CN110111343B - Middle-intelligence image segmentation method and device based on improved fuzzy C-means - Google Patents

Middle-intelligence image segmentation method and device based on improved fuzzy C-means Download PDF

Info

Publication number
CN110111343B
CN110111343B CN201910376778.5A CN201910376778A CN110111343B CN 110111343 B CN110111343 B CN 110111343B CN 201910376778 A CN201910376778 A CN 201910376778A CN 110111343 B CN110111343 B CN 110111343B
Authority
CN
China
Prior art keywords
image
value
subset
mean
alpha
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
Application number
CN201910376778.5A
Other languages
Chinese (zh)
Other versions
CN110111343A (en
Inventor
赵晶
王晓莉
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Shandong Huizhi Education Technology Development Co.,Ltd.
Original Assignee
Qilu University of Technology
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Qilu University of Technology filed Critical Qilu University of Technology
Priority to CN201910376778.5A priority Critical patent/CN110111343B/en
Publication of CN110111343A publication Critical patent/CN110111343A/en
Application granted granted Critical
Publication of CN110111343B publication Critical patent/CN110111343B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/11—Region-based segmentation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Image Analysis (AREA)
  • Image Processing (AREA)

Abstract

The disclosure discloses a middle-intelligence image segmentation method and a middle-intelligence image segmentation device based on improved fuzzy C mean value, wherein the method comprises the steps of receiving an original image to be segmented; converting an original image into a mesool image according to a mesool theory; denoising the middle-intelligence image by adopting alpha-mean operation, and enhancing the image; calculating the information entropy of the uncertain subset I in the image-enhanced middle-intelligence image; and when the information entropy calculation result is smaller than a threshold value, performing intelligent image segmentation by adopting a particle swarm optimization fuzzy C-means clustering algorithm, and otherwise, returning to the continuous denoising of the intelligent image.

Description

Middle-intelligence image segmentation method and device based on improved fuzzy C-means
Technical Field
The disclosure belongs to the technical field of image segmentation, and relates to a middle-intelligence image segmentation method and device based on improved fuzzy C-means.
Background
The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art.
Image segmentation is an important step in image processing, pattern recognition and computer vision, and is also a classic problem in computer vision. The result of the image segmentation may directly determine the final analysis quality of the image and the decision result of the pattern recognition. Image segmentation is one of three major tasks of image processing. Image segmentation is a technique of dividing an image into regions each having characteristics and extracting an object of interest. The purpose is to divide the image into non-overlapping regions. Through image segmentation, target extraction, feature extraction and parameter measurement, the image is converted into a more abstract and recognizable data form, so that high-level image analysis and image understanding become possible.
Image segmentation is the basis of image analysis and image understanding, and how to accurately and effectively segment images is always the key point in the fields of computer vision and image processing. The inventor finds in the research and development process that the traditional segmentation method hardly meets the modern requirements due to the inherent ambiguity of the image and the uncertainty in the segmentation process, which increase the complexity of the image segmentation. Many scholars have conducted intensive studies on image segmentation in recent decades, but there is no image segmentation method suitable for all images so far. Since the main basis of image segmentation is the similarity and discontinuity of gray levels, image segmentation algorithms can be divided into region-based and boundary-based segmentation algorithms, as well as segmentation methods that incorporate specific theoretical tools. And the image segmentation is deeply researched, so that the subsequent performance of image processing is improved. Since images are of many different types, a suitable image segmentation method has to be selected in a targeted manner. If the selected image segmentation algorithm is not suitable, the quality of the segmentation result of the image is greatly influenced.
In general, an image segmentation algorithm only considers information such as texture and edge of an image, but does not consider uncertainty information inherent in the image, which easily causes inaccuracy of an image segmentation result. However, noise is inevitable during the process of image acquisition, transmission and storage, so denoising has become an important issue for image processing research. For noise cancellation, many researchers at home and abroad have proposed various methods, and among these methods, playing a particular image in combination with a particular theoretical method usually has a good effect. People who suffer from Hengda and the like apply the theory of middle intelligence to the elimination of image noise to obtain a good effect. Noose is a branch of philosophy that, as a generalization of dialectics, studies the origin, scope and nature of neutrality and interactions with different ideas. Noose is the basis for noose logic, noose probability theory, noose set theory, and noose statistics. The research of the middle wisdom is the relationship between a real thing, a theory, a proposition, a concept or an entity 'A' and the opposite 'Anti-A' thereof, the negation 'Non-A' thereof and the 'Netu-A' (being neither 'A' nor 'Anti-A').
In image processing, ambiguity problems are usually handled by fuzzy logic, however, the ambiguity set cannot handle the uncertainty of the data itself. The intelligent theory is applied to image processing, and the uncertainty information in the image can be clearly quantized. The fuzzy C-means (FCM) algorithm is widely applied due to simple implementation. However, the algorithm is limited by the selection of the initial clustering center, is easy to fall into local optimum, is sensitive to noise, and needs to solve the problem that the FCM initial value falls into a local extreme value.
Disclosure of Invention
Aiming at the defects in the prior art, one or more embodiments of the disclosure provide a method and a device for segmenting a mesopic image based on an improved fuzzy C mean value, wherein a Particle Swarm Optimization (PSO) is applied to fuzzy C mean value clustering, so that the method has the advantages of higher randomness, difficulty in falling into local minimum and higher convergence rate when a next generation solution is generated, thereby improving the anti-noise interference capability and improving the segmentation accuracy of the image.
According to an aspect of one or more embodiments of the present disclosure, there is provided a method for improving blurred C-means based mesopic image segmentation.
A middle-intelligence image segmentation method based on improved fuzzy C mean value comprises the following steps:
receiving an original image to be segmented, and converting the original image into a mesology image;
denoising the middle-intelligence image by adopting alpha-mean operation, and enhancing the image;
calculating the information entropy of the uncertain subset I in the image-enhanced middle-intelligence image;
and when the information entropy calculation result is smaller than a threshold value, performing intelligent image segmentation by adopting a particle swarm optimization fuzzy C-means clustering algorithm, and otherwise, returning to the continuous denoising of the intelligent image.
Further, in the method, the mesopic images include a subset image T, a subset image I, and a subset image F;
the subset image T is expressed as an original image reality expression;
the subset image I is expressed as uncertainty expression of the original image;
the subset image F is represented as a non-authentic representation of the original image.
Further, in the method, the specific steps of converting the original image into the mesology image include:
calculating the subset image T according to the pixel value area mean value of the original image;
calculating the subset image I according to the pixel value region deviation of the original image;
and calculating the subset image F according to the range of the pixel values of the original image.
Further, in the method, the specific formula for denoising the mesology image by adopting alpha-mean operation is as follows:
Figure GDA0002080280920000031
Figure GDA0002080280920000032
Figure GDA0002080280920000033
Figure GDA0002080280920000041
Figure GDA0002080280920000042
Figure GDA0002080280920000043
Figure GDA0002080280920000044
Figure GDA0002080280920000045
wherein the content of the first and second substances,
Figure GDA0002080280920000046
subset imageThe alpha-average value of T is,
Figure GDA0002080280920000047
is the alpha-mean of the subset image I,
Figure GDA0002080280920000048
is the alpha-mean value of the subset image F, T is the set of true values of the original image,
Figure GDA0002080280920000049
the original image T set is converted into a noon set, F is a set of non-true values of the original image,
Figure GDA00020802809200000410
the method comprises the steps of converting an original image F set into a mesointelligence set, wherein I is a set of uncertain values of the original image, alpha generally takes a value of 0.85, w is the size of a limited region, (I, j) is a pixel point of the original image, (m, n) is a domain pixel information point in the w region, T (m, n) is a value of a point (m, n) of the mesointelligence set T, T (I, j) is a value of the point (I, j) of the mesointelligence set T, F (m, n) is a value of the point (m, n) of the mesointelligence set F, and F (I, j) is a value of the point (I, j) of the mesointelligence set T,
Figure GDA00020802809200000411
is the average intensity value of the subset of images T,
Figure GDA00020802809200000412
is the average value of the subset image T after alpha average operation,
Figure GDA00020802809200000413
is the average intensity value of the subset image F,
Figure GDA00020802809200000414
the mean intensity values of the subset image T in the w region,
Figure GDA00020802809200000415
is the average intensity value of the subset image I,
Figure GDA00020802809200000416
is the average intensity value of the pixel (i, j)
Figure GDA00020802809200000417
And after alpha mean operation
Figure GDA00020802809200000418
The absolute value of the difference between the two values,
Figure GDA00020802809200000419
is the average intensity value of the pixel point (i, j)
Figure GDA00020802809200000420
And after alpha mean operation
Figure GDA00020802809200000421
The minimum value of the absolute value of the difference,
Figure GDA00020802809200000422
is the average intensity value of the pixel point (i, j)
Figure GDA00020802809200000423
And after alpha mean operation
Figure GDA00020802809200000424
The maximum value of the absolute value of the difference.
Further, in the method, a calculation formula for performing image enhancement on the denoised mesology image is as follows:
Figure GDA0002080280920000051
Figure GDA0002080280920000052
Figure GDA0002080280920000053
Figure GDA0002080280920000054
Figure GDA0002080280920000055
Figure GDA0002080280920000056
Figure GDA0002080280920000057
wherein:
Figure GDA0002080280920000058
for the set of realisms of the wisdom set subjected to the beta enhancement operation,
Figure GDA0002080280920000059
is as follows
Figure GDA00020802809200000510
The authenticity subset of the noon set after beta enhancement operation,
Figure GDA00020802809200000511
to represent a subset of the uncertainty intelligence of the medium intelligence set, β is typically 0.85,
Figure GDA00020802809200000512
is composed of
Figure GDA00020802809200000513
The true value of the beta-enhancement operation is performed,
Figure GDA00020802809200000514
is the value of the set T after the mean operation,
Figure GDA00020802809200000515
for the non-authentic set of the wisdom set subjected to the beta enhancement operation,
Figure GDA00020802809200000516
is as follows
Figure GDA00020802809200000517
The non-reality subset of the noon set is subjected to beta enhancement operation,
Figure GDA00020802809200000518
is composed of
Figure GDA00020802809200000519
A non-true value of the beta boost operation is performed,
Figure GDA00020802809200000520
the non-real pixel points after the mean value operation,
Figure GDA00020802809200000521
is that
Figure GDA00020802809200000522
δ' (i, j) is the value of the pixel point (i, j) after the image enhancement operation
Figure GDA00020802809200000523
And mean value
Figure GDA00020802809200000524
Absolute value of the difference, δ'minValue after image enhancement operation for pixel point (i, j)
Figure GDA00020802809200000525
And mean value
Figure GDA00020802809200000526
Difference of differenceOf absolute value of δ'maxValue after image enhancement operation for pixel point (i, j)
Figure GDA00020802809200000527
And mean value
Figure GDA00020802809200000528
The maximum value of the absolute value of the difference,
Figure GDA00020802809200000529
as a subset
Figure GDA00020802809200000530
The average intensity in the w x w region,
Figure GDA0002080280920000061
as a subset
Figure GDA0002080280920000062
And (5) performing beta enhancement operation on the authenticity set in the w x w region.
Further, in the method, the information entropy calculation result is
Figure GDA0002080280920000063
Wherein, Enl(i +1) entropy, En, of element uncertainty for the (i +1) th iterationl(i) Entropy of uncertainty of elements for the ith iteration
The subset image T and the subset image I are combined into a new cluster value before image segmentation.
Further, in the method, the specific steps of performing the intelligent image segmentation by adopting the particle swarm optimization fuzzy C-means clustering algorithm include:
initial parameters are received, including a given class number, fuzzy index, cluster size, learning factor, and inertial weight.
Carrying out initialization coding on the N cluster centers to form N first-generation particles;
calculating the center vector of each clustering center and membership degree;
calculating the fitness of each particle, and if the fitness of the particle is better than the fitness of the current optimal position of the particle, updating the optimal position of a single particle; if the fitness of the current global optimal position is better than the fitness of the optimal positions in all the particles, updating the global optimal position;
updating the speed and the position of each particle to generate a next generation particle swarm;
and when the current iteration times reach the preset maximum times, stopping iteration, and performing intelligent image segmentation on the result in the last generation, otherwise, returning to the continuous denoising of the intelligent image.
According to an aspect of one or more embodiments of the present disclosure, there is provided a computer-readable storage medium.
A computer readable storage medium, wherein a plurality of instructions are stored, the instructions are suitable for being loaded by a processor of a terminal device and executing the method for improving the fuzzy C-means-based mesopic image segmentation.
According to an aspect of one or more embodiments of the present disclosure, there is provided a terminal device.
A terminal device comprising a processor and a computer-readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is used for storing a plurality of instructions which are suitable for being loaded by a processor and executing the middle-intelligence image segmentation method based on the improved fuzzy C mean value.
According to an aspect of one or more embodiments of the present disclosure, there is provided an intelligent image segmentation apparatus based on an improved blurred C-means.
The device for segmenting the mesopic image based on the improved fuzzy C mean value comprises the following steps of:
a data acquisition module configured to receive an original image to be segmented;
the image conversion module is configured to convert the original image into a mesool image according to a mesool theory;
the denoising enhancement module is configured to denoise the mesology image by adopting alpha-mean operation and enhance the image;
the image segmentation module is configured to calculate the information entropy of the uncertain subset I in the image-enhanced mesology image; and when the information entropy calculation result is smaller than a threshold value, performing intelligent image segmentation by adopting a particle swarm optimization fuzzy C-means clustering algorithm, and otherwise, returning to the continuous denoising of the intelligent image.
The beneficial effect of this disclosure:
according to the middle-intelligence image segmentation method and device based on the improved fuzzy C mean value, a Particle Swarm Optimization (PSO) is applied to fuzzy C mean value clustering, and the FCM improved by the PSO has high randomness when a next generation solution is generated and is not easy to fall into local minimum; the method has higher convergence rate, improves the anti-noise interference capability through alpha mean value operation and beta image enhancement, and improves the segmentation accuracy of the image.
Drawings
The accompanying drawings, which are included to provide a further understanding of the disclosure, illustrate embodiments of the disclosure and together with the description serve to explain the disclosure and are not to limit the disclosure.
FIG. 1 is a flow diagram of a method for improving blurred C-means based mesopic image segmentation in accordance with one or more embodiments;
FIG. 2 is a detailed flow diagram of a method for improving fuzzy C-means based inter-intelligent image segmentation, according to one or more embodiments;
fig. 3(a) is Lena original image (512 x 512);
FIG. 3(b) an effect graph of a mesoscopic image with an alpha-mean operation and an image enhancement operation;
FIG. 3(C) is based on the fuzzy C-means segmentation effect graph of FIG. 3 (b);
FIG. 3(d) is a graph based on the particle swarm optimization fuzzy C-means segmentation effect of FIG. 3 (b);
fig. 4(a) is a raw image of rice grains (256 × 256);
FIG. 4(b) is a graph of the effect of a mesopic image with an alpha mean operation and an image enhancement operation;
FIG. 4(C) is based on the fuzzy C-means segmentation effect graph of FIG. 4 (b);
fig. 4(d) is a graph based on the particle swarm optimization fuzzy C-means segmentation effect of fig. 4 (b).
The specific implementation mode is as follows:
technical solutions in one or more embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in one or more embodiments of the present disclosure, and it is to be understood that the described embodiments are only a part of the embodiments of the present disclosure, and not all of the embodiments. All other embodiments, which can be derived by one of ordinary skill in the art from one or more embodiments of the disclosure without making any creative effort, shall fall within the scope of protection of the disclosure.
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the disclosure. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure belongs.
It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of example embodiments according to the present disclosure. As used herein, the singular forms "a", "an" and "the" are intended to include the plural forms as well, and it should be understood that when the terms "comprises" and/or "comprising" are used in this specification, they specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof, unless the context clearly indicates otherwise.
It is noted that the flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of the present disclosure. It should be noted that each block in the flowchart or block diagrams may represent a module, a segment, or a portion of code, which may comprise one or more executable instructions for implementing the logical function specified in the respective embodiment. It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Without conflict, the embodiments and features of the embodiments in the present disclosure may be combined with each other, and the present disclosure will be further described with reference to the drawings and the embodiments.
Example one
According to an aspect of one or more embodiments of the present disclosure, there is provided a method for improving blurred C-means based mesopic image segmentation.
As shown in fig. 1, a method for middle-intelligence image segmentation based on improved fuzzy C-means includes:
s1, receiving an original image to be segmented; converting an original image into a mesool image according to a mesool theory;
s2, denoising the middle-intelligence image by adopting alpha-mean operation, and enhancing the image;
s3, calculating the information entropy of the uncertain subset I in the image-enhanced middle-intelligence image;
s4, when the information entropy calculation result is smaller than the threshold value, the particle swarm optimization fuzzy C-means clustering algorithm is adopted to carry out the segmentation of the mesology image, otherwise, the denoising of the mesology image is returned to continue.
(1) Image for middle intelligence
The image (PNS) is the expression of an image in the field of middle wisdom, which is realized by converting the characteristic information of the image into the field of middle wisdom by using the theory of middle wisdom set. Compared with the original image representation mode, the image representation method has the advantages that the image uncertainty is definitely quantized, and the description capability of uncertain information in the image is improved. Furthermore, it is important that the three noose elements T, I and F are mutually opposite. Through T, I and F, image information with different properties and different spaces can be used at the same time, so that the image can be more comprehensively and completely described, and the segmentation effect is improved. Therefore, the selection of the mesopic domain and the definition of the three mesopic elements T, I and F are the core of the segmentation algorithm based on mesopic images.
The middle wisdom image is: given an image as P, assume the full domain of discourse is U, the full set of image pixels is W, and W is a non-empty subset of U. The PNS image has three membership sets T, I and F. The subset image T is expressed as an original image authenticity expression, the subset image I is expressed as an original image uncertainty expression, and the subset image F is expressed as an original image non-authenticity expression.
The characteristic information used by the middle-intelligence image is the gray value of the image pixel point. A pixel P in the image is described as P (t, i, f), belonging to W for the element P (t, i, f) in the following way: t% true, I% uncertain, and F% false, where T ∈ T, I ∈ I, and F ∈ F. The pixel P (i, j) in the image domain is converted into an intellectual region.
PNS(i,j)＝{T(i,j),I(i,j),F(i,j)} (1)
In general, we represent the uncertainty of the data by the standard deviation. When judging whether the measurement result is accurate, the standard deviation of the obtained result plays a decisive role. If most of the measurements oscillate around a certain range, it can be said that the predicted values are reasonable to use the effect of the statistical feature values, and we describe T, I, F with the area mean, deviation and range of pixel values.
In the process of image transformation, the pixel points (i, j) are limited in a small area, and the size of the area is taken as w × w. The transformation process of the middle wisdom image is the acquisition process of the three sets T, I and F. The process is calculated by the following formula:
Figure GDA0002080280920000111
Figure GDA0002080280920000112
Figure GDA0002080280920000113
Figure GDA0002080280920000114
F(i,j)＝1-T(i,j) (6)
wherein:
Figure GDA0002080280920000115
is the local average of the image, and δ (i, j) is the pixel g (i, j) and its local average
Figure GDA0002080280920000116
The absolute value of the difference.
(2) Information entropy of Zhongzhi image
The entropy of an image is the evaluation and description of the image. If the entropy value is larger, the gray values are described as having the same probability and uniform distribution, and conversely, the gray values are described as having different probabilities and distribution asymmetry. In the image of the middle intelligence, the information entropy of the image is used for describing the aggregation characteristics of the gray level distribution of image pixels.
The information entropy of the mesopic image is the sum of the entropies of the three sets T, I and F, and is used for evaluating the distribution of elements in the mesopic domain:
EnNS＝EnT+EnI+EnF (7)
Figure GDA0002080280920000117
Figure GDA0002080280920000118
Figure GDA0002080280920000121
wherein: enT、EnI、EnFThe entropy of sets T, I and F, respectively. p is a radical ofT(i)、pI(i)、pF(i) The probabilities of element i in the three sets T, I and F, respectively. And EnTAnd EnFUsed to describe the distribution of elements in the mesology collection; enITo describe the distribution of the uncertainty of the elements.
According to the concept of mesology, the variation of elements in sets T and F should affect the uncertainty distribution of set I. The larger the entropy of the set T is, and the smaller the entropy of the set I is, the more uniform the gray level distribution of the image is, and the more reasonable the distribution of the gray level values is. In order to increase the entropy of the set T and reduce the entropy of the set I, a new operation, namely alpha-mean operation, is proposed in the image denoising of the mesology, and the operation utilizes the value of the pixel point in the uncertainty set to judge and update the current pixel value in the neural field neural.
The alpha-mean operation is to perform alpha-mean operation on the PNS image, and the PNS (alpha) is as follows:
Figure GDA0002080280920000122
Figure GDA0002080280920000123
Figure GDA0002080280920000124
Figure GDA0002080280920000125
Figure GDA0002080280920000126
Figure GDA0002080280920000127
Figure GDA0002080280920000128
Figure GDA0002080280920000129
wherein the content of the first and second substances,
Figure GDA0002080280920000131
the alpha-mean of the subset of images T,
Figure GDA0002080280920000132
is the alpha-mean of the subset image I,
Figure GDA0002080280920000133
is the alpha-mean value of the subset image F, T is the set of true values of the original image,
Figure GDA0002080280920000134
the original image T set is converted into a noon set, F is a set of non-true values of the original image,
Figure GDA0002080280920000135
the method comprises the steps of converting an original image F set into a mesointelligence set, wherein I is a set of uncertain values of the original image, alpha generally takes a value of 0.85, w is the size of a limited region, (I, j) is a pixel point of the original image, (m, n) is a domain pixel information point in the w region, T (m, n) is a value of a point (m, n) of the mesointelligence set T, T (I, j) is a value of the point (I, j) of the mesointelligence set T, F (m, n) is a value of the point (m, n) of the mesointelligence set F, and F (I, j) is a value of the point (I, j) of the mesointelligence set T,
Figure GDA0002080280920000136
is the average intensity value of the subset of images T,
Figure GDA0002080280920000137
is the average value of the subset image T after alpha average operation,
Figure GDA0002080280920000138
is the average intensity value of the subset image F,
Figure GDA0002080280920000139
the mean intensity values of the subset image T in the w region,
Figure GDA00020802809200001310
is the average intensity value of the subset image I,
Figure GDA00020802809200001311
is the average intensity value of the pixel (i, j)
Figure GDA00020802809200001312
And after alpha mean operation
Figure GDA00020802809200001313
The absolute value of the difference between the two values,
Figure GDA00020802809200001314
is the average intensity value of the pixel point (i, j)
Figure GDA00020802809200001315
And after alpha mean operation
Figure GDA00020802809200001316
The minimum value of the absolute value of the difference,
Figure GDA00020802809200001317
is the average intensity value of the pixel point (i, j)
Figure GDA00020802809200001318
And carry outAfter alpha mean operation
Figure GDA00020802809200001319
The maximum value of the absolute value of the difference.
After the alpha-mean operation, the entropy of the uncertain subset I decreases, and the distribution of elements in I becomes more uneven, and the non-uniformity reduces the uncertainty of the intelligent set PNS. Noise points and high-uncertainty pixel points in the image are reduced, and the distribution of image pixel information is more reasonable and beneficial to the subsequent processing of the image.
(4) Image enhancement operations
In the process of the mesopic image segmentation, the image is converted into the mesopic image, and the alpha mean value operation is carried out, so that the image contour becomes fuzzy, and therefore, the problem needs to be solved by utilizing the concept of image enhancement in the processing process. Image enhancement is the process of enhancing useful information in an image, the main purpose of which is to enhance the visual effect of the image, and possibly also a distortion. Generally, image enhancement is classified into a frequency domain method and a spatial domain method. Among the spatial domain methods, the local averaging method and the median filtering method (median of the domain pixels) are more representative. Directly routing out the gray levels of an image belongs to the scope of the spatial domain. Image enhancement generally highlights desired image information by adding some information to the original or by performing a certain transformation on the data of the original. In the mesopic method, the calculation formula of data enhancement is as follows:
Figure GDA0002080280920000141
Figure GDA0002080280920000142
Figure GDA0002080280920000143
Figure GDA0002080280920000144
Figure GDA0002080280920000145
Figure GDA0002080280920000146
Figure GDA0002080280920000147
wherein:
Figure GDA0002080280920000148
for the set of realisms of the wisdom set subjected to the beta enhancement operation,
Figure GDA0002080280920000149
is as follows
Figure GDA00020802809200001410
The authenticity subset of the noon set after beta enhancement operation,
Figure GDA00020802809200001411
to represent a subset of the uncertainty intelligence of the medium intelligence set, β is typically 0.85,
Figure GDA00020802809200001412
is composed of
Figure GDA00020802809200001413
The true value of the beta-enhancement operation is performed,
Figure GDA00020802809200001414
is the value of the set T after the mean operation,
Figure GDA00020802809200001415
for the non-authentic set of the wisdom set subjected to the beta enhancement operation,
Figure GDA00020802809200001416
is as follows
Figure GDA00020802809200001417
The non-reality subset of the noon set is subjected to beta enhancement operation,
Figure GDA00020802809200001418
is composed of
Figure GDA00020802809200001419
A non-true value of the beta boost operation is performed,
Figure GDA00020802809200001420
the non-real pixel points after the mean value operation,
Figure GDA00020802809200001421
is that
Figure GDA00020802809200001422
δ' (i, j) is the value of the pixel point (i, j) after the image enhancement operation
Figure GDA00020802809200001423
And mean value
Figure GDA0002080280920000151
Absolute value of the difference, δ'minValue after image enhancement operation for pixel point (i, j)
Figure GDA0002080280920000152
And mean value
Figure GDA0002080280920000153
Minimum value of absolute value of the difference, δ'maxValue after image enhancement operation for pixel point (i, j)
Figure GDA0002080280920000154
And mean value
Figure GDA0002080280920000155
The maximum value of the absolute value of the difference,
Figure GDA0002080280920000156
as a subset
Figure GDA0002080280920000157
The average intensity in the w x w region,
Figure GDA0002080280920000158
as a subset
Figure GDA0002080280920000159
And (5) performing beta enhancement operation on the authenticity set in the w x w region.
After enhancement processing, the outline of the image is clearer, and image segmentation is more facilitated.
(5) Fuzzy C-means algorithm
The fuzzy C-means algorithm realizes clustering by optimizing a fuzzy objective function, and unlike K-means clustering, each point can only belong to a certain class, but each point is endowed with membership to each class, so that the membership is used for better describing the characteristics of edge pixels, namely, the edge pixels and the like, and the fuzzy C-means algorithm is suitable for processing the inherent uncertainty of things. The image segmentation is carried out by using the characteristic of Fuzzy C Mean (FCM) unsupervised fuzzy clustering calibration, so that the human intervention can be reduced, and the method is more suitable for the characteristics of uncertainty and fuzziness in the image. The FCM algorithm is extremely sensitive to initial parameters, and sometimes manual intervention parameter initialization is needed to approach a global optimal solution, so that the segmentation speed is increased. In addition, the conventional FCM algorithm does not take spatial information into account and is sensitive to noise and gray scale non-uniformity.
For the noon subset, a new clustering method is defined, and PNS (alpha), namely the average value of alpha and the noon subset after the image enhancement operation, is processed.
Considering the influence of uncertainty, we combine the two sets T and I into a new cluster value.
Figure GDA00020802809200001510
Figure GDA00020802809200001511
An improved fuzzy c-means algorithm for the middle intelligent set is provided. The new objective function is defined as:
Figure GDA0002080280920000161
Figure GDA0002080280920000162
Figure GDA0002080280920000163
the FCM algorithm obtains fuzzy classification of the sample set by performing iterative optimization on the objective function.
The algorithm is sensitive to the initial value and depends largely on the choice of initial cluster center. When the initial cluster center deviates significantly from the global optimal cluster center, the FCM tends to fall into a local minimum. The disadvantage is more pronounced when the number of clusters is larger.
(6) Improved FCM
In order to solve the problems of the FCM algorithm, a Particle Swarm Optimization (PSO) is applied to the fuzzy C-means clustering.
Particle Swarm Optimization (PSO) is a population-based optimization algorithm that moves individuals to good areas according to their environmental suitability. However, it does not use evolutionary operators for individuals, but instead treats each individual as a particle with no volume in the D-dimensional search space, flying at a certain speed in the search space. This speed is dynamically adjusted based on the flight experience of the person and the flight experience of the group of companions. The particle i is represented by Xi ═ (Xi1, Xi2, …, xiD). xiD) and the best location it experiences (with best suitability) is labeled Pi ═ (Pi1, Pi2, …, piD), also known as pbesd. The number of indices of the best position experienced by all particles of a group is denoted by the symbol g, i.e. Pg, also called gbeat. The velocity of the particle i is represented by Vi (Vi1, Vi2, …, viD). For each generation, it is D-dimensional (1. ltoreq. D. ltoreq. D) is varied by the following equation
vid(t+1)＝wvid(t)+c1r1(pbestij-xid(t))+c2r2(gbest-xid(t)) (31)
xid(t+1)＝xid(t)+vid(t+1) (30)
Where w is the inertial weight, generally decreasing linearly from 0.9 to 0.2; c1 and c2 are acceleration constants; r1 and r2 are random functions that vary over a range of [0,1 ].
In the application of the particle swarm optimization fuzzy C-means clustering (PSOFCM) algorithm, the fitness function of each individual in the PSO is defined as follows:
f(xi)＝Jm(U,C) (32)
Figure GDA0002080280920000171
(7) intelligent image segmentation based on PSOFCM algorithm
As shown in fig. 2, the present embodiment proposes a new intermediate intelligent image segmentation algorithm, PSOFCM algorithm, which includes the following steps:
step 1: inputting an image;
step 2: transforming the image into the NS domain using equations (1) - (6);
and step 3: performing an alpha mean operation using equations (11) - (18);
and 4, step 4: performing a beta image enhancement operation using equations (19) - (24);
and 5: calculating the entropy of the indefinite subset I using equation (9);
step 6: if it is not
Figure GDA0002080280920000172
Entering step 7; otherwise, entering step 3;
and 7: applying the improved FCM algorithm to the set of nooses:
Figure GDA0002080280920000181
given the number of classes C, fuzzy index m, cluster size N, learning factors C1 and C2: the inertial weight w.
Figure GDA0002080280920000182
And carrying out initialization coding on the N cluster centers to form N first-generation particles. The number of cluster centers corresponds to the dimension of the particle. Pbest for each particle is its current location, and gbest is the best location for all particles in the current population.
Figure GDA0002080280920000183
Calculating a center vector U (k) of each cluster center C (k) and the degree of membership in k steps using equations (29) - (30);
Figure GDA0002080280920000184
the fitness of each particle is calculated according to equation (33). And if the fitness of the particle is better than the fitness of the current optimal position of the particle, updating the optimal position of the single particle. If the fitness of the current global optimal position is better than the fitness of the optimal positions in all the particles, the global optimal position is updated.
Figure GDA0002080280920000185
The velocity and position of each particle is updated using equations (31) - (32) to generate the next generation of particle clusters.
Figure GDA0002080280920000186
If the current iteration number reaches the previous settingAnd setting the maximum times, and stopping iteration. Finding the best solution in the last generation, otherwise repeating step (3).
And 8: and obtaining an image segmentation result.
The embodiment provides a middle-intelligence image segmentation method based on improved fuzzy c-means clustering. The feasibility of the algorithm and better noise interference resistance are verified through experiments, and the image characteristic information is more reasonably utilized during image processing. The obtained objective function is small, and the segmentation boundary is clear. The method can effectively solve the problems that the traditional FCM algorithm is sensitive to initialization and is easy to fall into a local minimum value, and has the characteristics of strong global search capability, high convergence speed and better image segmentation effect than the traditional FCM algorithm.
In the embodiment, the PSOFCM segmentation algorithm is applied to the mesopic image subjected to denoising of various real images, and is compared with the mesopic denoising image segmentation method based on the fuzzy C mean value.
Fig. 3(a) is Lena original image (512 x 512); FIG. 3(b) an effect graph of a mesoscopic image with an alpha-mean operation and an image enhancement operation; FIG. 3(C) is based on the fuzzy C-means segmentation effect graph of FIG. 3 (b); FIG. 3(d) is a graph based on the particle swarm optimization fuzzy C-means segmentation effect of FIG. 3 (b); table 1 shows the cluster centers obtained by FCM algorithm segmentation (fig. 3(c)) and the cluster centers obtained by PSOFCM algorithm segmentation (fig. 3 (d)).
TABLE 1
FCM PSOFCM
0.691304 0.704806
0.126243 0.125010
0.522027 0.535976
0.294190 0.418385
0.356919 0.739954
In both the FCM image segmentation method and the PSOFCM image segmentation method, the cluster center set in this embodiment is 5, so as to compare them with each other. Fig. 4(a) is a raw image of rice grains (256 × 256); FIG. 4(b) is a graph of the effect of a mesopic image with an alpha mean operation and an image enhancement operation; FIG. 4(C) is based on the fuzzy C-means segmentation effect graph of FIG. 4 (b); fig. 4(d) is a graph based on the particle swarm optimization fuzzy C-means segmentation effect of fig. 4 (b). Table 2 cluster centers obtained by FCM algorithm segmentation (fig. 4(c)) and cluster centers obtained by PSOFCM algorithm segmentation (fig. 4 (d)). Table 3 shows the difference between the two graphs on the objective function.
TABLE 2
FCM PSOFCM
0.2726 0.1257
0.6021 0.2642
0.4125 0.6067
0.1679 0.3710
0.6199 0.5164
TABLE 3
The objective function FCM PSOFCM
Figure.1 0.003000 0.002518
Figure.2 0.002361 0.001233
The final effect to be achieved by clustering is that the intra-class similarity is maximum, the inter-class similarity is minimum, and the sum of the weighted distances of the point and the center at this time is minimum. We can just make the value of the objective function as small as possible. As can be seen from the experimental data in fig. 3 and fig. 4, when PSOFCM is applied to the segmentation of the mesopic image, the obtained objective function is smaller than the objective function value obtained by FCM applied to the segmentation of the mesopic image.
Example two
According to an aspect of one or more embodiments of the present disclosure, there is provided a computer-readable storage medium.
A computer readable storage medium, wherein a plurality of instructions are stored, the instructions are suitable for being loaded by a processor of a terminal device and executing the method for improving the fuzzy C-means-based mesopic image segmentation.
EXAMPLE III
According to an aspect of one or more embodiments of the present disclosure, there is provided a terminal device.
A terminal device comprising a processor and a computer-readable storage medium, the processor being configured to implement instructions; the computer readable storage medium is used for storing a plurality of instructions which are suitable for being loaded by a processor and executing the middle-intelligence image segmentation method based on the improved fuzzy C mean value.
These computer-executable instructions, when executed in a device, cause the device to perform methods or processes described in accordance with various embodiments of the present disclosure.
In the present embodiments, a computer program product may include a computer-readable storage medium having computer-readable program instructions embodied thereon for performing various aspects of the present disclosure. The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic memory device, a magnetic memory device, an optical memory device, an electromagnetic memory device, a semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.
The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device over a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.
The computer program instructions for carrying out operations of the present disclosure may be assembler instructions, instruction set architecture (isa) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, the electronic circuitry can execute computer-readable program instructions to implement aspects of the present disclosure by utilizing state information of the computer-readable program instructions to personalize the electronic circuitry, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA).
Example four
According to an aspect of one or more embodiments of the present disclosure, there is provided an intelligent image segmentation apparatus based on an improved blurred C-means.
The device for segmenting the mesopic image based on the improved fuzzy C mean value comprises the following steps of:
a data acquisition module configured to receive an original image to be segmented;
the image conversion module is configured to convert the original image into a mesool image according to a mesool theory;
the denoising enhancement module is configured to denoise the mesology image by adopting alpha-mean operation and enhance the image;
the image segmentation module is configured to calculate the information entropy of the uncertain subset I in the image-enhanced mesology image; and when the information entropy calculation result is smaller than a threshold value, performing intelligent image segmentation by adopting a particle swarm optimization fuzzy C-means clustering algorithm, and otherwise, returning to the continuous denoising of the intelligent image.
It should be noted that although several modules or sub-modules of the device are mentioned in the above detailed description, such division is merely exemplary and not mandatory. Indeed, the features and functionality of two or more of the modules described above may be embodied in one module in accordance with embodiments of the present disclosure. Conversely, the features and functions of one module described above may be further divided into embodiments by a plurality of modules.
The above description is only a preferred embodiment of the present disclosure and is not intended to limit the present disclosure, and various modifications and changes may be made to the present disclosure by those skilled in the art. Any modification, equivalent replacement, improvement and the like made within the spirit and principle of the present disclosure should be included in the protection scope of the present disclosure. Thus, the present disclosure is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims (7)

1. A middle-intelligence image segmentation method based on improved fuzzy C mean value is characterized by comprising the following steps:
receiving an original image to be segmented, and converting the original image into a mesology image;
denoising the intermediate intelligent image by adopting alpha-mean value operation, and enhancing the image by alpha-mean value and beta enhancement operation; calculating the information entropy of the uncertain subset I in the image-enhanced middle-intelligence image; in the method, the specific formula for denoising the mesology image by adopting alpha-mean operation is as follows:
Figure FDA0003160251900000011
Figure FDA0003160251900000012
Figure FDA0003160251900000013
Figure FDA0003160251900000014
Figure FDA0003160251900000015
Figure FDA0003160251900000016
Figure FDA0003160251900000017
Figure FDA0003160251900000018
wherein the content of the first and second substances,
Figure FDA0003160251900000019
the alpha-mean of the subset of images T,
Figure FDA00031602519000000110
is the alpha-mean of the subset image I,
Figure FDA00031602519000000111
is the alpha-mean value of the subset image F, T is the set of true values of the original image,
Figure FDA00031602519000000112
the original image T set is converted into a noon set, F is a set of non-true values of the original image,
Figure FDA00031602519000000113
the method comprises the steps of converting an original image F set into a mesointelligence set, wherein I is a set of uncertain values of the original image, alpha generally takes a value of 0.85, w is the size of a limited region, (I, j) is a pixel point of the original image, (m, n) is a domain pixel information point in the w region, T (m, n) is a value of a point (m, n) of the mesointelligence set T, T (I, j) is a value of the point (I, j) of the mesointelligence set T, F (m, n) is a value of the point (m, n) of the mesointelligence set F, and F (I, j) is a value of the point (I, j) of the mesointelligence set T,
Figure FDA0003160251900000021
is the average intensity value of the subset of images T,
Figure FDA0003160251900000022
is the average value of the subset image T after alpha average operation,
Figure FDA0003160251900000023
is the average intensity value of the subset image F,
Figure FDA0003160251900000024
the mean intensity values of the subset image T in the w region,
Figure FDA0003160251900000025
is the average intensity value of the subset image I,
Figure FDA0003160251900000026
is the average intensity value of the pixel (i, j)
Figure FDA0003160251900000027
And after alpha mean operation
Figure FDA0003160251900000028
The absolute value of the difference between the two values,
Figure FDA0003160251900000029
is the average intensity value of the pixel point (i, j)
Figure FDA00031602519000000210
And after alpha mean operation
Figure FDA00031602519000000211
The minimum value of the absolute value of the difference,
Figure FDA00031602519000000212
is the average intensity value of the pixel point (i, j)
Figure FDA00031602519000000213
And after alpha mean operation
Figure FDA00031602519000000214
The maximum value of the absolute value of the difference;
in the method, a calculation formula for performing image enhancement on the denoised mesology image is as follows:
Figure FDA00031602519000000215
Figure FDA00031602519000000216
Figure FDA00031602519000000217
Figure FDA00031602519000000218
Figure FDA00031602519000000219
Figure FDA00031602519000000220
Figure FDA00031602519000000221
wherein:
Figure FDA00031602519000000222
for the set of realisms of the wisdom set subjected to the beta enhancement operation,
Figure FDA00031602519000000223
is as follows
Figure FDA00031602519000000224
The authenticity subset of the noon set after beta enhancement operation,
Figure FDA00031602519000000225
to represent a subset of the uncertainty intelligence of the medium intelligence set, β is typically 0.85,
Figure FDA0003160251900000031
is composed of
Figure FDA0003160251900000032
The true value of the beta-enhancement operation is performed,
Figure FDA0003160251900000033
is the value of the set T after the mean operation,
Figure FDA0003160251900000034
for the non-authentic set of the wisdom set subjected to the beta enhancement operation,
Figure FDA0003160251900000035
is as follows
Figure FDA0003160251900000036
The non-reality subset of the noon set is subjected to beta enhancement operation,
Figure FDA0003160251900000037
is composed of
Figure FDA0003160251900000038
A non-true value of the beta boost operation is performed,
Figure FDA0003160251900000039
the non-real pixel points after the mean value operation,
Figure FDA00031602519000000310
is that
Figure FDA00031602519000000311
δ' (i, j) is the value of the pixel point (i, j) after the image enhancement operation
Figure FDA00031602519000000312
And mean value
Figure FDA00031602519000000313
Absolute value of the difference, δ'minValue after image enhancement operation for pixel point (i, j)
Figure FDA00031602519000000314
And mean value
Figure FDA00031602519000000315
Minimum value of absolute value of the difference, δ'maxValue after image enhancement operation for pixel point (i, j)
Figure FDA00031602519000000316
And mean value
Figure FDA00031602519000000317
The maximum value of the absolute value of the difference,
Figure FDA00031602519000000318
as a subset
Figure FDA00031602519000000319
The average intensity in the w x w region,
Figure FDA00031602519000000320
as a subset
Figure FDA00031602519000000321
Performing beta enhancement operation on the authenticity set in the w x w region;
defining a new clustering method aiming at uncertain subsets in an image-enhanced mesology image, processing PNS (alpha), namely alpha mean and the image-enhanced mesology subset, combining two sets T and I into a new clustering value in consideration of the influence of uncertainty, providing an improved fuzzy C-mean algorithm of the mesology set, defining a new target function, and performing iterative optimization on the target function by the fuzzy C-mean algorithm to obtain fuzzy classification of a sample set; when the information entropy calculation result is smaller than a threshold value, performing intelligent image segmentation by adopting a particle swarm optimization fuzzy C-means clustering algorithm, and otherwise, returning to denoising of the continuous intelligent image; the new objective function of the improved fuzzy c-means algorithm of the middle intelligent set is defined as follows:
Figure FDA00031602519000000322
Figure FDA0003160251900000041
Figure FDA0003160251900000042
wherein C represents a given number of classes; m represents a blur index;
the specific steps of performing the intelligent image segmentation by adopting the particle swarm optimization fuzzy C-means clustering algorithm comprise:
receiving initial parameters, wherein the initial parameters comprise a given class number, a fuzzy index, a group size, a learning factor and an inertia weight;
carrying out initialization coding on the N cluster centers to form N first-generation particles;
calculating the center vector of each clustering center and membership degree;
calculating the fitness of each particle, and if the fitness of the particle is better than the fitness of the current optimal position of the particle, updating the optimal position of a single particle; if the fitness of the current global optimal position is better than the fitness of the optimal positions in all the particles, updating the global optimal position;
updating the speed and the position of each particle to generate a next generation particle swarm;
and when the current iteration times reach the preset maximum times, stopping iteration, and performing intelligent image segmentation on the result in the last generation, otherwise, returning to the continuous denoising of the intelligent image.
2. The method for improving blurred C mean value-based mesopic image segmentation as claimed in claim 1, wherein in the method, the mesopic image comprises a subset image T, a subset image I and a subset image F;
the subset image T is expressed as an original image reality expression;
the subset image I is expressed as uncertainty expression of the original image;
the subset image F is represented as a non-authentic representation of the original image.
3. The method as claimed in claim 2, wherein the step of converting the original image into the mesopic image comprises:
calculating the subset image T according to the pixel value area mean value of the original image;
calculating the subset image I according to the pixel value region deviation of the original image;
and calculating the subset image F according to the range of the pixel values of the original image.
4. The method as claimed in claim 1, wherein the entropy calculation is performed by using a mean value of the fuzzy C-meansThe fruit is
Figure FDA0003160251900000051
Wherein, Enl(i +1) entropy, En, of element uncertainty for the (i +1) th iterationl(i) Entropy, which is element uncertain for the ith iteration;
the subset image T and the subset image I are combined into a new cluster value before image segmentation.
5. A computer-readable storage medium, in which a plurality of instructions are stored, wherein the instructions are adapted to be loaded by a processor of a terminal device and to perform a method for improving blurred C-means based inter-intelligent image segmentation according to any of claims 1 to 4.
6. A terminal device comprising a processor and a computer-readable storage medium, the processor being configured to implement instructions; a computer-readable storage medium storing a plurality of instructions adapted to be loaded by a processor and to perform a method of improving blurred C-means based inter-intelligent image segmentation according to any of claims 1 to 4.
7. An improved blurred C mean value based mesopic image segmentation device, which is characterized in that the improved blurred C mean value based mesopic image segmentation method as claimed in any one of claims 1-4 comprises the following steps:
a data acquisition module configured to receive an original image to be segmented;
the image conversion module is configured to convert the original image into a mesool image according to a mesool theory;
the denoising enhancement module is configured to denoise the mesology image by adopting alpha-mean operation and enhance the image;
the image segmentation module is configured to calculate the information entropy of the uncertain subset I in the image-enhanced mesology image; and when the information entropy calculation result is smaller than a threshold value, performing intelligent image segmentation by adopting a particle swarm optimization fuzzy C-means clustering algorithm, and otherwise, returning to the continuous denoising of the intelligent image.
CN201910376778.5A 2019-05-07 2019-05-07 Middle-intelligence image segmentation method and device based on improved fuzzy C-means Active CN110111343B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910376778.5A CN110111343B (en) 2019-05-07 2019-05-07 Middle-intelligence image segmentation method and device based on improved fuzzy C-means

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910376778.5A CN110111343B (en) 2019-05-07 2019-05-07 Middle-intelligence image segmentation method and device based on improved fuzzy C-means

Publications (2)

Publication Number Publication Date
CN110111343A CN110111343A (en) 2019-08-09
CN110111343B true CN110111343B (en) 2021-08-31

Family

ID=67488541

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910376778.5A Active CN110111343B (en) 2019-05-07 2019-05-07 Middle-intelligence image segmentation method and device based on improved fuzzy C-means

Country Status (1)

Country Link
CN (1) CN110111343B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110751662B (en) * 2019-10-17 2022-10-25 齐鲁工业大学 Image segmentation method and system for quantum-behaved particle swarm optimization fuzzy C-means
CN110826472B (en) * 2019-11-01 2023-06-27 新疆大学 Image detection method and device
CN111798658A (en) * 2019-11-08 2020-10-20 方勤 Traffic lane passing efficiency detection platform
CN111080580B (en) * 2019-11-29 2023-06-09 山东大学 Ultrasonic breast tumor rapid threshold segmentation method and system based on Zhongzhi set
CN110956637B (en) * 2019-12-02 2022-08-02 兰州交通大学 Adaptive fuzzy C-means clustering noise image segmentation method and system
CN111833276A (en) * 2020-07-20 2020-10-27 浙江大华技术股份有限公司 Image median filtering processing method and device
US11788972B2 (en) 2021-04-29 2023-10-17 Industrial Technology Research Institute Method of automatically setting optical parameters and automated optical inspection system using the same
CN117496309B (en) * 2024-01-03 2024-03-26 华中科技大学 Building scene point cloud segmentation uncertainty evaluation method and system and electronic equipment

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103593855A (en) * 2013-12-04 2014-02-19 西安电子科技大学 Clustered image splitting method based on particle swarm optimization and spatial distance measurement
CN103942799A (en) * 2014-04-25 2014-07-23 哈尔滨医科大学 Breast ultrasounography image segmentation method and system
CN106023245A (en) * 2016-04-28 2016-10-12 绍兴文理学院 Static background moving object detection method based on neutrosophy set similarity measurement
CN108898602A (en) * 2018-06-27 2018-11-27 南京邮电大学 A kind of FCM medical image cutting method based on improvement QPSO

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015127464A1 (en) * 2014-02-24 2015-08-27 H. Lee Moffitt Cancer Center And Research Institute, Inc. Methods and systems for performing segmentation and registration of images using neutrosophic similarity scores
CN105227645A (en) * 2015-09-15 2016-01-06 齐鲁工业大学 A kind of cloud data migration method
CN109145921B (en) * 2018-08-29 2021-04-09 江南大学 Image segmentation method based on improved intuitive fuzzy C-means clustering

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103593855A (en) * 2013-12-04 2014-02-19 西安电子科技大学 Clustered image splitting method based on particle swarm optimization and spatial distance measurement
CN103942799A (en) * 2014-04-25 2014-07-23 哈尔滨医科大学 Breast ultrasounography image segmentation method and system
CN106023245A (en) * 2016-04-28 2016-10-12 绍兴文理学院 Static background moving object detection method based on neutrosophy set similarity measurement
CN108898602A (en) * 2018-06-27 2018-11-27 南京邮电大学 A kind of FCM medical image cutting method based on improvement QPSO

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
Computational intelligence optimization approach based on particle swarm optimizer and neutrosophic set for abdominal CT liver tumor segmentation;Ahmed M.Anter et al.;《Journal of Computational Science》;20180331;第25卷;第376-387页 *
基于能量的图像分割的应用及研究;赵鑫;《中国优秀硕士学位论文全文数据库 信息科技辑》;20120715;第19-30页 *

Also Published As

Publication number Publication date
CN110111343A (en) 2019-08-09

Similar Documents

Publication Publication Date Title
CN110111343B (en) Middle-intelligence image segmentation method and device based on improved fuzzy C-means
Deng et al. Adaptive top-hat filter based on quantum genetic algorithm for infrared small target detection
Chander et al. A new social and momentum component adaptive PSO algorithm for image segmentation
Li et al. Background data resampling for outlier-aware classification
Sathya et al. PSO-based Tsallis thresholding selection procedure for image segmentation
Shanmugavadivu et al. Particle swarm optimized multi-objective histogram equalization for image enhancement
Zhao et al. Optimal-selection-based suppressed fuzzy c-means clustering algorithm with self-tuning non local spatial information for image segmentation
Khokher et al. Image segmentation using multilevel graph cuts and graph development using fuzzy rule‐based system
JP2014041476A (en) Image processing apparatus, image processing method, and program
Chiranjeevi et al. New fuzzy texture features for robust detection of moving objects
CN111986125A (en) Method for multi-target task instance segmentation
Zhang et al. Superpixel generation for SAR imagery based on fast DBSCAN clustering with edge penalty
Dorrani et al. Noisy images edge detection: Ant colony optimization algorithm
Arish et al. Extraction of active regions and coronal holes from EUV images using the unsupervised segmentation method in the Bayesian framework
Reddy et al. A hybrid K-means algorithm improving low-density map-based medical image segmentation with density modification
Liu et al. Active contour driven by adaptive‐scale local‐energy signed pressure force function based on bias correction for medical image segmentation
Yuan et al. Deep variational instance segmentation
Xu et al. Vehicle detection based on improved multitask cascaded convolutional neural network and mixed image enhancement
CN114202746B (en) Pavement state identification method, device, terminal equipment and storage medium
Zhang et al. Solving the long-tailed problem via intra-and inter-category balance
CN110751662B (en) Image segmentation method and system for quantum-behaved particle swarm optimization fuzzy C-means
CN110310297B (en) Image segmentation method and system based on multi-resolution search particle swarm algorithm
CN114399635A (en) Image two-classification ensemble learning method based on feature definition and deep learning
Li et al. Image object detection algorithm based on improved Gaussian mixture model
Jiao et al. Salient region growing based on Gaussian pyramid

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
TR01 Transfer of patent right
TR01 Transfer of patent right

Effective date of registration: 20240108

Address after: Building 9, North Building 404-1, Shuntai Plaza, No. 2000 Shunhua Road, Jinan Area, China (Shandong) Pilot Free Trade Zone, Jinan City, Shandong Province, 250000

Patentee after: Shandong Huizhi Education Technology Development Co.,Ltd.

Address before: 250353 University Road, Changqing District, Ji'nan, Shandong Province, No. 3501

Patentee before: Qilu University of Technology