary of one or more target area.And one or more can be determined according to the boundary of one or more target areas
A target area, for example, being target area by target area boundaries area encompassed.
Fig. 2 shows the flow charts according to the image processing method of the embodiment of the present disclosure, as shown in Fig. 2, the method is also wrapped
It includes:
In step S14, one or more target areas addition number respectively in the first image is identified, and is obtained
The regional prediction figure of the first image.
In one possible implementation, it may respectively be the number mark that each target area is randomly assigned displacement, obtain
Obtain the regional prediction figure (entity prognostic chart) of the first image.In this example, the number of all pixels point in same target area
Mark can be consistent.Each pixel in regional prediction figure can have the data and number mark of description pixel position.
Fig. 3 shows the schematic diagram of the regional prediction figure according to the image processing method of the embodiment of the present disclosure.As shown in figure 3,
Each unit in regional prediction figure can indicate that a pixel in the first figure phase, each pixel can have description pixel
The data of position, such as [8,6] indicate the pixel that eighth row the 6th arranges, in addition, each pixel also has digital mark, example
Such as, the number for the pixel that eighth row the 6th arranges is identified as 5, and all pixels point that number is identified as 5 constitutes a target area.
In this example, 0 background area can be indicated, the pixel with other number marks separately constitutes different target areas.
Fig. 4 shows the flow chart of the image processing method according to the embodiment of the present disclosure, as shown in figure 4, the method is also wrapped
It includes:
It in step S15, is identified according to the number of the regional prediction figure, in respectively one or more target areas
Pixel distribution rgb value corresponding with the number mark, obtains the second colored image.
In one possible implementation, it can be identified according to the number of each target area, respectively each target area
Domain distribution rgb value corresponding with digital mark, the number mark of each target area can be uniquely, therefore, each target
The rgb value in region is also possible to uniquely, and the rgb value of target area is different from each other, that is, the color of different target areas is not
Together.
In this example, image to be processed is cell image, and it can be background area that target area, which is the region where nucleus,
Color be set as black or white, and distribute rgb value, the color of each nucleus region for each nucleus region
It is that uniquely, the discrimination between different nucleus regions can be made higher, keep the effect of visualization of the second image preferable.
It in this way, can be by distributing rgb value for target area, to improve the differentiation between different target region
Degree improves the effect of visualization of the second image.
In one possible implementation, before handling the first image using first network, first network can be carried out
Training.
Fig. 5 shows the flow chart of the image processing method according to the embodiment of the present disclosure, as shown in figure 5, the method is also wrapped
It includes:
In step s 16, described is trained by the training set being made of multiple sample images after normalized
One network.
In one possible implementation, the multiple sample image is the image carried out after normalized,
Any sample image can be inputted into first network, obtain the sampled pixel connection figure of multiple pixels of the sample image, institute
Stating can have the position of first sample pixel, first sample pixel to belong to the first of target area in sampled pixel connection figure
Probability value and first sample pixel are directed to the first connection value of the multiple pixels adjacent with first sample pixel,
In, the first sample pixel is any pixel point in the sample image.
In one possible implementation, one or more in sample image can be determined according to sampled pixel connection figure
A sample object region, further, can according to the true target area in the sample object region and the sample image,
Determine that each pixel belongs to the intersection entropy loss and each pixel and adjacent pixel of the probability in sample object region
Connection value intersection entropy loss, and then determine first network loss function.
In one possible implementation, the loss function can be used to adjust the network parameter values of first network,
In this example, the network parameter values can be adjusted according to the direction for minimizing loss function, makes first network adjusted
The goodness of fit with higher, while avoiding over-fitting.In this example, gradient descent method can be used to carry out the reversed of loss function
It propagates, to adjust the network parameter values of first network, for example, for the first network for carrying out tree-shaped connection between each neuron,
The methods of stochastic gradient descent method adjustment network parameter can be used to improve adjustment to reduce the complexity of adjustment network parameter
The efficiency of network parameter.
In this example, the sample image of predetermined quantity can be inputted to first network, that is, by the network parameter values of first network
Adjust pre-determined number.In this example, the number of adjustment can not be limited, and reduces to a certain extent or converge on one in loss function
When determining in threshold value, stops adjustment, obtain first network adjusted.And first network adjusted can be used to obtain the first figure
In the step of pixel connection figure of multiple pixels of picture.
Image processing method according to an embodiment of the present disclosure can be returned by multiple pixels to image to be processed
One change processing, reduces the diversity of image to be processed, improves treatment effeciency, the first probability value, the second probability value, the first connection
Value is connected to the classification that value determines the first pixel with second, reduces the complexity on identification object region boundary, reduces to instruction
The demand for practicing sample, improves the treatment effeciency of identification object region, and improve the accuracy of identification of target area, further
Ground, can be by distributing rgb value for target area, to improve the discrimination between different target region, and that improves the second image can
Depending on changing effect.
Fig. 6 shows the application schematic diagram of the image processing method according to the embodiment of the present disclosure, as shown in fig. 6, figure to be processed
As being cell image, target area is nucleus region.It can be by the rgb value of multiple pixels of image to be processed, gray scale
At least one of value and contrast are normalized, and obtain the first image.
In one possible implementation, the first image can be inputted first network to handle, obtains multiple pixels
The pixel connection figure of point, in this example, the first pixel are any one in multiple pixels of the first image, the first pixel
The pixel connection figure of point includes the first probability value and that the position coordinates of the first pixel, the first pixel belong to target area
One pixel is directed to the first connection value of 8 pixels adjacent with the first pixel.In this example, in addition to the first pixel
Except position coordinates, the pixel connection figure of the first pixel can also have 18 data, wherein the 1st data can indicate
One pixel belongs to the first probability value of nucleus region, and the 2nd can indicate that the first pixel is not belonging to nucleus for data
The probability value of region (for example, belonging to background area), the first pixel belong to the first probability value of nucleus region
It is 1 that the sum of probability value of nucleus region is not belonging to the first pixel.The 3rd of the pixel connection figure of first pixel
Data to the 18th data (totally 16 data) can be divided into 8 groups, and every group of data can have 2 data, respectively the first pixel
For the first connection value of 8 pixels adjacent with the first pixel, wherein for the two bits of each neighbor pixel
It may each be 0 or 1, when two bits are 0, indicate that the first pixel is not connected to for its neighbor pixel, for example, second
Pixel is any one in multiple pixels adjacent with first pixel, and the first pixel is directed to the second pixel
Two the first connection values be 0, then it represents that the first pixel is not connected to for the second pixel.
In one possible implementation, institute can be determined by the pixel connection figure of multiple pixels of the first image
State one or more nucleus regions in the first image.In this example, the multiple pixels adjacent with the second pixel
Including the first pixel, the pixel connection figure of the second pixel includes that the position of the second pixel, the second pixel belong to cell
Second probability value and the second pixel in the region where core are directed to the second of the multiple pixels adjacent with the second pixel
Connection value.It can be according to the first probability value, the second probability value, the first pixel for the first connection value of the second pixel and the
Two pixels are directed to the second connection value of the first pixel, determine the classification of the first pixel.
In one possible implementation, the probability threshold value of the first probability value and the second probability value can be set, such as
0.5, can the first probability value be greater than or equal to 0.5 and second probability value less than 0.5 in the case where, by the class of the first pixel
It is not determined as the boundary pixel point of nucleus region.Alternatively, 0.5, and the first picture can be greater than or equal in the first probability value
Under vegetarian refreshments and the disconnected situation of the second pixel, the classification of the first pixel is determined as to the boundary picture of nucleus region
Vegetarian refreshments.In this example, the first picture is directed to for the first connection value of the second pixel and the second pixel in the first pixel
When second connection value of vegetarian refreshments indicates not to be connected to, it may be determined that the first pixel is not connected to the second pixel, for example, the first picture
Vegetarian refreshments is 0 for two the first connection values of the second pixel, and the second pixel is directed to two second of the first pixel
Connection value is 0.
In one possible implementation, the classification of each pixel can be determined respectively, and according to multiple pixels
Classification determines the boundary of nucleus region, further, nucleus institute can be determined by the boundary of nucleus region
In region.
In one possible implementation, it may respectively be the number mark that each nucleus region is randomly assigned displacement
Know, obtains the regional prediction figure (as shown in Figure 3) of the first image.And respectively each nucleus region distribution is marked with number
Know corresponding rgb value, obtain the second colored image, promotes effect of visualization, in this example, the color of background area is set as black
Color, and unique color is distributed for each nucleus region.
It is appreciated that above-mentioned each embodiment of the method that the disclosure refers to, without prejudice to principle logic,
To engage one another while the embodiment to be formed after combining, as space is limited, the disclosure is repeated no more.
In addition, the disclosure additionally provides image processing apparatus, electronic equipment, computer readable storage medium, program, it is above-mentioned
It can be used to realize any image processing method that the disclosure provides, corresponding technical solution and description and referring to method part
It is corresponding to record, it repeats no more.
It will be understood by those skilled in the art that each step writes sequence simultaneously in the above method of specific embodiment
It does not mean that stringent execution sequence and any restriction is constituted to implementation process, the specific execution sequence of each step should be with its function
It can be determined with possible internal logic.
Fig. 7 shows the block diagram of the image processing apparatus according to the embodiment of the present disclosure, as shown in fig. 7, described device includes:
Preprocessing module 11 is pre-processed for multiple pixels to image to be processed, obtains the first image;
Pixel connection figure obtains module 12, for handling the first image input first network, described in acquisition
The pixel connection figure of multiple pixels of first image;
Target area determining module 13 is determined for the pixel connection figure according to multiple pixels of the first image
One or more target areas in the first image.
In one possible implementation, preprocessing module 11 is further used for:
Multiple pixels of the image to be processed are normalized, the first image is obtained.
In one possible implementation, preprocessing module 11 is further used for:
At least one of the rgb values of multiple pixels of image to be processed, sum of the grayscale values contrast are normalized
Processing obtains the first image.
In one possible implementation, the pixel connection figure of first pixel includes first pixel
Position, first pixel belong to the first probability value of target area and first pixel is directed to and first picture
First connection value of the adjacent multiple pixels of vegetarian refreshments, wherein first pixel is multiple pixels of the first image
Point in any one.
In one possible implementation, the pixel connection figure of the second pixel includes the position of second pixel
It sets, second pixel belongs to the second probability value of target area and second pixel is directed to and second pixel
Second connection value of the adjacent multiple pixels of point,
Wherein, second pixel is any one in multiple pixels adjacent with first pixel, with
The adjacent multiple pixels of second pixel include first pixel,
Wherein, target area determining module 13 is further used for:
According to first probability value, second probability value, first pixel for second pixel
First connection value and/or second pixel are directed to the second connection value of first pixel, determine first pixel
The classification of point;
According to the classification of multiple pixels of the first image, the boundary of one or more target area is determined;
According to the boundary of one or more of target areas, one or more of target areas are determined.
In one possible implementation, target area determining module 13 is further used for:
First probability value be greater than or equal to probability threshold value, and second probability value be less than probability threshold value the case where
Under, the classification of first pixel is determined as to the boundary pixel point of target area.
In one possible implementation, target area determining module 13 is further used for:
In the case where the first connection value with described second is connected to value and indicates disconnected situation, first pixel is determined
Point is not connected to second pixel;
It is greater than or equal to the probability threshold value, and first pixel and second pixel in first probability value
In the disconnected situation of point, the classification of first pixel is determined as to the boundary pixel point of target area.
Fig. 8 shows the block diagram of the image processing apparatus according to the embodiment of the present disclosure, as shown in figure 8, described device further include:
Adding module 14, for being respectively one or more target areas addition number mark in the first image,
Obtain the regional prediction figure of the first image.
In one possible implementation, described device further include:
Distribution module 15, for being identified according to the number of the regional prediction figure, respectively one or more target areas
In pixel distribution rgb value corresponding with the digital mark, obtain the second image of colour.
In one possible implementation, described device further include:
Training module 16, it is described for being trained by the training set being made of multiple sample images after normalized
First network.
In some embodiments, the embodiment of the present disclosure provides the function that has of device or comprising module can be used for holding
The method of row embodiment of the method description above, specific implementation are referred to the description of embodiment of the method above, for sake of simplicity, this
In repeat no more
The embodiment of the present disclosure also proposes a kind of computer readable storage medium, is stored thereon with computer program instructions, institute
It states when computer program instructions are executed by processor and realizes the above method.Computer readable storage medium can be non-volatile meter
Calculation machine readable storage medium storing program for executing.
The embodiment of the present disclosure also proposes a kind of electronic equipment, comprising: processor;For storage processor executable instruction
Memory;Wherein, the processor is configured to the above method.
The equipment that electronic equipment may be provided as terminal, server or other forms.
Fig. 9 is the block diagram of a kind of electronic equipment 800 shown according to an exemplary embodiment.For example, electronic equipment 800 can
To be mobile phone, computer, digital broadcasting terminal, messaging device, game console, tablet device, Medical Devices are good for
Body equipment, the terminals such as personal digital assistant.
Referring to Fig. 9, electronic equipment 800 may include following one or more components: processing component 802, memory 804,
Power supply module 806, multimedia component 808, audio component 810, the interface 812 of input/output (I/O), sensor module 814,
And communication component 816.
The integrated operation of the usual controlling electronic devices 800 of processing component 802, such as with display, call, data are logical
Letter, camera operation and record operate associated operation.Processing component 802 may include one or more processors 820 to hold
Row instruction, to perform all or part of the steps of the methods described above.In addition, processing component 802 may include one or more moulds
Block, convenient for the interaction between processing component 802 and other assemblies.For example, processing component 802 may include multi-media module, with
Facilitate the interaction between multimedia component 808 and processing component 802.
Memory 804 is configured as storing various types of data to support the operation in electronic equipment 800.These data
Example include any application or method for being operated on electronic equipment 800 instruction, contact data, telephone directory
Data, message, picture, video etc..Memory 804 can by any kind of volatibility or non-volatile memory device or it
Combination realize, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM) is erasable
Except programmable read only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, fastly
Flash memory, disk or CD.
Power supply module 806 provides electric power for the various assemblies of electronic equipment 800.Power supply module 806 may include power supply pipe
Reason system, one or more power supplys and other with for electronic equipment 800 generate, manage, and distribute the associated component of electric power.
Multimedia component 808 includes the screen of one output interface of offer between the electronic equipment 800 and user.
In some embodiments, screen may include liquid crystal display (LCD) and touch panel (TP).If screen includes touch surface
Plate, screen may be implemented as touch screen, to receive input signal from the user.Touch panel includes one or more touches
Sensor is to sense the gesture on touch, slide, and touch panel.The touch sensor can not only sense touch or sliding
The boundary of movement, but also detect duration and pressure associated with the touch or slide operation.In some embodiments,
Multimedia component 808 includes a front camera and/or rear camera.When electronic equipment 800 is in operation mode, as clapped
When taking the photograph mode or video mode, front camera and/or rear camera can receive external multi-medium data.It is each preposition
Camera and rear camera can be a fixed optical lens system or have focusing and optical zoom capabilities.
Audio component 810 is configured as output and/or input audio signal.For example, audio component 810 includes a Mike
Wind (MIC), when electronic equipment 800 is in operation mode, when such as call mode, recording mode, and voice recognition mode, microphone
It is configured as receiving external audio signal.The received audio signal can be further stored in memory 804 or via logical
Believe that component 816 is sent.In some embodiments, audio component 810 further includes a loudspeaker, is used for output audio signal.
I/O interface 812 provides interface between processing component 802 and peripheral interface module, and above-mentioned peripheral interface module can
To be keyboard, click wheel, button etc..These buttons may include, but are not limited to: home button, volume button, start button and lock
Determine button.
Sensor module 814 includes one or more sensors, for providing the state of various aspects for electronic equipment 800
Assessment.For example, sensor module 814 can detecte the state that opens/closes of electronic equipment 800, the relative positioning of component, example
As the component be electronic equipment 800 display and keypad, sensor module 814 can also detect electronic equipment 800 or
The position change of 800 1 components of electronic equipment, the existence or non-existence that user contacts with electronic equipment 800, electronic equipment 800
The temperature change of orientation or acceleration/deceleration and electronic equipment 800.Sensor module 814 may include proximity sensor, be configured
For detecting the presence of nearby objects without any physical contact.Sensor module 814 can also include optical sensor,
Such as CMOS or ccd image sensor, for being used in imaging applications.In some embodiments, which may be used also
To include acceleration transducer, gyro sensor, Magnetic Sensor, pressure sensor or temperature sensor.
Communication component 816 is configured to facilitate the communication of wired or wireless way between electronic equipment 800 and other equipment.
Electronic equipment 800 can access the wireless network based on communication standard, such as WiFi, 2G or 3G or their combination.Show at one
In example property embodiment, communication component 816 receives broadcast singal or broadcast from external broadcasting management system via broadcast channel
Relevant information.In one exemplary embodiment, the communication component 816 further includes near-field communication (NFC) module, short to promote
Cheng Tongxin.For example, radio frequency identification (RFID) technology, Infrared Data Association (IrDA) technology, ultra wide band can be based in NFC module
(UWB) technology, bluetooth (BT) technology and other technologies are realized.
In the exemplary embodiment, electronic equipment 800 can be by one or more application specific integrated circuit (ASIC), number
Word signal processor (DSP), digital signal processing appts (DSPD), programmable logic device (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components are realized, for executing the above method.
In the exemplary embodiment, a kind of non-volatile computer readable storage medium storing program for executing is additionally provided, for example including calculating
The memory 804 of machine program instruction, above-mentioned computer program instructions can be executed by the processor 820 of electronic equipment 800 to complete
The above method.
Figure 10 is the block diagram of a kind of electronic equipment 1900 shown according to an exemplary embodiment.For example, electronic equipment
1900 may be provided as a server.Referring to Fig.1 0, it further comprises one that electronic equipment 1900, which includes processing component 1922,
A or multiple processors and memory resource represented by a memory 1932, can be by processing component 1922 for storing
The instruction of execution, such as application program.The application program stored in memory 1932 may include one or more every
One corresponds to the module of one group of instruction.In addition, processing component 1922 is configured as executing instruction, to execute the above method.
Electronic equipment 1900 can also include that a power supply module 1926 is configured as executing the power supply of electronic equipment 1900
Management, a wired or wireless network interface 1950 is configured as electronic equipment 1900 being connected to network and an input is defeated
(I/O) interface 1958 out.Electronic equipment 1900 can be operated based on the operating system for being stored in memory 1932, such as
Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM or similar.
In the exemplary embodiment, a kind of non-volatile computer readable storage medium storing program for executing is additionally provided, for example including calculating
The memory 1932 of machine program instruction, above-mentioned computer program instructions can by the processing component 1922 of electronic equipment 1900 execute with
Complete the above method.
The disclosure can be system, method and/or computer program product.Computer program product may include computer
Readable storage medium storing program for executing, containing for making processor realize the computer-readable program instructions of various aspects of the disclosure.
Computer readable storage medium, which can be, can keep and store the tangible of the instruction used by instruction execution equipment
Equipment.Computer readable storage medium for example can be-- but it is not limited to-- storage device electric, magnetic storage apparatus, optical storage
Equipment, electric magnetic storage apparatus, semiconductor memory apparatus or above-mentioned any appropriate combination.Computer readable storage medium
More specific example (non exhaustive list) includes: portable computer diskette, hard disk, random access memory (RAM), read-only deposits
It is reservoir (ROM), erasable programmable read only memory (EPROM or flash memory), static random access memory (SRAM), portable
Compact disk read-only memory (CD-ROM), digital versatile disc (DVD), memory stick, floppy disk, mechanical coding equipment, for example thereon
It is stored with punch card or groove internal projection structure and the above-mentioned any appropriate combination of instruction.Calculating used herein above
Machine readable storage medium storing program for executing is not interpreted that instantaneous signal itself, the electromagnetic wave of such as radio wave or other Free propagations lead to
It crosses the electromagnetic wave (for example, the light pulse for passing through fiber optic cables) of waveguide or the propagation of other transmission mediums or is transmitted by electric wire
Electric signal.
Computer-readable program instructions as described herein can be downloaded to from computer readable storage medium it is each calculate/
Processing equipment, or outer computer or outer is downloaded to by network, such as internet, local area network, wide area network and/or wireless network
Portion stores equipment.Network may include copper transmission cable, optical fiber transmission, wireless transmission, router, firewall, interchanger, gateway
Computer and/or Edge Server.Adapter or network interface in each calculating/processing equipment are received from network to be counted
Calculation machine readable program instructions, and the computer-readable program instructions are forwarded, for the meter being stored in each calculating/processing equipment
In calculation machine readable storage medium storing program for executing.
Computer program instructions for executing disclosure operation can be assembly instruction, instruction set architecture (ISA) instructs,
Machine instruction, machine-dependent instructions, microcode, firmware instructions, condition setup data or with one or more programming languages
The source code or object code that any combination is write, the programming language include the programming language-of object-oriented such as
Smalltalk, C++ etc., and conventional procedural programming languages-such as " C " language or similar programming language.Computer
Readable program instructions can be executed fully on the user computer, partly execute on the user computer, be only as one
Vertical software package executes, part executes on the remote computer or completely in remote computer on the user computer for part
Or it is executed on server.In situations involving remote computers, remote computer can pass through network-packet of any kind
It includes local area network (LAN) or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as benefit
It is connected with ISP by internet).In some embodiments, by utilizing computer-readable program instructions
Status information carry out personalized customization electronic circuit, such as programmable logic circuit, field programmable gate array (FPGA) or can
Programmed logic array (PLA) (PLA), the electronic circuit can execute computer-readable program instructions, to realize each side of the disclosure
Face.
Referring herein to according to the flow chart of the method, apparatus (system) of the embodiment of the present disclosure and computer program product and/
Or block diagram describes various aspects of the disclosure.It should be appreciated that flowchart and or block diagram each box and flow chart and/
Or in block diagram each box combination, can be realized by computer-readable program instructions.
These computer-readable program instructions can be supplied to general purpose computer, special purpose computer or other programmable datas
The processor of processing unit, so that a kind of machine is produced, so that these instructions are passing through computer or other programmable datas
When the processor of processing unit executes, function specified in one or more boxes in implementation flow chart and/or block diagram is produced
The device of energy/movement.These computer-readable program instructions can also be stored in a computer-readable storage medium, these refer to
It enables so that computer, programmable data processing unit and/or other equipment work in a specific way, thus, it is stored with instruction
Computer-readable medium then includes a manufacture comprising in one or more boxes in implementation flow chart and/or block diagram
The instruction of the various aspects of defined function action.
Computer-readable program instructions can also be loaded into computer, other programmable data processing units or other
In equipment, so that series of operation steps are executed in computer, other programmable data processing units or other equipment, to produce
Raw computer implemented process, so that executed in computer, other programmable data processing units or other equipment
Instruct function action specified in one or more boxes in implementation flow chart and/or block diagram.
The flow chart and block diagram in the drawings show system, method and the computer journeys according to multiple embodiments of the disclosure
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
One module of table, program segment or a part of instruction, the module, program segment or a part of instruction include one or more use
The executable instruction of the logic function as defined in realizing.In some implementations as replacements, function marked in the box
It can occur in a different order than that indicated in the drawings.For example, two continuous boxes can actually be held substantially in parallel
Row, they can also be executed in the opposite order sometimes, and this depends on the function involved.It is also noted that block diagram and/or
The combination of each box in flow chart and the box in block diagram and or flow chart, can the function as defined in executing or dynamic
The dedicated hardware based system made is realized, or can be realized using a combination of dedicated hardware and computer instructions.
The presently disclosed embodiments is described above, above description is exemplary, and non-exclusive, and
It is not limited to disclosed each embodiment.Without departing from the scope and spirit of illustrated each embodiment, for this skill
Many modifications and changes are obvious for the those of ordinary skill in art field.The selection of term used herein, purport
In the principle, practical application or technological improvement to the technology in market for best explaining each embodiment, or lead this technology
Other those of ordinary skill in domain can understand each embodiment disclosed herein.
Claims (10)
1. a kind of image processing method, which is characterized in that the described method includes:
Multiple pixels of image to be processed are pre-processed, the first image is obtained;
The first image input first network is handled, the pixel connection of multiple pixels of the first image is obtained
Figure;
According to the pixel connection figure of multiple pixels of the first image, one or more mesh in the first image are determined
Mark region.
2. the method according to claim 1, wherein the pixel connection figure of first pixel includes described
The position of one pixel, first pixel belong to target area the first probability value and first pixel be directed to
First connection value of the adjacent multiple pixels of first pixel, wherein first pixel is the first image
Multiple pixels in any one.
3. method according to claim 1 or 2, which is characterized in that located in advance to multiple pixels of image to be processed
Reason obtains the first image, comprising:
Multiple pixels of the image to be processed are normalized, the first image is obtained.
4. according to the method in claim 2 or 3, which is characterized in that the pixel connection figure of the second pixel includes described
The position of two pixels, second pixel belong to target area the second probability value and second pixel be directed to
Second connection value of the adjacent multiple pixels of second pixel,
Wherein, second pixel is any one in multiple pixels adjacent with first pixel, and described
The adjacent multiple pixels of second pixel include first pixel,
Wherein, according to the pixel connection figure of multiple pixels of the first image, determine one in the first image or
Multiple target areas, comprising:
The first of second pixel is directed to according to first probability value, second probability value, first pixel
Connection value and/or second pixel are directed to the second connection value of first pixel, determine first pixel
Classification;
According to the classification of multiple pixels of the first image, the boundary of one or more target area is determined;
According to the boundary of one or more of target areas, one or more of target areas are determined.
5. according to the method described in claim 4, it is characterized in that, according to first probability value, second probability value, institute
The first pixel is stated for the first connection value of second pixel and/or second pixel for first pixel
Second connection value of point, determines the classification of first pixel, comprising:
In the case where first probability value is greater than or equal to probability threshold value and second probability value is less than probability threshold value,
The classification of first pixel is determined as to the boundary pixel point of target area.
6. method according to claim 4 or 5, which is characterized in that according to first probability value, second probability
Value, first pixel are for the first connection value of second pixel and/or second pixel for described the
Second connection value of one pixel, determines the classification of first pixel, comprising:
In the case where the first connection value with described second is connected to value and indicates disconnected situation, determine first pixel with
Second pixel is not connected to;
It is greater than or equal to the probability threshold value in first probability value, and first pixel and second pixel are not
In the case where connection, the classification of first pixel is determined as to the boundary pixel point of target area.
7. method according to claim 1 to 6, which is characterized in that the method also includes:
One or more target areas addition number mark respectively in the first image, obtains the area of the first image
Domain prognostic chart.
8. a kind of image processing apparatus characterized by comprising
Preprocessing module is pre-processed for multiple pixels to image to be processed, obtains the first image;
Pixel connection figure obtains module, for handling the first image input first network, obtains first figure
The pixel connection figure of multiple pixels of picture;
Target area determining module determines described for the pixel connection figure according to multiple pixels of the first image
One or more target areas in one image.
9. a kind of electronic equipment characterized by comprising
Processor;
Memory for storage processor executable instruction;
Wherein, the processor is configured to: perform claim require any one of 1 to 7 described in method.
10. a kind of computer readable storage medium, is stored thereon with computer program instructions, which is characterized in that the computer
Method described in any one of claim 1 to 7 is realized when program instruction is executed by processor.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811068237.8A CN109255784B (en) | 2018-09-13 | 2018-09-13 | Image processing method and device, electronic equipment and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811068237.8A CN109255784B (en) | 2018-09-13 | 2018-09-13 | Image processing method and device, electronic equipment and storage medium |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109255784A true CN109255784A (en) | 2019-01-22 |
CN109255784B CN109255784B (en) | 2021-06-25 |
Family
ID=65047443
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811068237.8A Active CN109255784B (en) | 2018-09-13 | 2018-09-13 | Image processing method and device, electronic equipment and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109255784B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109816650A (en) * | 2019-01-24 | 2019-05-28 | 强联智创(北京)科技有限公司 | A kind of target area recognition methods and its system based on two-dimentional DSA image |
CN111311623A (en) * | 2020-02-26 | 2020-06-19 | 歌尔股份有限公司 | Image boundary method, device, equipment and storage medium |
WO2021098300A1 (en) * | 2019-11-18 | 2021-05-27 | 北京京东尚科信息技术有限公司 | Facial parsing method and related devices |
CN116152043A (en) * | 2023-03-24 | 2023-05-23 | 摩尔线程智能科技(北京)有限责任公司 | Memory management method and device based on image processing and electronic equipment |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080170778A1 (en) * | 2007-01-15 | 2008-07-17 | Huitao Luo | Method and system for detection and removal of redeyes |
US20110243417A1 (en) * | 2008-09-03 | 2011-10-06 | Rutgers, The State University Of New Jersey | System and method for accurate and rapid identification of diseased regions on biological images with applications to disease diagnosis and prognosis |
CN105654475A (en) * | 2015-12-25 | 2016-06-08 | 中国人民解放军理工大学 | Image saliency detection method and image saliency detection device based on distinguishable boundaries and weight contrast |
CN105894517A (en) * | 2016-04-22 | 2016-08-24 | 北京理工大学 | CT image liver segmentation method and system based on characteristic learning |
CN106611427A (en) * | 2015-10-21 | 2017-05-03 | 中国人民解放军理工大学 | A video saliency detection method based on candidate area merging |
CN107028593A (en) * | 2017-04-14 | 2017-08-11 | 成都知识视觉科技有限公司 | A kind of aided detection method of breast ductal carcinoma in situ |
CN107424155A (en) * | 2017-04-17 | 2017-12-01 | 河海大学 | A kind of focusing dividing method towards light field refocusing image |
CN108305266A (en) * | 2017-12-26 | 2018-07-20 | 浙江工业大学 | Semantic image dividing method based on the study of condition random field graph structure |
-
2018
- 2018-09-13 CN CN201811068237.8A patent/CN109255784B/en active Active
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20080170778A1 (en) * | 2007-01-15 | 2008-07-17 | Huitao Luo | Method and system for detection and removal of redeyes |
US20110243417A1 (en) * | 2008-09-03 | 2011-10-06 | Rutgers, The State University Of New Jersey | System and method for accurate and rapid identification of diseased regions on biological images with applications to disease diagnosis and prognosis |
CN106611427A (en) * | 2015-10-21 | 2017-05-03 | 中国人民解放军理工大学 | A video saliency detection method based on candidate area merging |
CN105654475A (en) * | 2015-12-25 | 2016-06-08 | 中国人民解放军理工大学 | Image saliency detection method and image saliency detection device based on distinguishable boundaries and weight contrast |
CN105894517A (en) * | 2016-04-22 | 2016-08-24 | 北京理工大学 | CT image liver segmentation method and system based on characteristic learning |
CN107028593A (en) * | 2017-04-14 | 2017-08-11 | 成都知识视觉科技有限公司 | A kind of aided detection method of breast ductal carcinoma in situ |
CN107424155A (en) * | 2017-04-17 | 2017-12-01 | 河海大学 | A kind of focusing dividing method towards light field refocusing image |
CN108305266A (en) * | 2017-12-26 | 2018-07-20 | 浙江工业大学 | Semantic image dividing method based on the study of condition random field graph structure |
Non-Patent Citations (3)
Title |
---|
MICHAEL KAMPFFMEYER等: "ConnNet: A Long-Range Relation-Aware Pixel-Connectivity Network for Salient Segmentation", 《ARXIV》 * |
朱燕: "图像软分割算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
陈忠碧等: "一种适合于多目标检测的图像分割方法", 《光电工程》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109816650A (en) * | 2019-01-24 | 2019-05-28 | 强联智创(北京)科技有限公司 | A kind of target area recognition methods and its system based on two-dimentional DSA image |
WO2021098300A1 (en) * | 2019-11-18 | 2021-05-27 | 北京京东尚科信息技术有限公司 | Facial parsing method and related devices |
CN111311623A (en) * | 2020-02-26 | 2020-06-19 | 歌尔股份有限公司 | Image boundary method, device, equipment and storage medium |
CN116152043A (en) * | 2023-03-24 | 2023-05-23 | 摩尔线程智能科技(北京)有限责任公司 | Memory management method and device based on image processing and electronic equipment |
CN116152043B (en) * | 2023-03-24 | 2024-03-08 | 摩尔线程智能科技(北京)有限责任公司 | Memory management method and device based on image processing and electronic equipment |
Also Published As
Publication number | Publication date |
---|---|
CN109255784B (en) | 2021-06-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109829501A (en) | Image processing method and device, electronic equipment and storage medium | |
CN109522910A (en) | Critical point detection method and device, electronic equipment and storage medium | |
CN109255784A (en) | Image processing method and device, electronic equipment and storage medium | |
CN110503023A (en) | Biopsy method and device, electronic equipment and storage medium | |
CN109800744A (en) | Image clustering method and device, electronic equipment and storage medium | |
CN106651955A (en) | Method and device for positioning object in picture | |
CN107798669A (en) | Image defogging method, device and computer-readable recording medium | |
CN109658352A (en) | Optimization method and device, electronic equipment and the storage medium of image information | |
CN109544560A (en) | Image processing method and device, electronic equipment and storage medium | |
CN108764069A (en) | Biopsy method and device | |
CN109145970A (en) | Question and answer treating method and apparatus, electronic equipment and storage medium based on image | |
CN109344832A (en) | Image processing method and device, electronic equipment and storage medium | |
CN110287874A (en) | Target tracking method and device, electronic equipment and storage medium | |
CN109784255A (en) | Neural network training method and device and recognition methods and device | |
CN109658401A (en) | Image processing method and device, electronic equipment and storage medium | |
CN110298310A (en) | Image processing method and device, electronic equipment and storage medium | |
CN109801270A (en) | Anchor point determines method and device, electronic equipment and storage medium | |
CN109978891A (en) | Image processing method and device, electronic equipment and storage medium | |
CN109829863A (en) | Image processing method and device, electronic equipment and storage medium | |
CN106778773A (en) | The localization method and device of object in picture | |
CN107563994A (en) | The conspicuousness detection method and device of image | |
CN106875446B (en) | Camera method for relocating and device | |
CN106295499A (en) | Age estimation method and device | |
CN109584362A (en) | 3 D model construction method and device, electronic equipment and storage medium | |
CN109902738A (en) | Network module and distribution method and device, electronic equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |