Electronic apparatus, and skin quality analysis method and device used for the same

By controlling imaging devices to capture high-quality frames from a video stream and performing analysis on these frames, the method enhances the accuracy and consistency of skin quality analysis, addressing variability in conventional methods.

JP2024166053AActive Publication Date: 2024-11-28SHENZHEN ACCO TECH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
JP2023219108
Authority / Receiving Office
JP · JP
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-18
Filing Date
2023-12-26
Publication Date
2024-11-28
Estimated Expiration
2043-12-26

AI Technical Summary

Technical Problem

Conventional skin analysis methods are hindered by variability in image quality, leading to inconsistent and less effective analysis results due to factors like tilt angle and light irradiation, which affect the comparability of measurements across different individuals and sessions.

Method used

A method and device that control an imaging device to capture a video stream of skin location, select high-quality frames from the stream, and perform skin quality analysis on these frames to ensure consistent and accurate results.

Benefits of technology

This approach allows for real-time, efficient, and painless image acquisition, ensuring all images are of high quality, thereby improving the stability and comparability of skin quality analysis results and reducing noise from image quality issues.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 2024166053000001_ABST
    Figure 2024166053000001_ABST
Patent Text Reader

Abstract

To provide an electronic apparatus, and a skin quality analysis method and device used for the same.SOLUTION: A skin quality analysis method includes the steps of: controlling an imaging apparatus to acquire a video bit stream of a target skin portion including images of some frames in a time zone set in advance; selecting, from the images of some frames in the bit stream, an image of at least one frame matching an image quality request and taking the selected image as a target image of the at least one frame; performing skin quality analysis by using the target image of the at least one frame to obtain a skin quality analysis result; and outputting the skin quality analysis result. Consequently, a user experience is improved in an open environment by not acquiring an image statically, but by acquiring the video stream in real time and achieving efficient and insensitive acquisition. Further, all images for the skin quality analysis can be controlled to have the same or higher quality level, stability and comparability of the skin quality analysis results are improved, a noise problem caused by a problem of the quality of the image itself is solved, and the accuracy of skin quality analysis is improved.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Description

[Technical field]

[0001] The present application relates to the field of cosmetics, and in particular to an electronic device and a method and device for analyzing skin quality thereof. [Background technology]

[0002] In conventional skin analysis methods, input face images are generally directly processed and analyzed by AI algorithms to obtain parameters for each aspect of skin quality (roughness, spots, pores, wrinkles, acne, etc.). In the process of this method, image quality has a great influence on the analysis results, and even when the same person performs measurement analysis at the same time, there are large differences between the analysis results due to the influence of factors such as image quality, such as clarity, distance to the camera, tilt angle, and light irradiation. Such variability due to variations in image quality strongly reduces the effectiveness of skin quality analysis and the comparability between measurement results of different people and measurement results of the same person at different times. Summary of the Invention [Problem to be solved by the invention]

[0003] The present application aims to solve at least one of the technical problems existing in the prior art. [Means for solving the problem]

[0004] In order to solve the above technical problems, the technical solutions of this application are as follows:

[0005] In a first aspect, the present application provides a skin quality analysis method for use in an electronic device, comprising: controlling an image capture device to capture a video stream of the target skin location, the video stream including a number of frames of images within a pre-defined time period; selecting at least one frame of image that meets an image quality requirement from the several frame images of the video stream as at least one target frame of image; performing a skin quality analysis using the at least one frame of the target image to obtain a skin quality analysis result; and outputting the skin quality analysis result.

[0006] In a second aspect, the present application provides a method for analyzing skin quality comprising: a processor, a memory, and an imaging device, the imaging device and the memory being electrically connected to the processor, and a computer program being stored in the memory; and the processor executing the computer program to perform steps of a skin quality analysis method, namely: controlling an image capture device to capture a video stream of the target skin location, the video stream including a number of frames of images within a pre-defined time period; selecting at least one frame of image that meets an image quality requirement from the several frame images of the video stream as at least one target frame of image; performing a skin quality analysis using the at least one frame of the target image to obtain a skin quality analysis result; and outputting the skin quality analysis result.

[0007] In a third aspect, the present application provides a method for analyzing skin quality, the method being executed by a single processor, comprising the steps of: controlling an image capture device to capture a video stream of the target skin location, the video stream including a number of frames of images within a pre-defined time period; selecting at least one frame of image that meets an image quality requirement from the several frame images of the video stream as at least one target frame of image; performing a skin quality analysis using the at least one frame of the target image to obtain a skin quality analysis result; and outputting the skin quality analysis result.

[0008] In a fourth aspect, the present application provides a method for producing a method for the treatment of a disease comprising: an acquisition control module adapted to control the imaging device to acquire a video stream of the target skin location, the video stream including a number of frames of images within a predefined time period; an image selection module, which is used for selecting at least one frame of image that meets an image quality requirement from some frame images of the video stream, as at least one frame of target image; a skin quality analysis module used for performing a skin quality analysis using the at least one frame of the target image to obtain a skin quality analysis result; The present invention further provides a skin quality measuring device comprising: an output control module adapted to output the skin quality analysis result. Effect of the Invention

[0009] Compared with the prior art, the useful effects of the present invention are as follows: This allows real-time acquisition of video streams, rather than statically acquiring images in an open environment, realizing efficient and sensory-free acquisition and improving the user experience. In addition, all images for skin quality analysis can be controlled to be at least the same quality level, improving the stability and comparability of skin quality analysis results, eliminating noise issues caused by quality issues in the images themselves, and improving the accuracy of skin quality analysis. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 is a schematic diagram of a module of an electronic device according to an embodiment of the present application. [Diagram 2] 1 is a schematic flowchart of a skin quality analysis method according to an embodiment of the present application. [Diagram 3] 11 shows examples of statistics for various cases of labeling frame images in one embodiment of the present application. [Figure 4] 4 is a schematic flowchart of a skin quality analysis method according to another embodiment of the present application. [Diagram 5] FIG. 2 is a module schematic diagram of a skin quality analysis device provided in one embodiment of the present application. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] In the following, the embodiments of the present application are described in detail, and examples of the embodiments are shown in the drawings, in which the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the drawings are illustrative and are intended to interpret the present application, and should not be construed as limiting the present application.

[0012] Additionally, the terms "first", "second", etc. are for descriptive purposes only and should not be understood to indicate or imply a relative importance or number of the technical features being indicated. Thus, a feature qualified as "first" or "second" may expressly or imply the inclusion of one such feature or more of such features. In the present description, unless expressly and specifically limited, "plurality" means two or more than two.

[0013] In the present application, unless otherwise clearly defined or limited, the terms "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, they may be fixedly connected, detachably connected, or integrated. They may be mechanically connected or electrically connected. They may be directly connected, or further indirectly connected via an intermediate medium, the interiors of two elements may be in communication, or two elements may be in an interactive relationship. Those skilled in the art may understand the specific meaning of the above terms in the present application according to the specific situation.

[0014] Please refer to FIG. 1, which is a schematic diagram of a module of an electronic device 1 in an embodiment of the present application. The electronic device 1 is used to perform skin quality analysis. The electronic device 1 may be, but is not limited to, a skin analysis device, a beauty device or smart terminal having a skin detection function, a skin analysis system consisting of an imaging device having an image acquisition function and a skin analysis device having a data analysis function, or other devices capable of realizing image acquisition and data analysis functions, and is not limited here. Specifically, the electronic device 1 includes a memory 11 and a processor 12. The memory 11 is electrically connected to the processor 12. A computer program is stored in the memory 11. The processor 12 executes the computer program to execute steps of a skin quality analysis method. The electronic device 1 further includes an imaging device 13. The imaging device 13 is electrically connected to the processor 12. The imaging device 13 may include a camera system consisting of one camera or multiple cameras, and may be any device that can accurately acquire image data in a timely manner when measuring skin quality, and is not limited here. In this embodiment, when the electronic device 1 is a device that integrates image acquisition and data analysis functions, the imaging device 13 includes a front camera. It can be understood that in other embodiments, the imaging device 13 can not only include a front camera, but also include a rear camera, and in some cases, the camera for acquiring the video stream of the target skin location can be switched between the front camera and the rear camera.

[0015] Please refer to Fig. 2, which is a schematic flow chart of a skin type analysis method according to an embodiment of the present application. The skin type analysis method can be used in the electronic device 1 to realize skin type analysis. It is understood that some steps of the skin type analysis method can be added, deleted, or rearranged, and are not limited thereto. Specifically, the skin type analysis method includes the following steps S210, S220, S230, and S240. Step S210: Control the imaging device 13 to obtain a video stream of the target skin location, including images of a few frames within a preset time period. Step S220: Select at least one frame of image that meets an image quality requirement from some frames of images of the video stream, and set it as at least one frame of target image. Step S230: A skin quality analysis is performed using the at least one frame of the target image to obtain a skin quality analysis result. Step S240: The skin quality analysis result is output.

[0016] In this application, the steps of the skin quality analysis method are as follows: first, control the image capture device 13 to capture a video stream of a target skin location; then extract a plurality of frame images from the captured video stream sample information; and then select a frame image that meets the image quality requirements from the plurality of frame images to perform skin quality analysis. This allows real-time capture of video streams, rather than statically capturing images in an open environment, to achieve efficient and unobtrusive capture and improve user experience. In addition, all images for skin quality analysis can be controlled to be at or above the same quality level, improving the stability and comparability of skin quality analysis results, eliminating noise problems caused by quality problems of the images themselves, and improving the accuracy of skin quality analysis.

[0017] Here, the processor 12 controls the imaging device 13 to obtain a video stream of a target skin location within a predetermined time period, and the time length of the predetermined time period can be set according to actual needs, for example, 5 seconds, 10 seconds, 15 seconds, etc., and is not limited thereto. As can be understood, the longer the time length of the video stream, the more frame images the video stream contains, and therefore the higher the probability of selecting a high-quality image. However, it goes without saying that this will be accompanied by an increase in the amount of calculation data. Therefore, in practical applications, it is necessary to make a certain trade-off between high-quality images and the amount of calculation data.

[0018] Since the generation times of some frame images in the video stream are different, and the imaging parameters at different times may change slightly, the image quality of different frame images captured with different imaging parameters is not all the same, and some frame images have relatively good image quality and some frame images have relatively bad image quality. Therefore, by selecting at least one frame image that meets the image quality requirements from some frame images in the video stream, and using each frame image as a target image for one frame, the noise generated in the skin type analysis due to the frame images that do not meet the quality requirements can be removed, and the accuracy of the skin type analysis can be further improved.

[0019] In some embodiments, the manner of selecting at least one frame image that meets the image quality requirement from the several frame images of the video stream may be various and is not limited thereto.

[0020] Two of these selection methods will be introduced in detail below.

[0021] The first type of method is as follows:

[0022] The step of selecting at least one frame of an image that meets an image quality requirement from some frame images of the video stream and setting the at least one frame of the image as a target image includes: extracting a number of frames of images from the video stream; scoring each of the plurality of frames of images to obtain an image quality score for each frame of image; and selecting at least one frame of image that meets an image quality requirement from the images of the plurality of frames according to the image quality score of each frame of image, and setting the selected frame of image as a target image of the at least one frame.

[0023] As a result, in the first type of method, it is first necessary to perform image quality scoring on each frame image among the extracted multiple frame images, and after obtaining the image quality scores for each frame image among the extracted multiple frame images, at least one frame image can be selected in order of best quality to be used as a target image for the at least one frame.

[0024] In some embodiments, the step of selecting at least one frame of image that meets an image quality requirement from the plurality of frame images according to the image quality score of each frame of image and setting the at least one frame of image as a target image includes: A step of selecting one frame having the highest image quality score or at least two frames having the relatively highest image quality scores from the images of the plurality of frames according to the image quality scores of the images of each frame, and setting the selected frame as a target image of the at least one frame; or The method includes a step of randomly selecting at least one image frame that meets an image quality requirement from the images of the plurality of frames according to the image quality score of each image frame, and setting the selected image frame as a target image for the at least one image frame.

[0025] As a result, according to the above first type of method, the frame image with the highest or relatively highest image quality can be selected from the multiple frame images of the video stream in order of best quality and set as the target image, thereby further improving the accuracy of skin type analysis.

[0026] The second type of method is as follows:

[0027] The step of selecting at least one frame of an image that meets an image quality requirement from some frame images of the video stream and setting the at least one frame of the image as a target image includes: extracting a number of frames of images from the video stream; a step of scoring the images of the plurality of frames in order starting from a first frame image, scoring the images of each frame to obtain an image quality score, and then determining whether the image quality score of the image of this frame satisfies an image quality requirement, and if the image quality score of the image of this frame satisfies the image quality requirement, determining that the image of this frame is to be a target image; and if the image quality score of the image of this frame does not satisfy the image quality requirement, scoring an image of a next frame of the images of the plurality of frames until a target image is determined whose image quality score satisfies the image quality requirement.

[0028] For example, first, a first frame image of the plurality of frame images is scored to obtain an image quality score for the first frame image; determining whether an image quality score of the first frame image satisfies an image quality requirement, and determining that the first frame image is a target image if the image quality score of the first frame image satisfies the image quality requirement; if the image quality score of the first frame image does not satisfy the image quality requirement, scoring a second frame image of the plurality of frame images to obtain an image quality score of the second frame image; determining again whether the image quality score of the second frame image satisfies the image quality requirement, and if the image quality score of the second frame image satisfies the image quality requirement, determining that the second frame image is the target image; If the image quality score of the second frame image does not satisfy the image quality requirement, a third frame image of the plurality of frame images is scored, and this is repeated until a target image is determined whose image quality score satisfies the image quality requirement.

[0029] In this way, in the second type method, image quality scoring is performed from the first frame image, and it is judged whether the image quality score of the first frame image meets the image quality requirement, and if the first frame image meets the image quality requirement, the first frame image is set as the target image. In this case, if the purpose is to select only one frame of the target image and perform skin quality analysis, the target image selection procedure can be terminated without performing image quality scoring on the second frame image, the third frame image, etc., among the images of the multiple frames, and judging whether they meet the image quality requirement. Therefore, the amount of data calculation can be reduced to an extremely large extent. After the target image of the first frame is determined, when it is necessary to further determine the target image of the second frame, the target image of the third frame, etc., the similar judgment process is continued, and when the target image of the second frame, the target image of the third frame, etc. are selected, the target image selection procedure is terminated.

[0030] In some embodiments, even if image quality scoring is performed on all of the images of the multiple frames and a determination is made as to whether or not they meet the image quality requirement, if a frame image that meets the image quality requirement is not selected as a target image, a next group of images of multiple frames different from the images of the multiple frames is again extracted from the video stream until a target image that meets the image quality requirement is selected, and image quality scoring is performed again on the images of the multiple frames of the next group to select a target image that meets the image quality requirement.

[0031] In some embodiments, a group of frame images is extracted from a video stream, for example, the group of frame images may be 10 consecutive frame images, which may be a group of frame images selected from a start position of the video stream, and the next group of frame images is extracted again from the end of the group of frame images, for example, 10 consecutive frame images are extracted from the 11th frame image, and so on until a target image is determined whose image quality score satisfies the image quality requirement. In other embodiments, the group of frame images is extracted from a video stream, for example, the group of frame images is selected from an arbitrary position of the video stream, and the next group of frame images is extracted again from another position that does not overlap with the group of frame images, so long as it is ensured that the next group of frame images does not overlap with the frame images of the group of frame images. For example, the group of frame images may be 10 consecutive frame images from the 21st frame image, and the next group of frame images may be 10 consecutive frame images from the 41st frame image.

[0032] In some embodiments, performing image quality scoring for each frame of image comprises: performing quality labeling on each frame image from at least one dimension including at least one of sharpness, light illumination, and frontality to obtain at least one labeling result; The method includes a step of obtaining an image quality score for each frame image based on the labeling result of the image for each frame.

[0033] It will be appreciated that in other embodiments, the at least one dimension may include other dimensions and is not limited thereto.

[0034] Please refer to FIG. 3, which is a statistical example of various cases of labeling frame images in one embodiment of the present application.

[0035] FIG. 3 shows three dimensions for labeling the image of each frame, which are clarity, frontality, and light illumination, respectively. Each dimension includes pass and fail cases. If pass is satisfied, a first label is attached to the image of this frame in the corresponding dimension, and if fail is satisfied, a second label is attached to the image of this frame in the corresponding dimension. The first label and the second label each have different meanings. For example, in the case of the clarity dimension, the first label is 1 to indicate clarity, and the second label is 0 to indicate lack of clarity. The definition of pass clarity is that the face is clear and the aesthetic features can be identified. The cases of fail clarity include cases where the face is significantly blurred, where acne, bruises, and fine wrinkles cannot be identified, and where partial areas of the face are blurred. In the case of the frontality dimension, the first label is 1 to indicate frontal, and the second label is 0 to indicate non-frontal. The definition of pass frontality is that the face is captured from the front and is not tilted or occluded. Examples of unacceptable frontality include face down or face up, tilted to the left or right, incomplete or occluded, prominent facial expression, etc. In the case of the light exposure dimension, the first label being 1 indicates good light exposure, and the second label being 0 indicates poor light exposure. The definition of acceptable light exposure is that the light exposure is sufficient, uniform, there are no strong light spots, and color shifts do not affect the judgment of spots, bruises, or acne. Examples of unacceptable light exposure include light exposure that is too strong or too dark, light exposure that is uneven, too red or too green, and there are strong light spots.

[0036] In some embodiments, the step of outputting the labeling result for each frame image includes: The method includes the step of outputting at least one dimension labeling result for each frame image, where each dimension labeling result represents a probability value that the image of the frame will perform well in that dimension.

[0037] For example, in one embodiment, an example image quality assessment for a target image is as follows:

[0038] The output of the image quality assessment model for one frame of image is On=(Pqo, Pzo, Pgo), where Pqo, Pzo, and Pgo respectively represent the probability values ​​of good clarity, frontality, and light illumination, and the value range is [0,1]. For example, a Pqo value of 0.7 indicates that the probability that the frame image is clear is 0.7. Pzo=0.3 indicates that the probability that the frame image is frontal is 0.3 (i.e., not frontal), and Pgo=0.5 indicates that the probability that the light illumination of the frame image is good is 0.5 (between good and bad).

[0039] In some embodiments, the step of obtaining an image quality score for each frame of the image according to the labeling result of the image for each frame includes: The method includes a step of comprehensively scoring the evaluation probability of each dimension of the frame image.

[0040] For example, in the above example, the overall score of the image of this frame = (Pqo + Pzo + Pgo) / 3 * 100%. The overall score is the image quality score of each frame image. It can be understood that in other embodiments, after obtaining the probability value of each dimension of each frame image being good performance, the evaluation probability of each dimension of the frame image is comprehensively scored, and can be the method of giving a weighting coefficient to each dimension and then calculating the average score or other methods, and is not limited here.

[0041] In some embodiments, the step of selecting at least one frame of an image that meets the image quality requirement comprises: The method includes a step of comparing the image quality score of each frame image with a system-preset threshold, and if the image quality score exceeds the system-preset threshold, it indicates pass, and if the image quality score is lower than the system-preset threshold, it indicates fail, and continuing to select from the frame images until at least one frame image that meets the image quality requirements is selected as the target image for at least one frame.

[0042] In some embodiments, prior to the step of performing quality labeling on the image of each frame from at least one dimension to obtain at least one labeling result, the method further comprises: inputting a number of training images covering different cases of said at least one dimension; training an intelligent scoring model with said number of training images; After the training of the intelligent scoring model is completed, the image quality scoring for each frame of the image is performed by inputting the image of each frame into the intelligent scoring model to perform image quality scoring.

[0043] In some embodiments, the process of training the intelligent scoring model is as follows. 1. Data labeling. All training images are labeled and classified by the above method, and all training images are classified to obtain training samples and test samples. Here, in the labeling of each image, Ln=(Pq, Pz, Pg), where Ln represents the label of the nth image, where Pq, Pz, and Pg represent the values ​​in the sharpness, frontality, and illumination dimensions of the image, respectively, and take the value 0 or 1, where 0 represents not sharp, not frontal, or poor illumination, and 1 represents sharp, frontal, or good illumination. For example, if an image is sharp, not frontal, and poor illumination, the image label is (1,0,0). 2. Construct a CNN (convolutional neural network) model. Construct a CNN three-classification (clear, frontal, and light-illuminated) model for facial image evaluation, set the model loss function and optimizer, and initialize the model parameters. Model training: The training samples are fed into the CNN model, and multiple iterations are run on the model parameters, learning them through the loss function and optimizer settings until the model converges. Model evaluation: The test sample is input into the CNN model obtained by training, and the classification result predicted by the model is compared with the label to evaluate the model prediction accuracy. If the accuracy is greater than a control threshold (e.g., 99%), the model training is completed; if not, the learning parameters or the number of iterations in the model training and model evaluation are adjusted and training is performed again. 3. Model accuracy definition: Suppose the model predicts all positive samples (clear, frontal, well-illuminated), the number of correctly predicted samples is m, and the total number of positive samples in the sample set is n, then the model accuracy is m / n. The reason for selecting accuracy as the model evaluation index is that we want to ensure that the error rate of the selected positive sample images is small and ensure that the selected images have high quality. 4. Model storage and deployment: The trained model is stored and deployed in the actual system.

[0044] As will be appreciated, in other embodiments, the intelligent scoring model may be trained according to other algorithms and is not limited thereto.

[0045] In some embodiments, the step of performing a skin type analysis using the at least one frame of the target image to obtain a skin type analysis result comprises: Extracting key point information in each of the target images, including at least one of nasolabial folds, dark circles under the eyes, pores, wrinkles, blackheads, and freckles; and evaluating the quality of skin quality based on key point information in the at least one frame of the target image to obtain a skin quality analysis result.

[0046] In some embodiments, the at least one frame target image includes a one frame target image and a two frame target image. When the target image is a one frame target image, a skin type analysis is performed on the target image to obtain the skin type analysis result. When the target image is a multi-frame target image, a skin type analysis is performed on each of the multi-frame target images to obtain a plurality of analysis results, and the skin type analysis result is obtained after a comprehensive analysis based on the plurality of analysis results.

[0047] In some embodiments, the method further comprises: The method further includes outputting an image quality score for the at least one frame of the target image.

[0048] In some embodiments, when the at least one frame target image is a one frame target image, outputting an image quality score of the frame target image.

[0049] In another embodiment, when the at least one frame target image is a two-frame or multi-frame target image, an overall image quality score of the two-frame or multi-frame target image may be output.

[0050] Please refer to FIG. 4, which is a schematic flow chart of a skin type analysis method in another embodiment of the present application.

[0051] Step S410: Control the imaging device 13 to obtain a video stream of the target skin location, including images of a few frames within a preset time period. Step S420: Extract images from the video stream. Step S430: Image quality scoring is performed on the extracted images. Step S440: Determine whether the image quality score is greater than the system setting threshold. If YES, proceed to step S450; if NO, proceed to step S410. Step S450: A skin quality analysis is performed based on the at least one frame of the target image to obtain a skin quality analysis result. Step S460: Output the skin quality analysis result, and output the image quality score.

[0052] The overall concept of skin quality analysis in this application is as follows: After obtaining a total video stream, extract multiple consecutive frames of images from the video stream, then use an image quality detection AI algorithm to extract face image labels from the multiple consecutive frames of photos, perform image scoring, and save and output the image quality scores; discard images that do not meet the image quality requirements; if no image that meets the quality requirements is found from the multiple frames of images, extract multiple consecutive frames of images from the video stream again, and perform image label extraction and evaluation result scoring again until an image that meets the image quality requirements is selected as the target image; then use a skin quality analysis AI algorithm to perform skin quality analysis on the target image. Thus, by combining the image quality detection AI algorithm and the skin quality analysis AI algorithm, this application can realize skin quality analysis based on high quality images, making the analysis results more stable and the measurement results from different times more comparable.

[0053] The present invention is implemented by a single processor, and includes the steps of the skin quality analysis method, namely: controlling the image capture device 13 to capture a video stream of the target skin location, including a number of frames of images within a pre-defined time period; selecting at least one frame of image that meets an image quality requirement from the several frame images of the video stream as at least one target frame of image; performing a skin quality analysis using the at least one frame of the target image to obtain a skin quality analysis result; and outputting the skin quality analysis result.

[0054] The memory 11 may also include a flash drive, a read-only memory (abbreviated as ROM), a random access memory (abbreviated as RAM), a magnetic disk or an optical disk, etc. The processor 12 may be a common processor, a digital signal processor, a dedicated integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc. The processor may implement or execute each method, step and logic block diagram disclosed in the embodiments of the present invention. The processor may be an image processor, a microprocessor, any general processor, etc. The steps of the method disclosed in the embodiments of the present invention may be directly embodied to be performed and completed by a hardware decoding processor, or performed and completed by a combination of hardware and software modules in the decoding processor. The software modules may be in a storage medium commonly used in the field, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory, for example, the processor 12 can read applications, computer instructions or data in the memory and cooperate with the hardware to complete the steps of the above-mentioned method performed by the electronic device 1.

[0055] Referring to FIG. 5, the present application provides: an acquisition control module 510 used to control the imaging device to acquire a video stream of the target skin location, including a number of frames of images within a pre-defined time period; an image selection module 520, which is used for selecting at least one frame of image that meets an image quality requirement from some frame images of the video stream, as at least one frame of target image; a skin quality analysis module 530, which is used to perform a skin quality analysis according to the at least one frame of the target image to obtain a skin quality analysis result; The present invention further provides a skin quality analysis device 500 comprising an output control module 540 adapted to output the skin quality analysis result.

[0056] In some embodiments provided by the present application, it should be understood that the disclosed skin type analysis device 500 can be realized in other forms. For example, the above-mentioned embodiments of the skin type analysis device 500 are merely exemplary, and the distinction between the units is merely a distinction between logical functions, and may be distinguished in other forms in actual implementation, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted or not implemented. Meanwhile, the shown or described mutual couplings or direct couplings or communication connections may be indirect couplings or communication connections via some interfaces, devices or units, and may be in electrical or other forms.

[0057] As can be seen from the above, the present invention has the excellent properties as described above, and when used, it increases effects not found in the prior art, has practicality, and becomes a product of extremely practical value.

[0058] The above are only preferred embodiments of the present application, and are not intended to limit the present application. Any modifications, equivalent replacements or improvements made without departing from the concept and principles of the present application should be included in the scope of protection of the present application. [Explanation of symbols]

[0059] 1 Electronic equipment 11. Memory 12 processors 13 Imaging device 500 Skin quality analyzer 510 Acquisition Control Module 520 Image Selection Module 530 Skin Quality Analysis Module 540 Output Control Module

Claims

1. A skin quality analysis method for use in an electronic device, comprising: controlling an image capture device to capture a video stream of the target skin location, the video stream including a number of frames of images within a pre-defined time period; selecting at least one frame of an image that meets an image quality requirement from the several frame images of the video stream as at least one target frame of an image; performing a skin quality analysis using the at least one frame of the target image to obtain a skin quality analysis result; and outputting said skin quality analysis result.

2. The step of selecting at least one frame of an image that meets an image quality requirement from some frame images of the video stream and setting the at least one frame of an image as a target image includes: extracting a number of frames of images from the video stream; scoring each of the plurality of frames of images to obtain an image quality score for each frame of image; The skin quality analysis method according to claim 1, further comprising a step of selecting at least one frame of image that meets an image quality requirement from the plurality of frames of images based on the image quality score of each frame of image, and setting the at least one frame of image as a target image.

3. The step of selecting at least one frame of image that meets an image quality requirement from the plurality of frames of images according to the image quality score of each frame of image and setting the at least one frame of image as a target image includes: A step of selecting one frame having the highest image quality score or at least two frames having the relatively highest image quality scores from the images of the plurality of frames according to the image quality scores of the images of each frame, and setting the selected frame as a target image of the at least one frame; or The skin quality analysis method according to claim 2, further comprising a step of randomly selecting at least one frame of image that meets an image quality requirement from the plurality of frames of images based on the image quality score of each frame of image, and setting the selected frame of image as a target image for the at least one frame.

4. The step of selecting at least one frame of an image that meets an image quality requirement from some frame images of the video stream and setting the at least one frame of an image as a target image includes: extracting a number of frames of images from the video stream; a step of scoring the images of the plurality of frames in order starting from a first frame image, scoring the images of each frame to obtain an image quality score, and then determining whether or not the image quality score of the image of this frame satisfies an image quality requirement, and if the image quality score of the image of this frame satisfies the image quality requirement, determining that the image of this frame is to be a target image; The skin quality analysis method of claim 1, further comprising a step of scoring a next frame image among the plurality of frames of images if the image quality score of the image of this frame does not satisfy the image quality requirement, until a target image whose image quality score satisfies the image quality requirement is determined.

5. The skin quality analysis method according to any one of claims 2 to 4, further comprising the step of, when a target image satisfying the image quality requirement is not found from the images of the plurality of frames, re-extracting a next group of images of multiple frames different from the images of the plurality of frames from the video stream, and re-performing image quality scoring of the images of multiple frames of the next group until a target image satisfying the image quality requirement is selected.

6. The method for analyzing skin quality according to any one of claims 2 to 4, further comprising the step of outputting an image quality score of said at least one target image.

7. The step of performing a skin type analysis using the target image to obtain a skin type analysis result includes: When the target image is a one-frame target image, performing a skin type analysis on the target image to obtain the skin type analysis result; or 2. The skin quality analysis method according to claim 1, further comprising the steps of: when the target image is a target image of a plurality of frames, performing a skin quality analysis on each of the target images of the plurality of frames to obtain a plurality of analysis results; and performing a comprehensive analysis using the plurality of analysis results to obtain the skin quality analysis result.

8. The step of performing image quality scoring on the image of each frame includes: performing quality labeling on each frame image from at least one dimension including at least one of sharpness, light intensity, and frontality to obtain at least one labeling result; The skin quality analysis method according to any one of claims 2 to 4 and 7, further comprising a step of obtaining an image quality score for each frame image based on the labeling result of the image for each frame.

9. Prior to the step of performing quality labeling on the image of each frame from at least one dimension to obtain at least one labeling result, inputting a number of training images covering different cases of said at least one dimension; training an intelligent scoring model with said number of training images; The skin quality analysis method of claim 8, further comprising a step of: after the training of the intelligent scoring model is completed, inputting the image of each frame into the intelligent scoring model to perform image quality scoring for the image of each frame.

10. The step of performing a skin quality analysis using the at least one frame of the target image to obtain a skin quality analysis result includes: Extracting key point information in a target image of each frame, including at least one of nasolabial folds, dark circles under the eyes, pores, wrinkles, blackheads, and freckles; 2. The skin quality analysis method according to claim 1, further comprising the step of evaluating the quality of the skin quality based on key point information in the at least one frame of the target image to obtain the skin quality analysis result.

11. An electronic device comprising a processor, a memory, and an imaging device, the imaging device and the memory being electrically connected to the processor, a computer program being stored in the memory, and the processor executing the computer program to perform steps of the skin quality analysis method according to any one of claims 1 to 4, 7, and 10.

12. A computer-readable storage medium storing a computer program that, when executed by a single processor, can execute the steps of the skin quality analysis method according to any one of claims 1 to 4, 7, and 10.

13. an acquisition control module adapted to control the imaging device to acquire a video stream of the target skin location, the video stream including a number of frames of images within a predefined time period; an image selection module, which is used for selecting at least one frame of an image that meets an image quality requirement from some frame images of the video stream, as at least one frame of a target image; a skin quality analysis module used for performing a skin quality analysis using the at least one frame of the target image to obtain a skin quality analysis result; and an output control module used to output the skin quality analysis result.

Citation Information

Patent Citations

  • Chip, acceleration card, electronic equipment and data processing method

    CN114298295A

  • Imaging guidance device, imaging guidance method, and program

    JP2023053733A

  • Image processing apparatus, image processing method, and program

    WO2016199483A1