CN115032157A - Cloud platform-based identification method for detecting secretion by mobile phone photographing - Google Patents
Cloud platform-based identification method for detecting secretion by mobile phone photographing Download PDFInfo
- Publication number
- CN115032157A CN115032157A CN202210643512.4A CN202210643512A CN115032157A CN 115032157 A CN115032157 A CN 115032157A CN 202210643512 A CN202210643512 A CN 202210643512A CN 115032157 A CN115032157 A CN 115032157A
- Authority
- CN
- China
- Prior art keywords
- color
- test paper
- mobile phone
- cloud platform
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/8483—Investigating reagent band
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
Abstract
The invention relates to a cloud platform-based identification method for detecting secretions through mobile phone photographing. After the test paper is used for detection, a mobile phone is used for shooting, test paper images are automatically collected and uploaded to a cloud platform, the cloud platform integrates an image recognition method, and the results are returned to the mobile phone after being processed. Firstly, Harris corner detection calibration images are carried out; and then converting the image space from RGB to CIELAB, calculating the color difference between the test paper color and the color block of the standard colorimetric card, classifying the test paper color into the corresponding category of the colorimetric card by using a KNN nearest neighbor algorithm, and automatically completing the analysis of the test paper measurement index. The method is quick and simple, and avoids the problems of ambient light interference and low human eye identification accuracy.
Description
Technical Field
The invention belongs to the field of image recognition, and particularly relates to a recognition method for detecting secretions by photographing with a mobile phone based on a cloud platform, which can be used for family gynecological examination so as to diagnose and prevent gynecological diseases.
Background
In the medical field, diagnosis of diseases usually requires measurement of various parameters, most of which are from detection results of samples, and detection by using test strips is a widely applied detection method. In many gynecological examinations, vaginal secretion is a signal light of female health condition, vaginal secretion examination is also called leucorrhea examination, and leucorrhea detection and analysis is an important means for detecting many gynecological diseases, and has great significance in clinical diagnosis. The leucorrhea test object includes: vaginal pruritus, red swelling, abnormal secretion and fixed inspection after healing; before and after operation examination in gynecology, abortion or other uterine cavities; the routine physical examination in gynecology and obstetrics and the repeated examination of vaginosis.
At present, in order to change the current situation that the traditional manual detection depends on experience and the false judgment rate of missed detection is high, a plurality of electronic devices enter the medical field, and the leucorrhea analyzers appear in succession. The leucorrhea analyzer generally has a special reagent strip, and the reagent strip line contains a plurality of reagent pieces, and the equidistance distributes on the test paper. The test paper blocks are coated with chemical reagents specially used for detecting corresponding items, and the number of the reagent blocks determines the number of the detected items. The reagent block is coated with chemicals for measuring corresponding items, and color reaction occurs when the reagent block contacts with the white tape. According to the color change degree of the reagent block, the concentration of the corresponding substance in the white band can be judged and the detection result can be obtained.
The vaginal secretion is widely applied to the aspects of gynecological disease diagnosis, curative effect observation, prevention and the like. The traditional method is adopted for detecting the white bands, the process is complex, and the period is long. Although the white band analyzer can quickly and accurately measure various indexes of the white band, most of the white band analyzers are too expensive and have a limited detection range. The traditional human eye comparison and identification method is low in efficiency and is easily influenced by subjective factors of detection personnel, aiming at the problems, the western China gynecologic hospital provides an idea that a patient and a doctor can quickly and accurately judge test paper results, and the identification method for detecting female vaginal secretion by taking pictures through a mobile phone based on the patent cloud platform can be used for carrying out conventional detection on the vaginal secretion on mobile equipment and helping the user to carry out self-detection on the vaginal secretion at home. At present, the accuracy of the analysis result of the technology is reliably ensured by the clinical test of western China; by adopting the method of dry chemical analysis of the vaginal secretion and using the mobile equipment camera to scan the test paper and correct the color card to carry out the spectrum correction algorithm, a semi-quantitative measurement result can be obtained, and the method has the advantages of low cost, quickness in detection, convenience in use and the like.
Disclosure of Invention
The invention provides a cloud platform-based identification method for detecting secretions by photographing with a mobile phone, which is low in cost, simple and intuitive, shortens the detection time and has higher accuracy.
The invention provides a mobile phone photographing secretion detection identification method based on a cloud platform, which comprises the following steps:
and 4, classifying the colors of the test paper into classes corresponding to the color comparison card by using a KNN nearest neighbor algorithm, and automatically analyzing the index concentration.
The white band test paper comprises a two-dimensional code part, a standard color card part and a test paper part, wherein a link website of the test paper can be opened by scanning the two-dimensional code, and the test paper can be used for checking an instruction, a manufacturer, a standard color spectrum, an example and the like, and is used for photographing and identifying the color of each index after reaction.
In the step 3, the image space is converted, and the image is firstly converted from the RGB space to the CIEXYZ space and then to the CIELAB space.
And 4, classifying the sampled test paper colors into the color comparison card sample category in the CIELAB color space, firstly selecting the color comparison card sample color in the CIELAB color space, and classifying the test paper samples into the color category corresponding to the selected nearest color, thereby completing the analysis of the test paper index data.
The method also comprises the step of identifying the colors and the concentrations of the test paper blocks of different manufacturers, different models and different detection items.
The secretion inspection test paper image is obtained by photographing the mobile phone, and the result is returned to the mobile phone after the cloud platform finishes color recognition.
Drawings
FIG. 1 is a schematic diagram of the structure of the test strip for detecting leucorrhea in the example.
Fig. 2 is a flow chart of the image recognition method of the present invention.
Fig. 3 is a flow chart of Harris corner detection according to the present invention.
FIG. 4 is a flow chart of CIELAB spatial color difference calculation according to the present invention.
Detailed Description
The following describes in detail a method for recognizing secretions by mobile phone photographing based on a cloud platform according to the present invention with reference to the accompanying drawings.
Fig. 1 is a schematic structural diagram of the test paper for detecting white bands, which includes three parts: the test paper comprises a two-dimensional code part 1, a standard color card part 2 and a test paper part 3. The type, the use instruction, the manufacturer, the standard chromatogram, the example and the like of the test paper can be obtained by scanning the two-dimensional code part 1; the standard color card part 2 is used for judging the color similarity between the detection test paper 3 and the standard color card during detection; the test paper portion 3 is used for color reaction with the white band to be detected.
When the test paper is used, the white tape is firstly placed on the test paper, after the reaction is completed, the color of the test paper is firstly measured by human eyes, and the reference value of the test paper in a standard color card is determined.
Fig. 2 is a flow chart of a recognition method for detecting secretions by mobile phone photographing based on a cloud platform, which comprises the following steps:
and 4, classifying the colors of the test paper into classes corresponding to the color comparison card by using a KNN nearest neighbor algorithm, and automatically analyzing the index concentration.
The method for obtaining the images of the white band test paper and the standard color card comprises the steps of collecting images of the test paper and the standard color card after reaction by a camera of the mobile device, compressing an original image after the camera collects the original image, and simultaneously compressing new pixels which are the average value of the original surrounding average value to achieve the effect of average value filtering, wherein the camera is used for collecting images of the white band test paper and the standard color card after reaction, and the camera is used for collecting images of the white band test paper and the color card at the same time so as to avoid ambient light interference.
Fig. 3 is a flow chart of Harris corner detection, where the Harris corner detection calibrates an image, and is used to eliminate the influence of shake in the shooting process on the image acquisition result, and the basic mathematical formula of the Harris corner detection is as follows:
the Harris algorithm is carried out by utilizing autocorrelation of image gray in a window, a window is set and moved in an image, and autocorrelation coefficients of images of areas where the window is located before and after movement are calculated;
w (x, y) represents a window function, L represents a window, (u, v) refers to a point in the window, I (x, y) is the original image pixel, I (x + u, y + v) is the intensity after translation, and E (u, v) is desired.
The method comprises the following steps:
the method comprises the following steps: w (X, Y) represents a moving window from which the first order Gaussian partial derivative I is calculated in the X-and Y-axes of the image X And I Y ;I X Is the x-axis component of the image pixel; i is Y Is the y-axis component of the image pixel;
step two: obtaining I from the results of the first step X 2 ,I Y 2 And I X ·I Y A value;
step three: gaussian blur second step three valuesS X·Y (ii) a S represents the pixel value of the pixel point, namely the average brightness information in a certain block;
step four: from the Harris matrix of pixels, the matrix eigenvalue λ is calculated 1 ,λ 2 ;
m is a coefficient matrix relating only to x, y after the desired E (u, v) reduction;
step five: calculating the R value of each pixel;
the formula is as follows:
R=detM-k(traceM) 2
detM=λ 1 λ 2
traceM=λ 1 +λ 2
detM denotes the determinant of M, traceM denotes the trace of M, and R denotes the corner response value. k is an empirical constant, generally between 0.04 and 0.06.
Step six: non-maximum suppression is achieved using windows of 3 x 3 or 5 x 5;
step seven: according to the corner point detection result, finding color block matrix matching points in the extracted corner points, and performing self-adaptive sampling;
after the image is calibrated, the calibrated image needs to be uploaded to a server, after the server receives image data uploaded by a client, a color correction model is generated according to color correction color card color data and standard color card data in the image, brightness adjustment and color correction are carried out on the uploaded image, and accurate color restoration is carried out. In the aspect of Color Correction, an iccs (Image Color Correction system) is used for Color restoration, i.e. a specific process described in fig. 4.
Fig. 4 is a flowchart of a CIELAB spatial color difference calculation, which converts an image space from RGB to CIELAB, because the original image data collected by the camera is stored in an RGB format, and the CIELAB spatial color difference calculation requires that RGB color data be converted into a CIELAB color space, but the RGB color space cannot be directly converted into the CIELAB color space, and needs to be converted into the CIELAB color space after being converted into the CIELAB color space, that is: RGB-XYZ-LAB comprising the following steps:
the method comprises the following steps: calculating specific values of RGB pixel channels, and performing nonlinear tone editing on colors by using a gamma function to improve the contrast of the image;
the value ranges of the R, G and B pixel channels are [0, 255], and the R, G and B pixel values are calculated according to the following formula:
r, G and B represent specific pixel values of R, G and B pixel channels; r, G and B represent values of R, G and B pixel channels;
x is the color in RGB space, Gamma (x) is the Gamma correction formula;
step two: converting the RGB color space into a color space, wherein the color space is converted into a CIEXYZ color space according to the following formula:
m is a coefficient matrix in an initial conversion matrix formula between two spaces designed by the CIE committee;
wherein M is [0.4124, 0.3576, 0.1805, 0.2126, 0.7152, 0.0722, 0.0193, 0.1192, 0.9505]
Step three: converting the CIEXYZ color space into a CIELAB color space, wherein the formula is as follows:
L * the values represent the brightness of the color, ranging from 0 to 100;
a * the value represents the color change from green to red, -128 (pure green) to +128 (pure red), graded by 256;
b * the value represents the color change from yellow to blue, -128 (pure blue) to +128 (pure yellow), graded by 256
X, Y, Z-3 stimulus values for an object;
X n 、Y n 、Z n 3 stimulus values of the CIE standard illuminant (light source) of the reference white point;
the function f (t) is divided into two parts to avoid the infinite slope at t-0;
step four: and evaluating the color similarity of CIELAB space color difference by using Euclidean space distance, and correspondingly calculating the color difference between the white band image and a standard color card, wherein the formula is as follows:
L 1 * -a standard value of the target brightness;
a 1 * -a standard value of target chroma;
b 1 * -a standard value of target chroma;
L 2 * -a measure of brightness of the captured image;
a 2 * -a measure of the chroma of the captured image;
b 2 * -a measure of the chroma of the captured image;
ΔE * ab large total color differenceSmall;
the KNN nearest neighbor algorithm classifies the test paper colors into the classes corresponding to the color cards, the test paper colors sampled in the CIELAB color space are classified into the color card sample classes, the color card sample colors in the CIELAB color space are selected firstly, and the paper samples to be tested are classified into the color classes corresponding to the selected nearest neighbor colors according to the KNN nearest neighbor algorithm, so that the test paper index data analysis is completed.
Claims (6)
1. A mobile phone photographing secretion detection identification method based on a cloud platform is characterized by comprising the following steps:
step 1, obtaining a secretion inspection test paper image;
step 2, detecting a calibration image by Harris angular points;
step 3, transferring the image space from RGB to CIELAB, and calculating the color difference between the secretion inspection image and the standard color card;
and 4, classifying the test paper colors to the corresponding classes of the color comparison card by using a KNN nearest neighbor algorithm, and automatically analyzing the index concentration.
2. The identification method for detecting secretions by mobile phone photographing based on a cloud platform as claimed in claim 1, wherein: the secretion inspection test paper comprises a two-dimensional code part, a standard color card part and a detection test paper part, wherein the two-dimensional code is used for opening a link website of the test paper, checking information and taking pictures to identify the color of each index after reaction.
3. The identification method for detecting secretions by mobile phone photographing based on a cloud platform as claimed in claim 1, wherein: in the step 3, the image space is converted, and the image is firstly converted from the RGB space to the CIEXYZ space and then to the CIELAB space.
4. The identification method for detecting secretions by mobile phone photographing based on a cloud platform as claimed in claim 1, wherein: and 4, classifying the sampled test paper colors into the color comparison card sample category in the CIELAB color space, firstly selecting the color comparison card sample color in the CIELAB color space, and classifying the test paper samples into the color category corresponding to the selected nearest color, thereby completing the analysis of the test paper index data.
5. The identification method for detecting secretions by mobile phone photographing based on a cloud platform as claimed in claim 1, wherein: the method also comprises the step of identifying the colors and the concentrations of the test paper blocks of different manufacturers, different models and different detection items.
6. The identification method for detecting secretions by mobile phone photographing based on a cloud platform as claimed in claim 1, wherein: the secretion inspection test paper image is obtained by photographing the mobile phone, and the result is returned to the mobile phone after the cloud platform finishes color recognition.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210643512.4A CN115032157A (en) | 2022-06-08 | 2022-06-08 | Cloud platform-based identification method for detecting secretion by mobile phone photographing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210643512.4A CN115032157A (en) | 2022-06-08 | 2022-06-08 | Cloud platform-based identification method for detecting secretion by mobile phone photographing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN115032157A true CN115032157A (en) | 2022-09-09 |
Family
ID=83123432
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210643512.4A Pending CN115032157A (en) | 2022-06-08 | 2022-06-08 | Cloud platform-based identification method for detecting secretion by mobile phone photographing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115032157A (en) |
-
2022
- 2022-06-08 CN CN202210643512.4A patent/CN115032157A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101305906B (en) | System for detecting colorimetric abnormality in organism | |
CN106991679A (en) | One kind quantifies recognition methods based on cloud platform urine test paper physical signs | |
EP3087534B1 (en) | Analysis device, system and program | |
WO2018107983A1 (en) | Colour recognition method and system | |
JP4071186B2 (en) | Method and system for identifying an object of interest in a biological specimen | |
CN106546581A (en) | Detection paper card intelligent checking system and detection paper card intelligent analysis method | |
JP7062926B2 (en) | Color reaction detection system, color reaction detection method and program | |
CN110072470A (en) | Method and apparatus for estimating device for testing ovulation date | |
CN109030480B (en) | Sample analysis method, sample analysis device, readable storage medium and computer equipment | |
CN105466921A (en) | Simultaneous detection method of many samples | |
CN109242792B (en) | White balance correction method based on white object | |
CN110495888B (en) | Standard color card based on tongue and face images of traditional Chinese medicine and application thereof | |
CN113768452A (en) | Intelligent timing method and device for electronic endoscope | |
JP3522851B2 (en) | Detection method | |
CN106204523A (en) | A kind of image quality evaluation method and device | |
JPH09131323A (en) | Skin surface analyzer and skin surface evaluating method | |
CN115032157A (en) | Cloud platform-based identification method for detecting secretion by mobile phone photographing | |
CN106970074A (en) | The Urine test paper image-recognizing method of ability is corrected with color | |
Kibria et al. | Smartphone-based point-of-care urinalysis assessment | |
KR102399939B1 (en) | Detection method and detection pad | |
CN115100273A (en) | Immunochromatographic test strip quantitative analysis system and detection method based on image processing | |
KR100778370B1 (en) | System for analyzing skin color distribution and method therof | |
KR20200140500A (en) | Detection method and detection pad | |
CN115019059A (en) | Secretion test paper image identification method based on color correction mode | |
JP3265464B2 (en) | Sample evaluation method and apparatus using microscopic images |
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 |