A scar measurement method

Abstract

The invention discloses a scar measurement method, comprising the following steps: S1, a patient sticks a label sticker with positioning points on the skin, then uses a mobile phone to take a photo of the scar position and the label sticker, generates a picture, and sends the picture to the APP background through the mobile phone APP; S2, the APP background generates an overall three-dimensional model from the picture; S3, the normal skin and the keloid area of ​​the patient in the overall three-dimensional model are segmented to obtain the keloid three-dimensional model; at the same time, a normal skin model and a ruler model are separated from the overall three-dimensional model; S4, scar color difference data is obtained by comparing the normal skin model with the keloid three-dimensional model; the proportional relationship between the ruler model and the actual size is obtained by comparing the ruler model with the distance of the positioning points on the label sticker in the picture; and the actual measurement data of the scar is obtained through the proportional relationship.

Description

Technical Field

[0001] The invention relates to the field of skin beauty, and in particular to a scar measurement method. Background Art

[0002] Scars are caused by a variety of factors and are a common skin disease. They have a long course, change rapidly, and are difficult to evaluate in terms of treatment effectiveness. At present, hospitals do not have quantitative inspection and filing tools, and do not have effective comparison tools for evaluating the treatment cycle and recovery period of patients, making it difficult for doctors to assess the development of scars and formulate drug treatment plans. The following problems currently exist:

[0003] 1. Imprecise efficacy assessment

[0004] Keloids are three-dimensional structures with irregular shapes and uneven surfaces. The traditional measurement method using a ruler makes it difficult to accurately measure the area and thickness of keloids, resulting in rough evaluation standards and large errors.

[0005] 2. Difficulty in follow-up

[0006] Patients from other places cannot follow up on schedule, which not only affects the evaluation of the long-term effects of keloid treatment, but also affects the early detection and early treatment of keloid recurrence.

[0007] 3. Uncertain treatment methods

[0008] There is no objective standard for choosing the treatment method for keloids. In most cases, doctors choose the treatment plan based on their experience and habits.

[0009] 4. The treatment results are difficult to predict

[0010] When treating physicians are studying treatment plans and determining treatment techniques, most are unable to accurately predict the treatment outcomes for their patients. Summary of the invention

[0011] The purpose of the present invention is to provide a scar measurement method that can perform remote measurement and improve accuracy in order to solve the above problems.

[0012] In order to achieve the above object, the technical solution of the present invention is:

[0013] A scar measurement method comprises the following steps:

[0014] S1. The patient puts a label sticker with positioning points on the skin, then uses a mobile phone to take a photo of the scar location and the label sticker, generates a picture, and sends the picture to the APP background through the mobile phone APP;

[0015] S2. The APP background generates an overall three-dimensional model from the picture;

[0016] S3, segmenting the normal skin and keloid area of ​​the patient in the overall three-dimensional model to obtain the keloid three-dimensional model; and separating the normal skin model and the ruler model from the overall three-dimensional model;

[0017] S4. Obtain scar color difference data by comparing the normal skin model with the keloid three-dimensional model; obtain the proportional relationship between the ruler model and the actual size by comparing the distance between the ruler model and the positioning point on the label sticker in the picture; obtain the model measurement data of the scar by measuring the keloid three-dimensional model and obtain the actual measurement data of the scar by the proportional relationship between the ruler model and the actual size.

[0018] Furthermore, generating an overall three-dimensional model from the image in step S2 includes the following steps:

[0019] S21, extracting feature points from the image;

[0020] S22, search for neighbor feature points from k-dtree;

[0021] S23, performing RANSAC image matching;

[0022] S24, constructing cloud point data;

[0023] S25, constructing triangular surfaces and textures through cloud point data;

[0024] S26. Output the overall three-dimensional model through triangulated surfaces and mapping.

[0025] Furthermore, obtaining the keloid three-dimensional model from the overall three-dimensional model in step S3 comprises the following steps:

[0026] S31, performing triangular tiling processing on the entire three-dimensional model;

[0027] S32, analyzing each triangular face and extracting a texture image according to UV coordinates;

[0028] S33, separation of normal skin and keloid area by color difference;

[0029] S34, converting the texture image into a grayscale image;

[0030] S35, using Haar to decompose the grayscale image, obtain the feature vector and perform normalization processing;

[0031] S36, eliminating the remaining texture images and triangular faces to generate a keloid three-dimensional model.

[0032] Furthermore, obtaining scar model measurement data by measuring the keloid three-dimensional model in step S4 includes the following steps:

[0033] S41, calculating the overall orientation of the three-dimensional keloid model by using the normal of the three-dimensional keloid model;

[0034] S42, placing the keloid three-dimensional model horizontally as a whole, performing radiographic detection on the coverage area of ​​the keloid three-dimensional model, obtaining the undulation data within the pixel unit of the keloid three-dimensional model, and imaging the undulation data to obtain an elevation map;

[0035] S43, obtaining cross-sectional data of the three-dimensional model of the keloid through the elevation map, performing two linear fits on each section of the cross-sectional data to make each section of the cross-sectional data relatively horizontal, then calculating the area between the top curve and the bottom horizontal line in the cross-sectional data, accumulating the area in each section of the cross-sectional data, and obtaining the volume of the three-dimensional model of the keloid;

[0036] S44, obtaining the maximum thickness of the three-dimensional keloid model through the cross-sectional data of the three-dimensional keloid model;

[0037] S45, placing the keloid three-dimensional model as a whole horizontally, and performing a top-down projection on it to obtain a top-down polygonal image, wherein the perimeter of the top-down polygonal image is the perimeter of the keloid three-dimensional model;

[0038] S46, obtaining the minimum enclosing rectangle of the top-view polygonal image, the length and width of the minimum enclosing rectangle being the length and width of the keloid three-dimensional model.

[0039] Compared with the prior art, the present invention has the following advantages and positive effects:

[0040] The present invention enables customers to use APP for remote consultation, and doctors can obtain data such as color difference, volume, circumference, etc. of scars based on pictures taken by patients, which facilitates doctors to conduct long-term diagnosis and treatment tracking; at the same time, doctors can determine the next treatment plan based on the different treatment effects in each treatment stage, thereby effectively improving the treatment effect of scars; on the other hand, the present invention can perform three-dimensional modeling through pictures taken by patients, thereby obtaining various data of scars, which improves the accuracy of scar measurement and further improves the use effect of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0042] Figure 1 It is a framework flow chart of the present invention;

[0043] Figure 2 A framework flow chart for generating an overall 3D model for an image;

[0044] Figure 3 A framework flow chart for generating a keloid three-dimensional model for the overall three-dimensional model;

[0045] Figure 4 Flowchart of the framework for calculating measurement data for keloid 3D models. DETAILED DESCRIPTION

[0046] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work, any modifications, equivalent substitutions, improvements, etc., should be included in the protection scope of the present invention.

[0047] like Figures 1 to 4 As shown, the scar measurement method in this embodiment includes the following steps:

[0048] 1. Collect pictures by taking photos with a mobile phone. Users only need to pick up label stickers. There are multiple positioning points on the label stickers, and the distance between the positioning points is a limited value; log in to the APP to take pictures and transfer the photos to the APP background.

[0049] 2. Model Generation

[0050] The background program performs camera alignment, feature point extraction, cloud point construction, triangle surface construction, and UV and texture generation based on the image of the affected area taken by the camera. Generating an overall 3D model from the image is the most basic step of the application, and all subsequent data is calculated from the 3D model.

[0051] 3. Keloid segmentation

[0052] Based on the 3D model generated by the background program, we use the 3D visualization tool to split the model into several parts:

[0053] (1) Normal skin: select normal skin from the overall 3D model, with a reference ruler length of about 1 cm, and then separate it and export it as a normal skin model;

[0054] (2) Scale area: separate the scale area from the overall 3D model. This area can be the minimum rectangular range of the scale and exported as a scale model.

[0055] (3) Keloid: Segment the scar area from the overall 3D model based on the obvious protrusions or color changes of the scar, and export the scar area as a keloid model;

[0056] 4. Data calculation

[0057] (1) Color difference calculation: import the normal skin model and the keloid model into the analysis program to obtain the average color of the normal skin. Linear interpolation is performed between this color and the color of the keloid to obtain the color difference range of the keloid.

[0058] (2) Unified units: Import the ruler model and automatically obtain the positions of the three positioning points in the label sticker through image recognition. According to the distance between the three positioning points in the model and the actual size in reality, establish the unit ratio relationship between the model and the actual size, and the unit ratio relationship between the image pixels and the actual size;

[0059] (3) Generate an elevation map: import the keloid model, calculate the overall orientation through the model normal, place the keloid model horizontally, perform X-ray detection on the model coverage area in the virtual space, obtain the undulation data within the pixel unit of the keloid model, and visualize the data, i.e., the elevation map;

[0060] (4) Volume calculation: The row and column data, i.e., the cross-sectional data of the model, are obtained through the data in the elevation map. Two linear fits are performed on each cross-sectional data segment to make each cross-sectional data segment relatively horizontal. The closed area between the top curve of the cross-sectional data segment and the lowest horizontal line is then calculated. The accumulated cross-sectional area of ​​each segment is the model volume.

[0061] (5) Thickness calculation: The maximum thickness is obtained when calculating the cross-sectional area;

[0062] (6) Calculation of perimeter: After placing the keloid model horizontally, perform top-view projection, segment the obtained projection image, and obtain polygonal data of the image. The total length of the polygonal line segments is the perimeter.

[0063] (7) Calculation of length and width: Similar to the perimeter, obtain the top view projection image and segment it, and calculate the minimum enclosing rectangle of the projection image. The length and width of the rectangle are the length and width values ​​of the desired keloid model.

[0064] The present invention enables customers to use APP for remote consultation, and doctors can obtain data such as color difference, volume, circumference, etc. of scars based on pictures taken by patients, which facilitates doctors to conduct long-term diagnosis and treatment tracking; at the same time, doctors can determine the next treatment plan based on the different treatment effects in each treatment stage, thereby effectively improving the treatment effect of scars; on the other hand, the present invention can perform three-dimensional modeling through pictures taken by patients, thereby obtaining various data of scars, which improves the accuracy of scar measurement and further improves the use effect of the present invention.

Claims

1. A scar measurement method, characterized in that: The following steps are involved: S1. The patient puts a label sticker with positioning points on the skin, then uses a mobile phone to take a photo of the scar location and the label sticker, generates a picture, and sends the picture to the APP background through the mobile phone APP; S2. The APP background generates an overall three-dimensional model from the picture; S3, segmenting the normal skin and keloid area of ​​the patient in the overall three-dimensional model to obtain the keloid three-dimensional model; and separating the normal skin model and the ruler model from the overall three-dimensional model; S4, obtaining scar color difference data by comparing the normal skin model with the keloid three-dimensional model; The proportional relationship between the ruler model and the actual size is obtained by comparing the distance between the ruler model and the positioning point on the label sticker in the picture; the model measurement data of the scar is obtained by measuring the keloid three-dimensional model, and the actual measurement data of the scar is obtained by the proportional relationship between the ruler model and the actual size; The step S4 of obtaining scar model measurement data by measuring the keloid three-dimensional model comprises the following steps: S41, calculating the overall orientation of the three-dimensional keloid model by using the normal of the three-dimensional keloid model; S42, placing the keloid three-dimensional model horizontally as a whole, performing radiographic detection on the coverage area of ​​the keloid three-dimensional model, obtaining the undulation data within the pixel unit of the keloid three-dimensional model, and imaging the undulation data to obtain an elevation map; S43, obtaining cross-sectional data of the three-dimensional model of the keloid through the elevation map, performing two linear fits on each section of the cross-sectional data to make each section of the cross-sectional data relatively horizontal, then calculating the area between the top curve and the bottom horizontal line in the cross-sectional data, accumulating the area in each section of the cross-sectional data, and obtaining the volume of the three-dimensional model of the keloid; S44, obtaining the maximum thickness of the three-dimensional keloid model through the cross-sectional data of the three-dimensional keloid model; S45, placing the keloid three-dimensional model as a whole horizontally, and performing a top-down projection on it to obtain a top-down polygonal image, wherein the perimeter of the top-down polygonal image is the perimeter of the keloid three-dimensional model; S46, obtaining the minimum enclosing rectangle of the top-view polygonal image, the length and width of the minimum enclosing rectangle being the length and width of the keloid three-dimensional model.

2. The scar measurement method according to claim 1, characterized in that: The step S2 of generating an overall three-dimensional model from the image comprises the following steps: S21, extracting feature points from the image; S22, search for neighbor feature points from k-dtree; S23, performing RANSAC image matching; S24, constructing cloud point data; S25, constructing triangular surfaces and textures through cloud point data; S26. Output the overall three-dimensional model through triangulated surfaces and mapping.

3. The scar measurement method according to claim 1, characterized in that: The step S3 of obtaining the keloid three-dimensional model from the overall three-dimensional model comprises the following steps: S31, performing triangular tiling processing on the entire three-dimensional model; S32, analyzing each triangular face and extracting a texture image according to UV coordinates; S33, separation of normal skin and keloid area by color difference; S34, converting the texture image into a grayscale image; S35, using Haar to decompose the grayscale image, obtain the feature vector and perform normalization processing; S36, eliminating the remaining texture images and triangular faces to generate a keloid three-dimensional model.

Images