CN117953495A - Oral cavity microbial flora segmentation method - Google Patents
Oral cavity microbial flora segmentation method Download PDFInfo
- Publication number
- CN117953495A CN117953495A CN202410355036.5A CN202410355036A CN117953495A CN 117953495 A CN117953495 A CN 117953495A CN 202410355036 A CN202410355036 A CN 202410355036A CN 117953495 A CN117953495 A CN 117953495A
- Authority
- CN
- China
- Prior art keywords
- initial
- microorganism
- category
- hue
- microorganisms
- 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
- 238000000034 method Methods 0.000 title claims abstract description 85
- 230000000813 microbial effect Effects 0.000 title claims abstract description 56
- 210000000214 mouth Anatomy 0.000 title claims abstract description 55
- 230000011218 segmentation Effects 0.000 title claims abstract description 14
- 244000005700 microbiome Species 0.000 claims abstract description 445
- 230000002159 abnormal effect Effects 0.000 claims abstract description 43
- 230000008859 change Effects 0.000 claims abstract description 10
- 241000894007 species Species 0.000 claims description 83
- 238000004364 calculation method Methods 0.000 claims description 29
- 241000736262 Microbiota Species 0.000 claims description 20
- 238000012937 correction Methods 0.000 claims description 20
- 230000008569 process Effects 0.000 claims description 15
- 230000005856 abnormality Effects 0.000 claims description 12
- 230000003247 decreasing effect Effects 0.000 claims description 10
- 238000000638 solvent extraction Methods 0.000 claims description 8
- 238000013507 mapping Methods 0.000 claims description 2
- 238000003709 image segmentation Methods 0.000 abstract description 2
- 239000000203 mixture Substances 0.000 description 3
- 238000012544 monitoring process Methods 0.000 description 3
- 208000025157 Oral disease Diseases 0.000 description 2
- 230000002776 aggregation Effects 0.000 description 2
- 238000004220 aggregation Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 208000030194 mouth disease Diseases 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000894006 Bacteria Species 0.000 description 1
- 208000035473 Communicable disease Diseases 0.000 description 1
- 208000002064 Dental Plaque Diseases 0.000 description 1
- 241000233866 Fungi Species 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 239000003242 anti bacterial agent Substances 0.000 description 1
- 229940088710 antibiotic agent Drugs 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 210000002421 cell wall Anatomy 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 208000002925 dental caries Diseases 0.000 description 1
- 239000000975 dye Substances 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 244000005706 microflora Species 0.000 description 1
- 210000002200 mouth mucosa Anatomy 0.000 description 1
- 239000002324 mouth wash Substances 0.000 description 1
- 229940051866 mouthwash Drugs 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 201000001245 periodontitis Diseases 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 239000000243 solution Substances 0.000 description 1
- 230000003595 spectral effect Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000011269 treatment regimen Methods 0.000 description 1
Landscapes
- Measuring Or Testing Involving Enzymes Or Micro-Organisms (AREA)
Abstract
The invention relates to the technical field of image segmentation, in particular to a method for segmenting an oral microbial flora community. Dividing an oral cavity microorganism flora image to obtain an initial category of microorganisms; obtaining the possibility that the initial category belongs to the same category of microorganism according to the pixel point quantity change and the hue difference between the initial categories; obtaining the abnormal possibility of each pixel point in the initial category according to the hue difference and the possibility between the pixel points in the initial category; and obtaining shape indexes of the initial categories according to the shape characteristics of the connected domain and the abnormal possibility of the internal pixel points, further obtaining the similarity between the initial categories, further obtaining the clustering distance between the initial categories, and dividing the oral microbial flora. The method combines the position distribution and the color information of the microorganisms, determines the similarity between the initial categories of the microorganisms, and improves the segmentation accuracy of the oral cavity microorganism flora.
Description
Technical Field
The invention relates to the technical field of image segmentation, in particular to a method for segmenting an oral microbial flora community.
Background
Since microorganisms in the oral cavity are closely related to oral health, unbalance of the microorganisms may cause oral diseases such as dental plaque, periodontitis, caries, and the like. By monitoring the type and quantity of microorganisms in the oral cavity, the risk of oral diseases can be found early and corresponding prevention and treatment strategies can be formulated. Also, for some oral infectious diseases, monitoring microorganisms can also help select effective antibiotics and oral care products, while partitioning the microflora can help identify different species of microorganisms to understand the composition of the oral microbiota, including the distribution of bacteria, fungi, and other microorganisms. Thus, monitoring of microorganisms in the oral cavity is required.
In the prior art, when an oral cavity microorganism image is segmented by using a clustering algorithm, clustering is often performed only according to the similarity between the gray values of microorganisms under different categories, and the position distribution of the microorganisms is ignored, so that the microorganisms of different categories are divided into the same cluster, or the microorganisms of the same category but with certain color difference are divided into different clusters, and the segmentation of microbial colonies in the oral cavity is seriously interfered.
Disclosure of Invention
In order to solve the technical problems that in the prior art, when an oral cavity microorganism image is segmented by using a clustering algorithm, clustering is often carried out only according to the similarity between microorganism gray values under different categories, and the position distribution of microorganisms is ignored, so that the microorganisms of different categories are divided into the same cluster, or the microorganisms of the same category but with certain color difference are divided into different clusters, and the segmentation of microbial colonies in the oral cavity is seriously interfered, the invention aims to provide an oral cavity microorganism flora segmentation method, which adopts the following technical scheme:
a method of oral microbiota community segmentation, the method comprising:
acquiring an oral microbial flora image;
Dividing the oral cavity microbial flora image according to the color characteristics of the oral cavity microbial flora image to obtain all initial categories of microorganisms; obtaining the possibility that any two microorganism initial categories belong to the same microorganism according to the quantity change characteristics and the hue difference degree of the pixel points between the any two microorganism initial categories;
Obtaining the abnormal possibility of each pixel point in each microorganism initial category according to the possibility distribution difference and the hue distribution difference of each pixel point and other pixel points in each microorganism initial category in a preset adjacent area;
Obtaining the shape index of each microorganism initial category according to the abnormal possibility of the pixel points in each communicating domain and the shape characteristics of the communicating domains in each microorganism initial category; obtaining the similarity between any two initial categories of microorganisms according to the shape index difference and the hue difference between any two initial categories of microorganisms; obtaining a clustering distance when the two microorganism initial categories are clustered according to the similarity of the two microorganism initial categories; and dividing the oral cavity microbial flora according to the clustering distance.
Further, the method for acquiring the initial category of the microorganism comprises the following steps:
mapping the oral cavity microorganism flora image into an HSL color space to obtain a microorganism hue space;
and uniformly dividing the microbial hue space into 360 areas according to hue, and taking all pixel points corresponding to the oral cavity microbial flora image in each area as an initial microbial category.
Further, the method for acquiring the possibility that any two kinds of microorganisms belong to the same kind of microorganisms comprises the following steps:
Taking the hue average value of the pixel points in the initial microorganism category as an abscissa and the number of the pixel points in the initial microorganism category as an ordinate to obtain hue curves formed by all the initial microorganism categories;
in the hue curve, calculating the monotonically decreasing frequency and the monotonically increasing frequency of the number of the pixel points between any two microorganism initial categories;
acquiring initial possibility of microorganisms belonging to the same category between any two initial categories of microorganisms according to the monotonically decreasing number frequency and the monotonically increasing number frequency;
Obtaining a correction coefficient of the initial possibility according to the hue difference degree between any two microorganism initial categories;
and taking the product of the correction coefficient and the initial probability as the probability that any two microorganism initial categories belong to the same microorganism category.
Further, the method for acquiring the initial likelihood includes:
the initial likelihood is obtained according to an initial likelihood calculation formula, wherein the initial likelihood calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>Initial class and/>, of species microorganismThe initial likelihood that an initial class of species of microorganism belongs to the same class of microorganism; /(I)Represents the/>, in the hue curveInitial class and/>, of species microorganismAn initial microorganism initial category between species microorganism initial categories; /(I)Represents the/>, in the hue curveInitial class and/>, of species microorganismTerminal microorganism initial categories between species microorganism initial categories; /(I)Expressed in hue curve as-Initial class and/>, of species microorganismA serial number of all initial categories of microorganisms between initial categories of species microorganisms; /(I)Represents the/>Color phase average value of pixel points in the initial category of the seed microorganisms; /(I)Represents the/>Color phase average value of pixel points in the initial category of the seed microorganisms; /(I)As an indication function; /(I)Representing a quantitative function; /(I)Representing the maximum function.
Further, the method for obtaining the correction coefficient includes:
In the hue curve, obtaining a hue maximum value with the smallest hue average value difference from pixel points in the initial category of microorganisms as a hue reference value of each initial category of microorganisms; taking the inverse of the hue reference value difference between any two microorganism initial categories as a hue reference coefficient; taking the reciprocal of the average difference of the hue between any two initial categories of microorganisms as a hue difference coefficient; and taking the sum of the hue reference coefficient and the hue difference coefficient as a correction coefficient of the initial possibility.
Further, the method for acquiring the abnormal possibility comprises the following steps:
Obtaining the possibility of abnormality according to an abnormality possibility calculation formula, wherein the possibility of abnormality calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>A serial number of the target pixel point in the initial category of the seed microorganism; /(I)Represents the/>Abnormal probability of target pixel points in the initial category of the seed microorganism; /(I)Represents the/>The number of other pixels in the initial category of species microorganism; represents the/> Serial numbers of other pixels in the initial category of the seed microorganism; /(I)Represents the/>The number of pixels in each pixel neighborhood window in the initial category of the seed microorganism; /(I)Represents the/>Sequence numbers of pixel points in each pixel point neighborhood window in the initial category of the seed microorganism; /(I)Expressed as/>First/>, in a neighborhood window established by taking target pixel points in initial category of species microorganisms as centersA plurality of pixel points; /(I)Expressed as/>First/>, of the species microorganism initial categoryFirst/>, in a neighborhood window built by taking other pixel points as centersA plurality of pixel points; /(I)Representation/>Hue values of (2); /(I)Representation/>Hue values of (2); /(I)Representation/>And (3) withThe possibility of belonging to the same species of microorganism.
Further, the method for obtaining the shape index comprises the following steps:
Calculating the credibility of each connected domain in each microorganism initial category;
calculating the maximum distance and the minimum distance between pixel points in each connected domain in each microorganism initial category;
taking the ratio of the maximum distance to the minimum distance as the distance coefficient of each connected domain in each microorganism initial category;
Taking the ratio of the number of boundary pixels of each connected domain in each microorganism initial class to the total number of the pixels of the connected domains as the shape coefficient of each connected domain in each microorganism initial class;
Taking the ratio between the distance coefficient and the shape coefficient as the shape characteristic of each connected domain in each microorganism initial category;
And accumulating and summing the products of the credibility of all the connected domains in each microorganism initial category and the shape characteristics to obtain the shape index of each microorganism initial category.
Further, the method for obtaining the credibility comprises the following steps:
and normalizing the inverse of the sum value of the abnormal possibility of all the pixel points in each connected domain in each microorganism initial class to obtain the credibility of each connected domain in each microorganism initial class.
Further, the method for obtaining the similarity comprises the following steps:
obtaining the similarity according to a similarity calculation formula, wherein the similarity calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>Initial class and/>, of species microorganismSimilarity between initial categories of species microorganisms; represents the/> Shape index of the initial class of species microorganism; /(I)Represents the/>Shape index of the initial class of species microorganism; /(I)Represents the/>A hue average value of an initial class of species microorganism; /(I)Represents the/>The average hue of the initial class of microorganisms.
Further, the method for acquiring the clustering distance comprises the following steps:
Taking the pixel point with the smallest abnormal possibility in each microorganism initial category as a characteristic pixel point in each microorganism initial category;
Calculating hue differences between characteristic pixel points of any two microorganism initial categories to be used as first hue differences;
And taking the ratio of the similarity between the first hue difference and any two initial categories of microorganisms as the clustering distance when the any two initial categories of microorganisms are clustered.
The invention has the following beneficial effects:
The method comprises the steps of obtaining an oral cavity microorganism flora image; the color information is the most intuitive characteristic in the oral cavity microbial flora image, so that the oral cavity microbial flora image is divided according to the color characteristic of the oral cavity microbial flora image to obtain an initial category of microorganisms, and an initial center point or a seed point can be provided for a clustering algorithm; the more the number of microorganisms with the same color characteristics in each type of microorganism flora, the more the color characteristics are possible to be the most obvious color characteristics in the microorganism flora, so that the possibility that any two microorganism initial categories belong to the same type of microorganisms is obtained according to the number change characteristics and the hue difference degree of pixel points between the any two microorganism initial categories; obtaining the abnormal possibility of each pixel point in each microorganism initial category according to the possibility distribution difference and the hue distribution difference of each pixel point and other pixel points in the same preset neighborhood range in each microorganism initial category, and reflecting the characteristic difference of each pixel point and other pixel points in the neighborhood range in the same preset neighborhood range; since the initial category of the microorganism may contain abnormal pixel points, the connected domain in the initial category of the microorganism may not have the referential property, and since the shape characteristics of the connected domain in different initial categories of the microorganism of the same category of the microorganism should be similar, the shape index of each initial category of the microorganism is obtained according to the abnormal possibility of the pixel points in each connected domain in each initial category of the microorganism and the shape characteristics of the connected domain; according to the shape index difference and the hue difference between any two initial categories of microorganisms, obtaining the similarity between any two initial categories of microorganisms, wherein the greater the similarity is, the more likely the any two initial categories of microorganisms belong to the same kind of microorganisms; obtaining a clustering distance when the two microorganism initial categories are clustered according to the similarity between the two microorganism initial categories; the oral cavity microbial flora is segmented according to the clustering distance. The method combines the position distribution and the color information of the microorganisms, determines the similarity between the initial categories of the microorganisms, adjusts the clustering distance according to the similarity, and improves the segmentation accuracy of the oral cavity microbial flora.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a method for dividing oral microbiota communities according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the present invention to achieve the preset purpose, the following detailed description refers to the specific implementation, structure, characteristics and effects of an oral microbiota community segmentation method according to the present invention with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
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 invention belongs.
The following specifically describes a specific scheme of the oral cavity microbial community dividing method provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a flowchart of a method for dividing oral microbiota communities according to an embodiment of the present invention is shown, the method includes:
Step S1: an image of the oral microbiota is acquired.
The embodiment of the invention provides a method for dividing oral microbial flora. Different kinds of segmentation are performed on the oral microbial flora, and an image of the oral microbial flora needs to be acquired first.
In one embodiment of the invention, an oral cavity microorganism sample is collected by using an oral cavity mouthwash method, the collected microorganism sample is dyed by using a gram, and different microorganisms show different colors due to different absorption and retention of dyes caused by different chemical compositions of cell walls of the different microorganisms, and a prepared sample glass is observed by using a microscope, and an oral cavity microorganism flora image is obtained by using the microscope.
Thus, an image of the oral microbiota is obtained.
Step S2: dividing the oral cavity microbial flora image according to the color characteristics of the oral cavity microbial flora image to obtain all initial categories of microorganisms; and obtaining the possibility that the two microorganism initial categories belong to the same microorganism according to the quantity change characteristics and the hue difference degree of the pixel points between the two microorganism initial categories.
In order to cluster different types of microorganisms in the oral cavity microorganism flora image subsequently, an initial center point or seed point is needed to be provided for a clustering algorithm, so that the accuracy and stability of clustering are improved. Since the color information is the most intuitive feature in the oral cavity microbial flora image, if the grayscale oral cavity microbial flora image is analyzed, a large amount of color information is lost, so that the clustering is inaccurate, if the RGB image is analyzed, the clustering algorithm efficiency is seriously reduced because each pixel point is singly classified under the condition that the RGB value range is large, and therefore, in the embodiment of the invention, the oral cavity microbial flora image is divided according to the color of the oral cavity microbial flora image relative to the oral cavity microbial flora image, so that the initial microbial category is obtained.
Preferably, in one embodiment of the present invention, the method for obtaining the initial category of microorganisms includes:
the oral cavity microbial flora image is mapped into an HSL color space to obtain a microbial hue space, and color information of the oral cavity microbial flora image is extracted in the microbial hue space, so that the dimension of data in the oral cavity microbial flora image can be reduced;
Because the value range of the hue in the HSL color space is 0-360 degrees, all possible colors in the oral cavity microbial flora image are covered, the microbial hue space is uniformly divided into 360 areas according to the hue, and all pixel points corresponding to the oral cavity microbial flora image in each area are used as an initial microbial category, so that all color information of the oral cavity microbial flora image can be presented.
To this end, all initial categories of microorganisms in the oral microbiota image are obtained.
Since in practice the hue of the same kind of microorganism will float over a range of hues and the more microorganisms in each kind of microorganism flora that have the same color characteristics, the more likely that color characteristic will be the most pronounced color characteristic in that microorganism flora. In order to determine the possibility that any two initial categories of microorganisms belong to the same type of microorganism, the number of pixels in all initial categories of microorganisms between any two initial categories of microorganisms needs to be counted, if any two initial categories of microorganisms do not belong to the same type of microorganism, the number of pixels between the two initial categories of microorganisms may be maximized, so that in the embodiment of the invention, the possibility that any two initial categories of microorganisms belong to the same type of microorganism is obtained according to the number change characteristics and the hue difference degree of the pixels between the any two initial categories of microorganisms.
Preferably, in one embodiment of the present invention, the method for acquiring the possibility that any two kinds of microorganisms belong to the same kind of microorganism includes:
Taking the hue average value of the pixel points in the initial microorganism category as an abscissa and the number of the pixel points in the initial microorganism category as an ordinate to obtain hue curves composed of all the initial microorganism categories, wherein the hue curves can intuitively reflect the number distribution condition of the pixel points with different color characteristics; in the hue curve, if two kinds of microorganisms belong to the same microorganism, as one kind of microorganism has the most obvious color characteristic, the most obvious color characteristic corresponds to the most pixels, and the hue wave band in the hue curve corresponding to one kind of microorganism has the characteristic of a peak, namely the most obvious color characteristic corresponds to the most pixels in the hue wave band of one kind of microorganism as the maximum value in the hue wave band, and has certain monotonicity at two sides of the peak, so that each kind of microorganism in the hue curve has the maximum value; in practical situations, different microorganism initial categories are often arranged on two sides of a peak of a hue curve composed of microorganisms of one category, so that the number monotonically decreasing frequency and the number monotonically increasing frequency of pixel points between any two microorganism initial categories are calculated in the hue curve; acquiring the initial possibility of microorganisms belonging to the same category between any two types of microorganisms according to the monotonically decreasing number frequency and the monotonically increasing number frequency; since the initial categories of microorganisms are respectively located at two sides of the peak of the hue curve composed of one type of microorganisms, the number difference degree of pixel points in every two adjacent initial categories of microorganisms is greatly different, and further the two initial categories of microorganisms respectively located at two sides of the peak of the hue curve composed of one type of microorganisms are considered to be different from the same type of microorganisms, so that the initial possibility is required to be corrected; since the initial types of different microorganisms in the same type of microorganisms have smaller hue difference, a correction coefficient of the initial possibility is obtained according to the hue difference degree between any two initial types of microorganisms; the product of the correction coefficient and the initial probability is taken as the probability that any two initial categories of microorganisms belong to the same category of microorganisms. In one embodiment of the present invention, the calculation formula of the probability that any two microorganism initial categories belong to the same category of microorganism is as follows:
；
in the method, in the process of the invention, Represents the/>Initial class and/>, of species microorganismThe likelihood that the initial class of species microorganisms belongs to the same class of microorganisms; /(I)Represents the/>Initial class and/>, of species microorganismA correction factor for the initial likelihood that the initial category of species of microorganism belongs to the same category of microorganism; /(I)Represents the/>Initial class and/>, of species microorganismThe initial class of microorganisms belongs to the initial possibilities of the same class of microorganisms.
In the probability calculation formula, the firstPixel points and/>, within the initial category of species microorganismThe smaller the difference in pixel color phase in the initial category of the species microorganism, the/>Initial class and/>, of species microorganismThe greater the likelihood that the initial species of microorganism belongs to the same species of microorganism, if at this point the/>Pixel points and/>, within the initial category of species microorganismIf the initial probability that the microorganism belonging to the same species is of a smaller initial class is small, the initial probability needs to be corrected, and the initial probability is increased, namely the/>Initial class and/>, of species microorganismThe greater the correction factor for the initial probability that the initial category of species of microorganism belongs to the same category of microorganism; since the/>Pixel points and/>, within the initial category of species microorganismThe difference of the hue between the pixel points in the initial category of the microorganism is larger, the initial probability is required to be further corrected, the initial probability is reduced, and the correction coefficient is smaller.
Preferably, in one embodiment of the present invention, the method for acquiring the initial likelihood includes:
The initial likelihood is obtained according to an initial likelihood calculation formula, which is shown as follows:
；
in the method, in the process of the invention, Represents the/>Initial class and/>, of species microorganismThe initial likelihood that an initial class of species of microorganism belongs to the same class of microorganism; /(I)Represents the/>, in the hue curveInitial class and/>, of species microorganismAn initial microorganism initial category between species microorganism initial categories; /(I)Represents the/>, in the hue curveInitial class and/>, of species microorganismTerminal microorganism initial categories between species microorganism initial categories; /(I)Expressed in hue curve as-Initial class and/>, of species microorganismA serial number of all initial categories of microorganisms between initial categories of species microorganisms; /(I)Represents the/>Color phase average value of pixel points in the initial category of the seed microorganisms; /(I)Represents the/>Color phase average value of pixel points in the initial category of the seed microorganisms; /(I)Represents the/>Initial class and/>, of species microorganismThe number of pixel points between the initial categories of the species microorganisms monotonically decreases; represents the/> Initial class and/>, of species microorganismThe number of pixel points between the initial categories of the species microorganisms monotonically increases in frequency; /(I)As an indication function; /(I)Representing a quantitative function; /(I)Representing the maximum function.
In the initial likelihood calculation formula,Represents the/>Initial class and/>, of species microorganismThe number of times of monotonically decreasing change of the pixel point number of every two adjacent microorganism initial categories among the microorganism initial categories reflects the/>Initial class of species microorganism to/>The number of the initial category pixel points of the species microorganism changes, the monotone decreasing times are more, and at the moment, the/>Initial class and/>, of species microorganismThe greater the monotonically decreasing frequency of the number of pixels per two adjacent microorganism initial categories between the microorganism initial categories, at this time the/>Initial class and/>, of species microorganismThe more likely the initial species of microorganism is in the hue curve of one species of microorganism and appears to the right of the peak of the hue curve of one species of microorganism, i.e. >, the/>Initial class of species microorganism to/>The higher the likelihood that the initial class of species of microorganism belongs to the same microorganism; the smaller the monotonically decreasing frequency is, the/>Initial class and/>, of species microorganismThe higher the monotonically increasing frequency of the number of the pixel points among the initial categories of the seed microorganisms, the description of the/>Initial class and/>, of species microorganismThe more likely the initial species of microorganism is to be to the left of the peak of the hue curve of one species of microorganism composition; maximum/>Reflecting the/>Initial class and/>, of species microorganismMonotonicity of pixel point number change of every two adjacent microorganism initial categories among species microorganism initial categories, namely reflecting the/>Initial class and/>, of species microorganismThe initial class of microorganisms belongs to the initial possibilities of the same microorganism.
Preferably, in one embodiment of the present invention, the method for obtaining the correction coefficient includes:
In order to avoid that two different initial categories of microorganisms are respectively present in hue curves of the two categories of microorganisms and the hue difference of the two different initial categories of microorganisms is smaller, a hue maximum value with the smallest hue average value difference with pixel points in the initial categories of microorganisms is obtained as a hue reference value of each initial category of microorganisms in the hue curves; taking the inverse of the hue reference value difference between any two microorganism initial categories as a hue reference coefficient; taking the reciprocal of the average difference of the hue between any two initial categories of microorganisms as a hue difference coefficient; the sum of the hue reference coefficient and the hue difference coefficient is taken as a correction coefficient of the initial possibility. In one embodiment of the present invention, the correction coefficient calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>Initial class and/>, of species microorganismA correction factor for the initial likelihood that the initial category of species of microorganism belongs to the same category of microorganism; /(I)Represents the/>Color phase average value of pixel points in initial category of species microorganism; /(I)Represents the/>Color phase average value of pixel points in initial category of species microorganism; /(I)Represents the/>Hue reference value of the initial class of species microorganism; represents the/> Hue reference values of pixels in an initial class of microorganisms.
In the correction coefficient calculation formula, the firstColor phase average value and/>, of pixel points in initial category of species microorganismThe smaller the difference between the hue means of the pixel points in the initial class of species microorganism, the/>Initial class and/>, of species microorganismThe greater the likelihood that the initial class of species of microorganism belongs to the same class of microorganism; since the/>Color phase average value and/>, of pixel points in initial category of species microorganismSince the difference between the color phase average values of the pixel points in the microorganism-derived initial category is small but the two microorganism-derived initial categories belong to different microorganism species, the difference between the color phase average values of the pixel points in the microorganism-derived initial category and the color phase average value of the pixel points in the microorganism-derived initial category is minimized is compared with the color phase reference value of the microorganism-derived initial category in the color phase curve, and the difference is the smallestHue maximum value with minimum hue average value difference of pixel points in initial category of species microorganism and/>The smaller the difference between hue maxima with the smallest hue mean difference of pixel points in the initial category of species microorganism, namely the first/>Initial class and/>, of species microorganismThe smaller the difference in hue reference values between the initial categories of the species microorganisms, the greater the degree of correction for the initial possibilities at this time, the greater the correction coefficient.
Step S3: and obtaining the abnormal possibility of each pixel point in each microorganism initial category according to the difference of the possibility distribution and the hue distribution of each pixel point and other pixel points in the preset adjacent area in each microorganism initial category.
In practical situations, the situation that pixels which do not belong to the same type of microorganism are divided into the same type of microorganism may occur, so that pixels which are abnormal in the initial category of the microorganism may occur, the pixels which are abnormal in color phase difference with other pixels in the initial category of the microorganism are larger, the possibility that the pixels which are abnormal in color phase difference with each other pixel in the initial category of the microorganism belong to the same type of microorganism is lower, and the pixels which are abnormal in color phase difference with the pixels in the neighborhood of the pixels which are abnormal in color phase difference with the pixels in the same neighborhood of the other pixels may be larger. If there are abnormal pixel points in the initial category of the microorganism, the calculation of the clustering distance between the initial categories of the microorganism is affected in the subsequent clustering process, and in order to reduce the influence of the abnormal pixel points on the initial category of the microorganism, in one embodiment of the invention, the abnormal possibility of each pixel point in each initial category of the microorganism is obtained according to the hue distribution difference and the possibility distribution difference of each pixel point and other pixel points in each initial category of the microorganism.
Preferably, in one embodiment of the present invention, the method for acquiring the likelihood of an abnormality includes:
obtaining the possibility of abnormality according to an abnormality possibility calculation formula, wherein the possibility of abnormality calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>A serial number of the target pixel point in the initial category of the seed microorganism; /(I)Represents the/>Abnormal probability of target pixel points in the initial category of the seed microorganism; /(I)Represents the/>The number of other pixels in the initial category of species microorganism; represents the/> Serial numbers of other pixels in the initial category of the seed microorganism; /(I)Represents the/>The number of pixels in each pixel neighborhood window in the initial category of the seed microorganism; /(I)Represents the/>Sequence numbers of pixel points in each pixel point neighborhood window in the initial category of the seed microorganism; /(I)Expressed as/>First/>, in a neighborhood window established by taking target pixel points in initial category of species microorganisms as centersA plurality of pixel points; /(I)Expressed as/>First/>, of the species microorganism initial categoryFirst/>, in a neighborhood window built by taking other pixel points as centersA plurality of pixel points; /(I)Representation/>Hue values of (2); /(I)Representation/>Hue values of (2); /(I)Representation/>And/>The possibility of belonging to the same species of microorganism.
In the calculation formula of the anomaly possibility, the greater the hue difference between each pixel point in the neighborhood window of the target pixel point and the corresponding pixel point in the neighborhood window of each other pixel point is, the greater the hue difference between the target pixel point and the first pixel point isThe hue of other pixels in the same neighborhood range has larger change, namely the target pixel and the/>The difference among other pixels is larger, which indicates that the integral characteristic in the neighborhood range of the target pixel is in the/>The more prominent the species microorganism is in the initial category, the more abnormal the target pixel is expressed, and the more likely the target pixel is an abnormal pixel; /(I)The smaller the target pixel point and the/>, the descriptionThe more unlikely that the other pixels belong to the same kind of microorganism, the more abnormal the target pixel is.
In one embodiment of the present invention, the preset neighborhood is set to be centered on the target pixel pointIs a square window of (c). It should be noted that, the preset neighborhood setting method may be set by an implementation personnel, and is not limited herein.
To this end, the anomaly possibility of all pixel points in each microorganism initial class is obtained.
Step S4: obtaining the shape index of each microorganism initial category according to the abnormal possibility of the pixel points in each communicating domain and the shape characteristics of the communicating domains in each microorganism initial category; obtaining the similarity between any two initial categories of microorganisms according to the shape index difference and the hue difference between any two initial categories of microorganisms; obtaining characteristic pixel points in each microorganism initial category; obtaining a clustering distance when the two microorganism initial categories are clustered according to the similarity of the two microorganism initial categories and the hue difference between the characteristic pixel points; the oral cavity microbial flora is segmented according to the clustering distance.
In clustering different initial categories of microorganisms, it is necessary to compare the similarity between the two initial categories of microorganisms to determine whether to cluster them. The shapes of areas formed by different types of microorganisms are different, for example, gram-positive microorganisms have strong adhesion capability, so that biofilm can be formed on the surfaces of teeth and oral mucosa, and colonisation and propagation in the oral cavity can be promoted. Thus, gram-positive microorganisms show a certain aggregation in the image; in contrast, gram-negative microorganisms have lower aggregation in the oral cavity and thus are more widely distributed, resulting in a structure in which the gram-negative microorganisms overall exhibit a tree-like distribution; the shape characteristics of the connected domains within the initial class of different microorganisms of the same species of microorganism should be similar. In order to compare the similarity of different initial categories of microorganisms, the shape characteristics of all connected domains in the initial categories of microorganisms need to be obtained; in addition, since the initial microorganism category may contain abnormal pixels, the connected domain in the initial microorganism category cannot represent the hue feature of the initial microorganism category, and the connected domain may not have the referential property, the shape feature of all the connected domains in the initial microorganism category needs to be considered for the abnormal pixels, and the shape index of the initial microorganism category is obtained. Therefore, in one embodiment of the present invention, the shape index of each microorganism initial class is obtained according to the abnormal probability of the pixel point in each connected domain and the shape characteristic of the connected domain in each microorganism initial class.
Preferably, in one embodiment of the present invention, the method for obtaining the confidence level includes:
calculating the inverse of the sum value of the abnormal possibility of all the pixel points in each connected domain in each microorganism initial class, and carrying out normalization processing to obtain the credibility of each connected domain in each microorganism initial class, wherein the larger the sum value of the abnormal possibility of all the pixel points in the connected domain is, the larger the abnormality degree of the connected domain is, the connected domain cannot represent the hue characteristics of the microorganism initial class, and the smaller the credibility of the connected domain is.
Preferably, in one embodiment of the present invention, the method for obtaining a shape index includes:
Calculating the credibility of each connected domain in each microorganism initial category; calculating the maximum distance and the minimum distance between pixel points in each connected domain in each microorganism initial category; taking the ratio of the maximum distance to the minimum distance as the distance coefficient of each connected domain in each microorganism initial category; taking the ratio of the number of boundary pixels of each connected domain in each microorganism initial class to the total number of the pixels of the connected domains as the shape coefficient of each connected domain in each microorganism initial class; taking the ratio between the distance coefficient and the shape coefficient as the shape characteristic of each connected domain in each microorganism initial category; and accumulating and summing the products of the credibility of all the connected domains in each microorganism initial category and the shape characteristics to obtain the shape index of each microorganism initial category. In one embodiment of the present invention, the shape index calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>Shape index of the initial class of species microorganism; /(I)Represents the/>The number of communicating domains within the initial class of species microorganism; /(I)Represents the/>Serial number of connected domain in initial class of microorganism; /(I)Represents the/>Within the initial category of species microorganism/>The number of pixels in each connected domain; /(I)Represents the/>Within the initial category of species microorganism/>Sequence numbers of pixel points in the connected domain; /(I)Represents the/>Abnormal probability of each pixel point in the initial category of the species microorganism; /(I)Represents the/>Within the initial category of species microorganism/>A plurality of connected domains; /(I)Represents the/>Within the initial category of species microorganism/>The sum of the abnormal possibility of all the pixel points in the connected domain; /(I)Represents the/>Within the initial category of species microorganism/>Boundary pixel points of the connected domains; /(I)Represents the/>Within the initial category of species microorganism/>A distance maximum value of the connected domains; represents the/> Within the initial category of species microorganism/>Minimum distance of each connected domain; /(I)Represents the/>Within the initial category of species microorganism/>The number of boundary pixel points of the connected domains; /(I)Represents the/>Within the initial category of species microorganism/>The number of pixels in each connected domain; /(I)Represents pixel point/>Within the initial category of species microorganism/>The/>, of the connected domainPixel dot and/>Maximum distance between individual pixel points; /(I)Represents the/>Within the initial category of species microorganism/>The/>, of the connected domainPixel dot and/>Minimum distance between individual pixel points; /(I)Representing a distance function; /(I)Representing a quantitative function; /(I)Representing a maximum function; /(I)Representing a minimum function.
In the shape index calculation formula, the firstWithin the initial category of species microorganism/>The ratio of the maximum value to the minimum value of the distance between the pixel points in the connected domains can reflect the/>Shape information of connected Domain,/>The closer to 1, the description of the/>The closer the shape of the individual communicating domains is to a circle, when/>Above 1, the first/>The shape of the individual connected domains may be elongated, and the/>Within the initial category of species microorganism/>The ratio of the number of boundary pixels in each connected domain to the total number of pixels can reflect the/>Boundary complexity of individual connected domains, wherein/>The closer to 1, the more/>The more tortuous the boundary line of the connected domains is, and/>The closer to 0, the description of the/>The smoother the boundary of each connected domain, so use/>Reflect the/>Within the initial category of species microorganism/>Shape characteristics of the individual connected domains; /(I)Represents the/>Within the initial category of species microorganism/>Abnormal possibility of connected Domain, no./>The smaller the probability of abnormality of the individual connected domain, the description of the/>The smaller the number of abnormal pixels in the connected domain, the/>The more the individual connected domains can reflect the/>Hue characteristics of the initial species of microorganism at this point/>The larger the (th)/>Degree of confidence of individual connected domainThe larger; will/>The product of the credibility of all connected domains in the initial category of the seed microorganism and the shape characteristic is accumulated and summed to obtain the/>Shape index of an initial class of microorganisms.
The shape index and the hue mean value between the initial categories of different microorganisms in the same kind of microorganisms should be similar, so in one embodiment of the invention, the similarity between the initial categories of any two microorganisms is obtained according to the shape index difference and the hue difference between the initial categories of any two microorganisms.
Preferably, in one embodiment of the present invention, the method for obtaining the similarity includes:
obtaining similarity according to a similarity calculation formula, wherein the similarity calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>Initial class and/>, of species microorganismSimilarity between initial categories of species microorganisms; represents the/> Shape index of the initial class of species microorganism; /(I)Represents the/>Shape index of the initial class of species microorganism; /(I)Represents the/>A hue average value of an initial class of species microorganism; /(I)Represents the/>The average hue of the initial class of microorganisms.
In the similarity calculation formula, the smaller the shape index difference is, the firstInitial class and/>, of species microorganismThe more likely the initial class of species microorganisms belong to the same class of microorganisms; the smaller the hue difference, the/>Initial class and/>, of species microorganismThe more likely the initial species of microorganism belongs to the same species of microorganism, the/>Initial class and/>, of species microorganismThe greater the similarity between the initial categories of species of microorganism.
Since the smaller the clustering distance between the two initial categories of microorganisms is, the greater the similarity between the initial categories of microorganisms is, in the embodiment of the invention, the clustering distance when the initial categories of any two microorganisms are clustered is obtained according to the similarity between the initial categories of any two microorganisms.
Preferably, in one embodiment of the present invention, the method for obtaining a cluster distance includes:
Taking the pixel point with the smallest abnormal possibility in each microorganism initial category as a characteristic pixel point in each microorganism initial category; calculating hue differences between characteristic pixel points of any two microorganism initial categories to be used as first hue differences; and taking the ratio of the similarity between the first hue difference and the initial categories of any two microorganisms as the clustering distance when the initial categories of any two microorganisms are clustered. In one embodiment of the present invention, the clustering distance calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>Initial class and/>, of species microorganismClustering distances between initial categories of species microorganisms; represents the/> Hue values of the characteristic pixel points in the initial category of the seed microorganisms; /(I)Represents the/>Hue values of the characteristic pixel points in the initial category of the seed microorganisms; /(I)Represents the/>Initial class and/>, of species microorganismSimilarity between initial categories of species of microorganism.
In the clustering distance calculation formula, the firstCharacteristic pixel points and/>, in initial category of species microorganismThe smaller the hue difference between the characteristic pixel points in the initial category of the species microorganism is, the description of the/>Initial class and/>, of species microorganismThe closer the hue characteristics of the initial species of microorganism are, the greater the/>Initial class and/>, of species microorganismThe more likely the initial species of microorganism belongs to the same species of microorganism, the/>Initial class and/>, of species microorganismThe more should the initial species of microorganism be clustered, at this point/>Initial class and/>, of species microorganismThe smaller the clustering distance between the initial categories of species microorganisms; and/>Initial class and/>, of species microorganismThe greater the similarity between the initial categories of species microorganisms, the greater the similarity between the initial categories of species microorganismsInitial class and/>, of species microorganismThe more should the initial species of microorganism be clustered, at this point/>Initial class and/>, of species microorganismThe smaller the clustering distance between the initial categories of species of microorganism.
Thus, the clustering distance between any two initial categories of microorganisms is obtained.
In one embodiment of the invention, hierarchical clustering algorithm is adopted for the oral cavity microbial flora images, and clustering segmentation is carried out on all the oral cavity microbial flora by utilizing the clustering distance between any two initial microbial categories.
It should be noted that the hierarchical clustering algorithm is a technical means well known to those skilled in the art, and will not be described herein. In other embodiments of the present invention, other clustering algorithms such as a spectral clustering algorithm and a kernel clustering algorithm may be used to perform clustering segmentation on the oral microbial flora, which is not limited herein.
Thus, the division of the oral microbiota community is completed.
In summary, the invention acquires images of oral microbiota; dividing the oral cavity microbial flora image according to the color characteristics of the oral cavity microbial flora image to obtain an initial category of microorganisms; obtaining the possibility that any two microorganism initial categories belong to the same microorganism according to the quantity change characteristics and the hue difference degree of the pixel points between the any two microorganism initial categories; obtaining the abnormal possibility of each pixel point in each microorganism initial category according to the possibility distribution difference and the hue distribution difference of each pixel point and other pixel points in the preset adjacent areas in each microorganism initial category; obtaining the shape index of each microorganism initial category according to the abnormal possibility of the pixel points in each communicating domain and the shape characteristics of the communicating domains in each microorganism initial category; obtaining the similarity between any two initial categories of microorganisms according to the shape index difference and the hue difference between any two initial categories of microorganisms; obtaining a clustering distance when the two microorganism initial categories are clustered according to the similarity between the two microorganism initial categories; the oral cavity microbial flora is segmented according to the clustering distance. The method combines the position distribution and the color information of the microorganisms, determines the similarity between the clustering categories, adjusts the clustering distance according to the similarity, and improves the segmentation accuracy of the oral cavity microorganism flora.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments.
Claims (10)
1. A method for dividing an oral microbiota community, the method comprising:
acquiring an oral microbial flora image;
Dividing the oral cavity microbial flora image according to the color characteristics of the oral cavity microbial flora image to obtain all initial categories of microorganisms; obtaining the possibility that any two microorganism initial categories belong to the same microorganism according to the quantity change characteristics and the hue difference degree of the pixel points between the any two microorganism initial categories;
Obtaining the abnormal possibility of each pixel point in each microorganism initial category according to the possibility distribution difference and the hue distribution difference of each pixel point and other pixel points in each microorganism initial category in a preset adjacent area;
Obtaining the shape index of each microorganism initial category according to the abnormal possibility of the pixel points in each communicating domain and the shape characteristics of the communicating domains in each microorganism initial category; obtaining the similarity between any two initial categories of microorganisms according to the shape index difference and the hue difference between any two initial categories of microorganisms; obtaining a clustering distance when the two microorganism initial categories are clustered according to the similarity of the two microorganism initial categories; and dividing the oral cavity microbial flora according to the clustering distance.
2. The method for partitioning an oral microbiota community according to claim 1, wherein the method for obtaining the initial class of microorganisms comprises:
mapping the oral cavity microorganism flora image into an HSL color space to obtain a microorganism hue space;
and uniformly dividing the microbial hue space into 360 areas according to hue, and taking all pixel points corresponding to the oral cavity microbial flora image in each area as an initial microbial category.
3. The method for partitioning an oral microbiota community according to claim 1, wherein the method for obtaining the likelihood that any two initial categories of microorganisms belong to the same category of microorganisms comprises:
Taking the hue average value of the pixel points in the initial microorganism category as an abscissa and the number of the pixel points in the initial microorganism category as an ordinate to obtain hue curves formed by all the initial microorganism categories;
in the hue curve, calculating the monotonically decreasing frequency and the monotonically increasing frequency of the number of the pixel points between any two microorganism initial categories;
acquiring initial possibility of microorganisms belonging to the same category between any two initial categories of microorganisms according to the monotonically decreasing number frequency and the monotonically increasing number frequency;
Obtaining a correction coefficient of the initial possibility according to the hue difference degree between any two microorganism initial categories;
and taking the product of the correction coefficient and the initial probability as the probability that any two microorganism initial categories belong to the same microorganism category.
4. A method of oral microbiota population segmentation according to claim 3, characterized in that the initial likelihood acquisition method comprises:
the initial likelihood is obtained according to an initial likelihood calculation formula, wherein the initial likelihood calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>Initial class and/>, of species microorganismThe initial likelihood that an initial class of species of microorganism belongs to the same class of microorganism; /(I)Represents the/>, in the hue curveInitial class and/>, of species microorganismAn initial microorganism initial category between species microorganism initial categories; /(I)Represents the/>, in the hue curveInitial class and/>, of species microorganismTerminal microorganism initial categories between species microorganism initial categories; /(I)Expressed in hue curve as-Initial class and/>, of species microorganismA serial number of all initial categories of microorganisms between initial categories of species microorganisms; /(I)Represents the/>Color phase average value of pixel points in the initial category of the seed microorganisms; /(I)Represents the/>Color phase average value of pixel points in the initial category of the seed microorganisms; /(I)As an indication function; /(I)Representing a quantitative function; /(I)Representing the maximum function.
5. A method for partitioning an oral microbiota population according to claim 3, characterized in that the method for obtaining the correction factor comprises:
In the hue curve, obtaining a hue maximum value with the smallest hue average value difference from pixel points in the initial category of microorganisms as a hue reference value of each initial category of microorganisms; taking the inverse of the hue reference value difference between any two microorganism initial categories as a hue reference coefficient; taking the reciprocal of the average difference of the hue between any two initial categories of microorganisms as a hue difference coefficient; and taking the sum of the hue reference coefficient and the hue difference coefficient as a correction coefficient of the initial possibility.
6. The method for dividing oral microbiota community according to claim 1, wherein the method for obtaining the abnormal likelihood comprises:
Obtaining the possibility of abnormality according to an abnormality possibility calculation formula, wherein the possibility of abnormality calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>A serial number of the target pixel point in the initial category of the seed microorganism; /(I)Represents the/>Abnormal probability of target pixel points in the initial category of the seed microorganism; /(I)Represents the/>The number of other pixels in the initial category of species microorganism; /(I)Represents the/>Serial numbers of other pixels in the initial category of the seed microorganism; /(I)Represents the/>The number of pixels in each pixel neighborhood window in the initial category of the seed microorganism; /(I)Represents the/>Sequence numbers of pixel points in each pixel point neighborhood window in the initial category of the seed microorganism; /(I)Expressed as/>First/>, in a neighborhood window established by taking target pixel points in initial category of species microorganisms as centersA plurality of pixel points; /(I)Expressed as/>First/>, of the species microorganism initial categoryFirst/>, in a neighborhood window built by taking other pixel points as centersA plurality of pixel points; /(I)Representation/>Hue values of (2); /(I)Representation/>Hue values of (2); /(I)Representation/>And/>The possibility of belonging to the same species of microorganism.
7. The method for partitioning an oral microbiota community according to claim 1, wherein the method for obtaining the shape index comprises:
Calculating the credibility of each connected domain in each microorganism initial category;
calculating the maximum distance and the minimum distance between pixel points in each connected domain in each microorganism initial category;
taking the ratio of the maximum distance to the minimum distance as the distance coefficient of each connected domain in each microorganism initial category;
Taking the ratio of the number of boundary pixels of each connected domain in each microorganism initial class to the total number of the pixels of the connected domains as the shape coefficient of each connected domain in each microorganism initial class;
Taking the ratio between the distance coefficient and the shape coefficient as the shape characteristic of each connected domain in each microorganism initial category;
And accumulating and summing the products of the credibility of all the connected domains in each microorganism initial category and the shape characteristics to obtain the shape index of each microorganism initial category.
8. The method for partitioning an oral microbiota community according to claim 7, wherein the method for obtaining the confidence level comprises:
and normalizing the inverse of the sum value of the abnormal possibility of all the pixel points in each connected domain in each microorganism initial class to obtain the credibility of each connected domain in each microorganism initial class.
9. The method for partitioning an oral microbiota community according to claim 1, wherein the method for obtaining similarity comprises:
obtaining the similarity according to a similarity calculation formula, wherein the similarity calculation formula is as follows:
；
in the method, in the process of the invention, Represents the/>Initial class and/>, of species microorganismSimilarity between initial categories of species microorganisms; /(I)Represents the/>Shape index of the initial class of species microorganism; /(I)Represents the/>Shape index of the initial class of species microorganism; /(I)Represents the/>A hue average value of an initial class of species microorganism; /(I)Represents the/>The average hue of the initial class of microorganisms.
10. The method for partitioning oral microbiota community according to claim 1, wherein the method for obtaining the clustering distance comprises:
Taking the pixel point with the smallest abnormal possibility in each microorganism initial category as a characteristic pixel point in each microorganism initial category;
Calculating hue differences between characteristic pixel points of any two microorganism initial categories to be used as first hue differences;
And taking the ratio of the similarity between the first hue difference and any two initial categories of microorganisms as the clustering distance when the any two initial categories of microorganisms are clustered.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410355036.5A CN117953495B (en) | 2024-03-27 | 2024-03-27 | Oral cavity microbial flora segmentation method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410355036.5A CN117953495B (en) | 2024-03-27 | 2024-03-27 | Oral cavity microbial flora segmentation method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117953495A true CN117953495A (en) | 2024-04-30 |
| CN117953495B CN117953495B (en) | 2024-06-04 |
Family
ID=90794724
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410355036.5A Active CN117953495B (en) | 2024-03-27 | 2024-03-27 | Oral cavity microbial flora segmentation method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117953495B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2728353A1 (en) * | 2012-10-31 | 2014-05-07 | Polyor SARL | Method for assigning an agro/micro-pedoclimate (AMPC) to an agricultural plot, and for forming microbial consortia |
| CN106769893A (en) * | 2016-11-22 | 2017-05-31 | 西安建筑科技大学 | A kind of characterizing method of the sludge microbe COMMUNITY CHARACTERISTICS based on color space |
| US20180105444A1 (en) * | 2015-04-03 | 2018-04-19 | Sumitomo Chemical Company, Limited | Microbial Flora Analysis System, Determination System, Microbial Flora Analysis Method, and Determination Method |
| CN109661473A (en) * | 2016-09-02 | 2019-04-19 | 生物梅里埃公司 | For determining the presence of microorganism and identifying the method, system and computer program product of the microorganism |
| CN112151191A (en) * | 2020-10-16 | 2020-12-29 | 山东管理学院 | Microorganism and disease association relation prediction method and system based on attention mechanism |
| CN112651305A (en) * | 2020-12-10 | 2021-04-13 | 哈尔滨师范大学 | Method for identifying microorganism species |
| CN113012810A (en) * | 2019-12-18 | 2021-06-22 | 中国科学院昆明动物研究所 | FBA oral flora functional group obtained based on microbial functional group mining method |
| RU2791813C1 (en) * | 2022-05-03 | 2023-03-13 | Курочкин Евгений Владимирович | System and method for detecting and classifying microorganism colonies using images based on artificial intelligence and computer vision technologies |
| CN116992314A (en) * | 2023-07-03 | 2023-11-03 | 武汉理工大学 | Microbial community clustering analysis method |
-
2024
- 2024-03-27 CN CN202410355036.5A patent/CN117953495B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2728353A1 (en) * | 2012-10-31 | 2014-05-07 | Polyor SARL | Method for assigning an agro/micro-pedoclimate (AMPC) to an agricultural plot, and for forming microbial consortia |
| US20180105444A1 (en) * | 2015-04-03 | 2018-04-19 | Sumitomo Chemical Company, Limited | Microbial Flora Analysis System, Determination System, Microbial Flora Analysis Method, and Determination Method |
| CN109661473A (en) * | 2016-09-02 | 2019-04-19 | 生物梅里埃公司 | For determining the presence of microorganism and identifying the method, system and computer program product of the microorganism |
| CN106769893A (en) * | 2016-11-22 | 2017-05-31 | 西安建筑科技大学 | A kind of characterizing method of the sludge microbe COMMUNITY CHARACTERISTICS based on color space |
| CN113012810A (en) * | 2019-12-18 | 2021-06-22 | 中国科学院昆明动物研究所 | FBA oral flora functional group obtained based on microbial functional group mining method |
| CN112151191A (en) * | 2020-10-16 | 2020-12-29 | 山东管理学院 | Microorganism and disease association relation prediction method and system based on attention mechanism |
| CN112651305A (en) * | 2020-12-10 | 2021-04-13 | 哈尔滨师范大学 | Method for identifying microorganism species |
| RU2791813C1 (en) * | 2022-05-03 | 2023-03-13 | Курочкин Евгений Владимирович | System and method for detecting and classifying microorganism colonies using images based on artificial intelligence and computer vision technologies |
| CN116992314A (en) * | 2023-07-03 | 2023-11-03 | 武汉理工大学 | Microbial community clustering analysis method |
Non-Patent Citations (1)
| Title |
|---|
| 刘文洁;王洋;周礼红;方月月;解文利;: "贵州地区肥胖人群肠道微生物群落特征分析", 山地农业生物学报, no. 01, 15 January 2020 (2020-01-15) * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117953495B (en) | 2024-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107578423B (en) | Multi-feature hierarchical fusion related filtering robust tracking method | |
| CN111863232B (en) | Remote disease intelligent diagnosis system based on block chain and medical image | |
| CN112767416B (en) | Fundus blood vessel segmentation method based on space and channel dual attention mechanism | |
| CN109145848A (en) | A kind of wheat head method of counting | |
| CN113298777A (en) | Cotton leaf blight detection method and system based on color features and super-pixel clustering | |
| CN114567528B (en) | Communication signal modulation mode open set recognition method and system based on deep learning | |
| CN117953495B (en) | Oral cavity microbial flora segmentation method | |
| CN109753969A (en) | A kind of insulator based on shape feature and image segmentation identifies and positions method | |
| CN113052859A (en) | Super-pixel segmentation method based on self-adaptive seed point density clustering | |
| CN117764864B (en) | Nuclear magnetic resonance tumor visual detection method based on image denoising | |
| CN114822823A (en) | Tumor fine classification system based on cloud computing and artificial intelligence fusion multi-dimensional medical data | |
| CN117314940B (en) | Laser cutting part contour rapid segmentation method based on artificial intelligence | |
| CN117575953A (en) | Detail enhancement method for high-resolution forestry remote sensing image | |
| CN117173049A (en) | Image enhancement method for ureteroscope lithotripsy and lithotomy | |
| CN114677384B (en) | Solar cell coating defect detection method | |
| Ikonomakis et al. | Unsupervised seed determination for a region-based color image segmentation scheme | |
| CN114366116B (en) | Parameter acquisition method based on Mask R-CNN network and electrocardiogram | |
| WO2023284528A1 (en) | Image enhancement method and apparatus, computer device, and storage medium | |
| CN113139930B (en) | Thyroid slice image classification method and device, computer equipment and storage medium | |
| CN115147445A (en) | Multi-threshold human brain x-ray image segmentation method based on mixed wolf algorithm | |
| CN111340761B (en) | Remote sensing image change detection method based on fractal attribute and decision fusion | |
| CN112837293A (en) | Hyperspectral image change detection method based on Gaussian function typical correlation analysis | |
| CN116824586B (en) | Image processing method and black garlic production quality online detection system applying same | |
| CN104504681A (en) | Threshold image segmentation method with minimal clustering distortion | |
| CN116309186B (en) | Infrared image dynamic range compression method based on multi-section S-curve mapping |
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 |