Illumination projection-based real color restoration method and device for lesion picture
By combining light projection and temperature information, a skin color restoration network model is constructed, which solves the skin color restoration problem under the influence of light and temperature differences in skin medical care, and achieves a more accurate skin color restoration effect.
Patent Information
- Application Number
- CN202410979560.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-12-06
AI Technical Summary
In skin medical care, accurately reducing the true color of the skin is an urgent problem, especially under different lighting conditions and the environmental temperature difference is large, and the existing technology is difficult to effectively solve.
Using a light projection method, by casting light to the patient's skin surface, obtaining skin pictures and temperature information, and building a skin color restoration network model, combining error constraints for model training, eliminating the impact of temperature and light differences, and finally achieving skin color reduction.
This method can more accurately restore skin color, adapt to people of different skin colors, and improve the accuracy of the reduction results, eliminating the impact of environmental temperature differences.
Smart Images

Figure CN119090775A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent restoration of true skin color, and in particular to a method and device for restoring the true color of a lesion image based on light projection. Background Art
[0002] As people's living standards improve, their requirements for appearance and skin quality are getting higher and higher, and the number of dermatology patients is increasing. In skin medicine, how to accurately restore the true color of the skin is an urgent problem to be solved. Summary of the invention
[0003] The present application provides a method for restoring the true color of a lesion image based on light projection, comprising:
[0004] Project light onto the patient's skin surface.
[0005] Obtain a picture of the patient's skin containing the lesion area under light exposure;
[0006] Obtain the patient's skin temperature information under light;
[0007] The skin image is preprocessed, a skin color restoration network model is pre-constructed using the preprocessed skin image and the temperature information, and a skin color restoration result output by the color restoration network model is obtained.
[0008] Optionally, before inputting the temperature information and the skin image into a pre-built skin color restoration network model, the method further includes:
[0009] Obtain historical skin pictures taken under different light intensities, and obtain the skin temperature of the patient under each light intensity, and use the historical skin pictures and the skin temperature of the patient under each light intensity as a training set;
[0010] Construct error constraints;
[0011] Model training is performed based on the training set, the error constraint condition and the set number of iterations to obtain a final skin color restoration network model.
[0012] Optionally, the error constraint condition includes:
[0013] Temperature error correction conditions and / or radiation error correction conditions.
[0014] Optionally, the temperature error correction condition satisfies the following relationship:
[0015] a=[1+α(T-T0)]a0;
[0016] Wherein, α represents the expansion coefficient, T represents the theoretical temperature, T0 represents the actual temperature, and a0 represents a constant.
[0017] Optionally, the reflection error correction condition satisfies the following relationship:
[0018]
[0019] Where n is a constant, λ is the wavelength, and θ is the reflection angle.
[0020] Optionally, preprocessing the skin image includes:
[0021] The skin image is subjected to scale normalization, color space conversion, and color scale normalization processing.
[0022] Optionally, the illumination includes: infrared light and visible light.
[0023] Optionally, the illumination intensity is divided into N types, where N is a positive integer.
[0024] The present application also provides a system for restoring the true colors of lesion images based on light projection, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the method described in the first aspect when executing the computer program.
[0025] The present application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of the method described in the first aspect are implemented.
[0026] Beneficial effects:
[0027] The method for restoring the true color of lesion images based on light projection provided by the present invention can more clearly obtain the skin image of the patient including the lesion area under light illumination by projecting light onto the skin surface of the patient. It is suitable for people of all skin colors. Moreover, by considering temperature information to establish a skin color restoration network model, it can eliminate the influence caused by the temperature difference in the environment during shooting, making the final restoration result more accurate. BRIEF DESCRIPTION OF THE DRAWINGS
[0028] Figure 1 The present invention is a flowchart of a method for restoring the true color of a lesion image based on light projection according to a preferred embodiment of the present invention. DETAILED DESCRIPTION
[0029] The technical solution of the present invention is described clearly and completely below. Obviously, the described embodiments are only a 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 are within the scope of protection of the present invention.
[0030] Unless otherwise defined, the technical terms or scientific terms used in the present invention shall have the usual meanings understood by persons with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. Similarly, words such as "one" or "one" do not indicate quantity restrictions, but indicate the existence of at least one. Words such as "connect" or "connected" are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship also changes accordingly.
[0031] See also Figure 1 The present application provides a method for restoring the true color of a lesion image based on light projection, comprising:
[0032] Project light onto the patient's skin surface.
[0033] Obtain a picture of the patient's skin containing the lesion area under light exposure;
[0034] Obtain the patient's skin temperature information under light;
[0035] The skin image is preprocessed, a skin color restoration network model is pre-constructed using the preprocessed skin image and the temperature information, and a skin color restoration result output by the color restoration network model is obtained.
[0036] The above-mentioned method for restoring the true color of lesion images based on light projection can more clearly obtain the skin image of the patient containing the lesion area under light by projecting light onto the patient's skin surface. It is suitable for people of all skin colors. Moreover, by considering temperature information to establish a skin color restoration network model, it can eliminate the influence of temperature differences in the environment during shooting, making the final restoration result more accurate.
[0037] Optionally, before inputting the temperature information and the skin image into a pre-built skin color restoration network model, the method further includes:
[0038] Obtain historical skin pictures taken under different light intensities, and obtain the skin temperature of the patient under each light intensity, and use the historical skin pictures and the skin temperature of the patient under each light intensity as a training set;
[0039] Construct error constraints;
[0040] Model training is performed based on the training set, the error constraint condition and the set number of iterations to obtain a final skin color restoration network model.
[0041] First of all, it is worth pointing out that different light intensities will have different effects on the shooting results during specific shooting. The pigmentation of human skin is not a simple feature, and is not determined by just one gene or a few simple groups of genes, but by the simultaneous influence of many genes and the living environment. Based on this complexity, in this optional implementation, by considering light intensity and temperature to train the skin color restoration network model, the learning features of the trained model can be more, and the output results obtained can be more accurate.
[0042] In one example, the error constraints include:
[0043] Temperature error correction condition and\or reflection error correction condition.
[0044] Optionally, the temperature error correction condition satisfies the following relationship:
[0045] a=[1+α(T-T0)]a0;
[0046] Wherein, α represents the expansion coefficient, T represents the theoretical temperature, T0 represents the actual temperature, and a0 represents a constant.
[0047] Since the influence of temperature can only be eliminated in a constant temperature environment, and the actual shooting conditions cannot be in a constant temperature environment such as a constant temperature box, and the actual temperature will be affected to a certain extent due to climate or light influences, in this embodiment, by setting the temperature error correction condition, the temperature influence can be eliminated and the final restoration accuracy can be improved.
[0048] In one example, the reflection error correction condition satisfies the following relationship:
[0049]
[0050] Where n is a constant, λ is the wavelength, and θ is the reflection angle.
[0051] Light skin reflects more light than dark skin, and different wavelengths of light are reflected to different degrees. Therefore, it is very meaningful to use skin reflectivity as an indicator for studying skin color. In this embodiment, the accuracy of color matching can be improved by setting reflection error correction conditions.
[0052] Optionally, preprocessing the skin image includes:
[0053] The skin image is subjected to scale normalization, color space conversion, and color scale normalization processing.
[0054] Optionally, the illumination includes: infrared light and visible light.
[0055] Optionally, the illumination intensity is divided into N types, where N is a positive integer.
[0056] The present application also provides a system for restoring the true color of lesion images based on light projection, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the computer program.
[0057] The system for restoring the true color of lesion images based on light projection can implement the various embodiments of the above-mentioned method for restoring the true color of lesion images based on light projection, and can achieve the same beneficial effects, which will not be described in detail here.
[0058] The present application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.
[0059] The computer device can implement the various embodiments of the above-mentioned method for restoring the true color of lesion images based on light projection, and can achieve the same beneficial effects, which will not be described in detail here.
[0060] The preferred specific embodiments of the present invention are described in detail above. It should be understood that a person skilled in the art can make many modifications and changes based on the concept of the present invention without creative work. Therefore, any technical solution that can be obtained by a person skilled in the art through logical analysis, reasoning or limited experiments based on the concept of the present invention on the basis of the prior art should be within the scope of protection determined by the claims.
Claims
1. A method for restoring the true color of lesion images based on light projection, It is characterized in that include: Projecting light onto the surface of the patient's skin; Obtain a picture of the patient's skin containing the lesion area under light exposure; Obtain the patient's skin temperature information under light; The skin image is preprocessed, a skin color restoration network model is pre-constructed using the preprocessed skin image and the temperature information, and a skin color restoration result output by the color restoration network model is obtained.
2. The method for restoring the true color of a lesion image based on light projection according to claim 1, It is characterized in that Before inputting the temperature information and the skin image into a pre-built skin color restoration network model, the method further includes: Obtain historical skin pictures taken under different light intensities, and obtain the skin temperature of the patient under each light intensity, and use the historical skin pictures and the skin temperature of the patient under each light intensity as a training set; Construct error constraints; Model training is performed based on the training set, the error constraint condition and the set number of iterations to obtain a final skin color restoration network model.
3. The method for restoring the true color of a lesion image based on light projection according to claim 2, It is characterized in that The error constraints include: Temperature error correction condition and\or reflection error correction condition.
4. The method for restoring the true color of a lesion image based on light projection according to claim 3, It is characterized in that The temperature error correction condition satisfies the following relationship: a=[1+α(T-T 0 )]a 0 ; In the formula, α represents the expansion coefficient, T represents the theoretical temperature, and T 0 Indicates the actual temperature, a 0 Represents a constant.
5. The method for restoring the true color of a lesion image based on light projection according to claim 3, It is characterized in that The reflection error correction condition satisfies the following relationship: Where n is a constant, λ is the wavelength, and θ is the reflection angle.
6. The method for restoring the true color of a lesion image based on light projection according to claim 1, It is characterized in that The preprocessing of the skin image comprises: The skin image is subjected to scale normalization, color space conversion, and color scale normalization processing.
7. The method for restoring the true color of a lesion image based on light projection according to claim 1, It is characterized in that The illumination includes: infrared light and visible light.
8. The method for restoring the true color of a lesion image based on light projection according to claim 1, It is characterized in that The intensity of the illumination is divided into N types, where N is a positive integer.
9. A system for restoring the true color of a lesion image based on light projection, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor. It is characterized in that When the processor executes the computer program, the steps of the method described in any one of claims 1 to 8 are implemented.
10. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.
