Scalp abnormity dynamic detection method, device and equipment and storage medium

Abstract

The invention discloses a scalp abnormity dynamic detection method, device and equipment and a storage medium, and the method comprises the steps: collecting an image sequence of head massage equipment when the head massage equipment moves on a head, and obtaining the posture change data of the head massage equipment; calculating the pixel offset of adjacent image frames of the image sequence by using the attitude change data; performing position correction on a detection target of each frame of image in the image sequence according to the pixel offset; and performing anomaly analysis on the detection target in the image sequence, marking an abnormal region based on an analysis result, and generating a detection report. The problem that the image recognition accuracy is insufficient when the head massage equipment rapidly moves is solved.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and particularly to a method, device, equipment and storage medium for dynamically detecting scalp abnormalities. Background Art

[0002] With the continuous development of scalp health monitoring technology, as a convenient detection tool, the head massage device is widely used in the daily monitoring of scalp conditions. Common head massage device technologies usually combine image acquisition and recognition technologies, which can continuously acquire scalp images when the head massage device moves, and analyze the images through image processing algorithms to identify the health status of the scalp.

[0003] However, in actual applications, when the head massage device moves rapidly on the head, due to problems such as jitter and blurring in the acquired image sequence, the accuracy of image recognition is insufficient. In addition, the minute changes in the scalp are difficult to be accurately captured during rapid movement, further increasing the difficulty of recognition. These problems limit the accuracy of the existing head massage device technology in detecting scalp conditions during rapid movement and cannot meet the user's requirements for high-precision and real-time scalp health monitoring. Summary of the Invention

[0004] The main purpose of this application is to provide a method, device, equipment and storage medium for dynamically detecting scalp abnormalities, aiming to solve the technical problem of insufficient recognition accuracy in detecting scalp conditions when the head massage device moves rapidly.

[0005] To achieve the above purpose, this application provides a method for dynamically detecting scalp abnormalities, and the method for dynamically detecting scalp abnormalities includes the following steps:

[0006] While collecting the image sequence when the head massage device moves on the head, obtain the attitude change data of the head massage device;

[0007] Use the attitude change data to calculate the pixel offset of adjacent image frames of the image sequence;

[0008] According to the pixel offset, correct the position of the detection target in each frame of the image sequence;

[0009] Perform abnormal analysis on the detection target in the image sequence, mark the abnormal area based on the analysis result and generate a detection report.

[0010] In one embodiment, the step of using the attitude change data to calculate the pixel offset of adjacent image frames of the image sequence includes:

[0011] Calculate the attitude matrix of the head massage device when collecting each frame of the image according to the attitude change data;

[0012] By extracting the transformation information in the pose matrices of adjacent image frames, a translation vector between adjacent image frames is obtained;

[0013] The translation vector is converted into a pixel offset in the image coordinate system.

[0014] In one embodiment, the step of correcting the position of the detection target in each frame of the image sequence according to the pixel offset includes:

[0015] Using image processing techniques to identify the detection target in each frame of the image sequence;

[0016] For each frame of the image sequence, calculate the cumulative offset from the first frame to the current frame according to the pixel offset;

[0017] Using the cumulative offset, perform a geometric transformation on the detection target in each frame of the image, so as to move the detection target to the correct position.

[0018] In one embodiment, after the step of correcting the position of the detection target in each frame of the image sequence according to the pixel offset includes:

[0019] Calculate the error data between the corrected position and the initial position of the detection target in each frame of the image sequence;

[0020] According to the error data, evaluate the correction effect of the detection target in the image sequence.

[0021] In one embodiment, the step of performing anomaly analysis on the detection target in the image sequence, marking the anomaly area based on the analysis result and generating a detection report includes:

[0022] Obtain the statistical features of the detection target by analyzing multiple frames of images;

[0023] Compare the statistical features of the detection target with a preset threshold to determine whether there is an anomaly;

[0024] If an anomaly is detected, mark the anomaly area on the image and generate a detection report based on the detection result.

[0025] In one embodiment, the step of performing anomaly analysis on the detection target in the image sequence further includes:

[0026] Obtain the anomaly recognition result of the detection target in each frame of the image sequence;

[0027] Perform time series analysis on the abnormal recognition results of the image sequence to detect the change trend of abnormal conditions;

[0028] Analyze the consistency of the abnormal recognition results in the same area of different frame images of the image sequence to obtain the spatial consistency analysis result;

[0029] Combine the time series analysis result and the spatial consistency analysis result to conduct a comprehensive analysis of abnormal conditions.

[0030] In one embodiment, the step of acquiring the attitude change data of the head massage device while acquiring the image sequence of the head massage device moving on the head includes:

[0031] When the head massage device moves on the head, continuously acquire scalp images at a preset frequency to obtain an image sequence, and acquire the attitude change data of the head massage device at the same frequency as that of acquiring scalp images.

[0032] In addition, to achieve the above object, the present application further provides a scalp abnormal dynamic detection device, and the scalp abnormal dynamic detection device includes:

[0033] An acquisition module, configured to acquire the attitude change data of the head massage device while acquiring the image sequence of the head massage device moving on the head;

[0034] A calculation module, configured to calculate the pixel offset of adjacent image frames of the image sequence by using the attitude change data;

[0035] A correction module, configured to perform position correction on the detection target of each frame image in the image sequence according to the pixel offset;

[0036] An analysis module, configured to perform abnormal analysis on the detection target in the image sequence, mark the abnormal area based on the analysis result, and generate a detection report.

[0037] In addition, to achieve the above object, the present application further provides a terminal device, and the terminal device includes a memory, a processor, and a scalp abnormal dynamic detection program stored on the memory and executable on the processor. When the scalp abnormal dynamic detection program is executed by the processor, the steps of the scalp abnormal dynamic detection method as described above are implemented.

[0038] In addition, to achieve the above object, the present application further provides a computer-readable storage medium, and a scalp abnormal dynamic detection program is stored on the computer-readable storage medium. When the scalp abnormal dynamic detection program is executed by a processor, the steps of the scalp abnormal dynamic detection method as described above are implemented.

[0039] One or more technical solutions proposed by this application have at least the following technical effects:

[0040] In this application, while collecting the image sequence when the head massage device moves on the head, the attitude change data of the head massage device is obtained; using the attitude change data, the pixel offset of adjacent image frames of the image sequence is calculated; according to the pixel offset, the position of the detection target in each frame of the image sequence is corrected; the detection target in the image sequence is analyzed for anomalies, and the abnormal area is marked based on the analysis result and a detection report is generated. By these technical means, the problem of insufficient accuracy in image recognition when the head massage device moves rapidly is solved.

[0041] Compared with the prior art, by real-time monitoring of the attitude change of the head massage device, this application can capture the minute changes of the scalp more accurately, thereby improving the accuracy and real-time performance of scalp health monitoring. Specifically, through the real-time analysis of the attitude change data and the calculation of the pixel offset, the influence of image jitter and blur on the detection result can be reduced, and the accuracy of the detection target position correction is improved. In addition, correcting the position of the detection target in each frame of the image according to the pixel offset solves the difficulties in tracking and analysis caused by the change of the target position in the continuous image sequence. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a schematic flowchart of the first exemplary embodiment of the scalp anomaly dynamic detection method of this application;

[0043] Figure 2 It is a schematic flowchart of the second exemplary embodiment of the scalp anomaly dynamic detection method of this application;

[0044] Figure 3 It is a schematic diagram of the module structure of the scalp anomaly dynamic detection device in the embodiment of this application;

[0045] Figure 4 It is a schematic diagram of the device structure of the hardware operating environment involved in the scalp anomaly dynamic detection method in the embodiment of this application.

[0046] The implementation, functional features and advantages of the purpose of this application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] It should be understood that the specific embodiments described herein are only used to explain this application and are not used to limit this application.

[0048] The main technical solution of this application is as follows: while collecting the image sequence when the head massage device moves on the head, obtain the attitude change data of the head massage device; use the attitude change data to calculate the pixel offset of adjacent image frames of the image sequence; according to the pixel offset, correct the position of the detection target in each frame of the image sequence;

[0049] Perform anomaly analysis on the detection target in the image sequence, mark the abnormal area based on the analysis result, and generate a detection report.

[0050] This application actually takes into account that although common scalp health monitoring technologies can combine image acquisition and recognition technologies, during the rapid movement of the head massage device, due to image jitter and blur, as well as the difficulty in capturing minute scalp changes, the recognition accuracy is limited. In addition, the change in the target position in the continuous image sequence also brings difficulties to tracking and analysis.

[0051] Based on this, the embodiments of this application propose a solution: when the head massage device moves on the head, continuously collect scalp images at a preset frequency to obtain an image sequence, obtain the attitude change data of the head massage device at the same frequency as the collection of scalp images, use the attitude change data to calculate the pixel offset of adjacent image frames of the image sequence, according to the pixel offset, correct the position of the detection target in each frame of the image sequence, obtain the statistical features of the detection target by analyzing multiple frames of images, compare the statistical features with a preset threshold, if an anomaly is detected, mark the abnormal area on the image, and generate a detection report based on the detection result. It is also possible to perform a comprehensive analysis of the abnormal situation by combining the time series analysis result and the spatial consistency analysis result.

[0052] Specifically, the following are the detailed steps of the first exemplary embodiment of the scalp anomaly dynamic detection method of this application:

[0053] See Figure 1 , Figure 1 which is a schematic flowchart of the first exemplary embodiment of the scalp anomaly dynamic detection method of this application. In this embodiment, the scalp anomaly dynamic detection method includes steps S10 to S40:

[0054] Step S10, while collecting the image sequence when the head massage device moves on the head, obtain the attitude change data of the head massage device;

[0055] In a feasible implementation manner, step S10 may include step S11:

[0056] Step S11: When the head massage device moves on the head, continuously collect scalp images at a preset frequency to obtain an image sequence, and acquire the attitude change data of the head massage device at the same frequency as that of collecting the scalp images.

[0057] Specifically, when the head massage device is placed on the head and starts to move, continuously collect images of the scalp at a preset frequency. These images are arranged in chronological order to form a series of image sequences. The preset frequency is set according to actual needs and the requirements of scalp health monitoring to ensure that sufficient quantity and quality of image data can be collected. Meanwhile, acquire the attitude change data of the head massage device at the same frequency as that of collecting the scalp images. These data include parameters such as the position, angle, and speed of the head massage device during movement, which can reflect the specific state of the head massage device when collecting each frame of image.

[0058] During the collection process, ensure the synchronization of the image sequence and the attitude change data, that is, each frame of image corresponds to a set of attitude change data. In this way, these data can be used to correct the position of each frame of image in the image sequence, thereby reducing or eliminating the image jitter and blurring problems caused by the movement of the head massage device.

[0059] Step S20: Use the attitude change data to calculate the pixel offset of adjacent image frames of the image sequence;

[0060] In a feasible implementation manner, step S20 may include steps S21 - S23:

[0061] Step S21: Calculate the attitude matrix of the head massage device when collecting each frame of image according to the attitude change data;

[0062] Specifically, use these attitude change data and adopt a rotation matrix or quaternion to describe the rotation of the head massage device relative to a fixed reference coordinate system (such as the initial position of the head massage device). The rotation matrix is a 3x3 matrix that can describe the rotation transformation in three-dimensional space, while the quaternion is a rotation representation method that avoids the gimbal lock problem and represents rotation with a scalar and three vector components.

[0063] For the rotation matrix R, if the rotation angles of the head massage device relative to the three axes of the reference coordinate system (denoted as α, β, γ) are known, then the rotation matrix can be expressed as:

[0064] R = R x (α)R y (β)R z (γ)

[0065] Wherein, R x (α), R y(β), R z (γ) are rotation matrices about the x-axis, y-axis, and z-axis respectively, and the calculation formulas are as follows:

[0066]

[0067]

[0068] By multiplying these rotation matrices, the pose matrix R of the head massage device at any moment can be obtained. This matrix not only contains rotation information but also can include translation information by being extended to a 4x4 matrix, thus completely describing the position and orientation of the head massage device in space.

[0069] Step S22, by extracting the transformation information in the pose matrices of adjacent image frames, obtain the translation vector between adjacent image frames;

[0070] Specifically, in the above steps, the pose matrix of the head massage device at the time of each frame of image acquisition is obtained. These matrices not only contain rotation information but also implicitly contain translation information because the position of the head massage device relative to the reference coordinate system changes during movement. To extract the translation vector from the pose matrix, the displacement of the head massage device between two consecutive frames of images needs to be considered.

[0071] Suppose there are two consecutive pose matrices R and R2, corresponding to the poses of the first and second frames of images respectively, and two corresponding translation vectors T1 and T2. The translation vector can be expressed as:

[0072]

[0073] where t x , t y , t z represent the translation distances along the x-axis, y-axis, and z-axis respectively;

[0074] To obtain the translation vector ΔT between adjacent image frames, the difference between the two translation vectors can be calculated, and the calculation formula is as follows:

[0075]

[0076] The difference vector ΔT represents the actual movement distance and direction of the head massage device from the first frame to the second frame of image acquisition. In practical applications, the movement of the camera (or head massage device) in space can also be accurately estimated through visual odometry (VO) or simultaneous localization and mapping (SLAM) technology.

[0077] By calculating the translation vectors between all adjacent frames, a complete movement trajectory can be constructed, which is crucial for subsequent pixel offset calculation and image correction.

[0078] Step S23: Convert the translation vector into a pixel offset in the image coordinate system.

[0079] Specifically, first obtain the relationship between the image coordinate system and the world coordinate system. The image coordinate system usually takes the upper left corner of the image as the origin, with the positive directions of the x-axis and y-axis being to the right and down respectively. The world coordinate system is a coordinate system in three-dimensional space, usually taking the initial position of the head massage device as the origin. The translation vector is represented in the world coordinate system and needs to be converted into a pixel offset in the image coordinate system.

[0080] To perform the conversion, first obtain the internal parameters of the camera, including the focal lengths fx and fy corresponding to the x-axis and y-axis respectively, and the center point coordinates (c x , c y ) of the image. These parameters can be obtained through the camera calibration process.

[0081] Assume the translation vector is where Δt x , Δt y , Δt z represent the translation distances along the x-axis, y-axis, and z-axis respectively. Then, the corresponding pixel offset ΔP can be expressed as:

[0082]

[0083] where Δp x and Δp y represent the pixel offsets in the x-axis and y-axis directions of the image respectively. The conversion formula is as follows:

[0084]

[0085] Here, assume that Δt z is not zero, that is, the head massage device moves in the z-axis direction. If Δt z is zero, it means that the head massage device does not move in the z-axis direction. At this time, the pixel offset will be directly proportional to the translation distances in the x-axis and y-axis.

[0086] Through the above formula, the translation vector of each frame of the image can be converted into a pixel offset. These pixel offsets will be used to correct the positions of the detection targets in each frame of the image.

[0087] Step S30: Correct the positions of the detection targets in each frame of the image sequence according to the pixel offset;

[0088] In a feasible implementation manner, step S30 may include steps S31 to S33:

[0089] Step S31: Use image processing technology to identify the detection target in each frame of the image sequence;

[0090] Specifically, it is first necessary to preprocess the image sequence to improve the accuracy of subsequent recognition. The preprocessing steps may include denoising, enhancing contrast, adjusting brightness, etc., to eliminate interference factors in the image and make the features of the detection target more obvious. The preprocessed image will provide clearer image data for feature extraction and recognition of the detection target.

[0091] Next, use image recognition algorithms to identify the detection target in each frame of the image. It involves a variety of image processing technologies, such as edge detection, feature point matching, pattern recognition, etc. For example, the edge detection algorithm (such as the Canny edge detector) can be used to find the edge information in the image, and these edge information helps to determine the contour of the detection target. At the same time, feature point matching technologies (such as SIFT, SURF or ORB) can be used to identify specific points or patterns in the image, and these points or patterns are unique and matchable in different image frames.

[0092] For example, for each frame of the image, the recognition process may be as follows: extract key feature points and descriptors from the preprocessed image; match the extracted features with the known scalp abnormality feature database to determine the detection target; according to the matching result, locate the position of the detection target in the image.

[0093] Step S32: For each frame of the image sequence, calculate the cumulative offset from the first frame to the current frame according to the pixel offset;

[0094] Specifically, accumulate the offsets to obtain the total offset from the first frame to the current frame.

[0095] Suppose there are n frames of images, and the pixel offset of the i-th frame of the image is where Δp ix and Δp iy represent the pixel offsets in the x-axis and y-axis directions respectively. Then, the cumulative offset C i from the first frame to the i-th frame can be expressed as:

[0096]

[0097] This cumulative offset represents the total moving distance of the detection target from the first frame to the i-th frame. By calculating the cumulative offset of each frame, the moving trajectory of the detection target in the entire image sequence can be obtained.

[0098] This calculation process can be implemented through programming. For example, a loop can be used to iterate through each frame of the image, and the pixel offset of each frame is added to an accumulative variable. This accumulative variable is initialized to zero at the beginning of the loop and then updated during the calculation of each frame.

[0099] Step S33: Using the cumulative offset, perform geometric transformation on the detection target in each frame of the image, so as to move the detection target to the correct position.

[0100] Specifically, before performing geometric transformation, it is first necessary to determine the initial position of the detection target in each frame of the image. This can be done by manual marking or automatic detection in the first frame of the image. After obtaining the initial position, the cumulative offset mentioned above can be used to update the position of the detection target in each frame.

[0101] Geometric transformation usually includes operations such as translation, rotation, and scaling. In this step, the focus is mainly on translation transformation, that is, directly adjusting the position of the detection target according to the cumulative offset. For each frame of the image, the translation transformation can be expressed by the following formula:

[0102] P′ = P + C

[0103] Where P is the position vector of the detection target in the original image, P′ is the transformed position vector, and C is the cumulative offset vector from the first frame to the current frame. The position vectors P and P′ can be expressed as:

[0104]

[0105] According to the above formula, the transformed position P′ can be calculated by adding the cumulative offset C to the initial position P:

[0106] x′ = x + Δx

[0107] y′ = y + Δy

[0108] This process needs to be repeated for each frame of the image sequence to ensure that the detection targets in all frames are corrected to the correct positions.

[0109] Step S40: Perform anomaly analysis on the detection targets in the image sequence, mark the abnormal areas based on the analysis results, and generate a detection report.

[0110] In a feasible implementation, step S40 may include steps A1 to A3:

[0111] Step A1: Obtain the statistical features of the detection target by analyzing multiple frames of images;

[0112] Specifically, relevant features of the detection target are first extracted from each frame of the image. These features may include color histograms, texture features, shape descriptors, edge information, etc. For example, color histograms can be used to describe the color distribution of the detection target, local binary patterns (LBP) can be used to describe its texture features, or contour tracking algorithms can be used to extract its shape information.

[0113] After the features are extracted, statistical analysis is performed on these features to obtain the statistical features of the detection target. The mean, variance, maximum, and minimum values of the features can be calculated. For example, for a color histogram, the mean and standard deviation of each color channel can be calculated; for texture features, the mean and variance of the LBP features can be calculated.

[0114] Based on these statistical features, quantitative analysis of the changes of the detection target in a series of images can be carried out. These statistical data can not only reflect the stability of the detection target but also reveal its possible abnormal changes. For example, if the color histogram of the detection target shows significant variability in a series of images, this may indicate an abnormal situation.

[0115] Step A2: Compare the statistical features of the detection target with a preset threshold to determine whether there is an abnormality.

[0116] Specifically, the thresholds for judging abnormalities are first determined. These thresholds can be set based on the statistical features of normal scalps. For example, a large amount of image data of normal scalps can be collected, the mean and standard deviation of their statistical features can be calculated, and then the thresholds can be determined based on these data. For example, the mean plus or minus two standard deviations is used as the threshold.

[0117] Next, the statistical features of the detection target obtained in the above steps are compared with these thresholds. If the statistical features exceed the threshold range, it is considered that the detection target has an abnormality. For example, if the mean of the color histogram of the detection target exceeds the normal range, this may indicate inflammation or other problems on the scalp.

[0118] Step A3: If an abnormality is detected, mark the abnormal area on the image and generate a detection report based on the detection result.

[0119] Specifically, when it is determined that the detection target has an abnormality, these abnormal areas are clearly marked on the image. This usually involves drawing bounding boxes, highlighting, or using color coding on the image to identify the abnormal areas. For example, a red mark can be used to indicate the inflammation area, or different color labels can be used to distinguish different types of abnormalities.

[0120] The marking process can be achieved through image processing techniques, such as drawing graphics or text on the image. If color coding is used, a color mapping table can be defined to associate different types of abnormalities with specific colors. For example, it can be defined that:

[0121] Red: Inflammation; Yellow: Injury; Blue: Other abnormalities.

[0122] After marking the abnormal areas, a detection report is generated based on the detection results. The detection report will include a description of the abnormal areas, the nature of the abnormalities, possible causes, and recommended follow-up actions. For example, if the detected abnormality is scalp inflammation, the report may indicate the location, extent, and severity of the inflammation and recommend that the user consult a doctor or undergo further examinations.

[0123] In another feasible implementation, step S40 may further include steps B1 to B4:

[0124] Step B1, obtaining the abnormal recognition results of the detection target for each frame of the image sequence;

[0125] Specifically, an abnormal detection algorithm is applied to each frame of the image, which may include machine learning models, deep learning networks, or other image analysis techniques. These techniques can identify abnormal patterns in the image and generate an abnormal recognition result for each frame of the image, usually a binary label (abnormal / normal) or a probability value indicating the likelihood of the presence of an abnormality.

[0126] The acquisition of the abnormal recognition results can be represented by the following formula:

[0127] Let A i be the abnormal recognition result of the i-th frame of the image, where A i can be a binary variable, taking a value of 1 to indicate abnormality and a value of 0 to indicate normality. The formula for A i can be expressed as:

[0128] A i = f(I i )

[0129] where I i is the i-th frame of the image, and f is the abnormal detection function that takes the image as input and outputs the abnormal recognition result.

[0130] After obtaining the abnormal recognition results for each frame of the image, these results are stored in a time series for subsequent time series analysis. This time series will contain the abnormal status corresponding to each frame of the image sequence, providing a complete view of the abnormal conditions of the detection target over time.

[0131] Step B2: Perform time series analysis on the anomaly recognition results of the image sequence to detect the changing trend of the abnormal conditions;

[0132] Specifically, first preprocess the time series data obtained in the above step. After preprocessing, use time series analysis techniques such as autoregressive model (AR), moving average model (MA), autoregressive moving average model (ARMA), or autoregressive integrated moving average model (ARIMA) to model the time series of the anomaly recognition results.

[0133] The time series analysis can be represented by the following formula:

[0134] Let T t be the value of the time series at time point t, be the predicted value, and the model parameters be θ. Then the time series model can be expressed as:

[0135]

[0136] where g is the time series model function, and p is the order of the model, representing the number of past time points used to predict the value at the current time point.

[0137] Identify the changing trend of the abnormal conditions through time series analysis, such as the rising or falling of the trend, as well as the frequency and duration of the occurrence of anomalies. This information is very valuable for evaluating the scalp health condition and formulating corresponding treatment plans. For example, if the time series analysis shows that the abnormal conditions increase over time, it may indicate that more aggressive intervention measures are needed. On the contrary, if the abnormal conditions decrease over time, it may indicate that the current treatment plan is effective.

[0138] Step B3: Analyze the consistency of the anomaly recognition results in the same region of different frame images of the image sequence to obtain the spatial consistency analysis result;

[0139] Specifically, determine the positions of the abnormal regions in each frame image of the image sequence. This can be achieved through the anomaly recognition results obtained in Step B1, where the abnormal regions of each frame image have been marked. Then perform spatial consistency analysis on these abnormal regions to determine whether they remain consistent at different time points.

[0140] The spatial consistency analysis can be represented by the following formula:

[0141] Let A i,j be the anomaly recognition result of the j-th abnormal region in the i-th frame image, where A i,j can be a binary variable, taking the value of 1 indicating anomaly and 0 indicating normal. Let C j be the spatial consistency score of the j-th abnormal region, and the calculation formula is:

[0142]

[0143] Among them, n is the number of frames in the image sequence.

[0144] Spatial consistency score C j represents the frequency at which the j-th abnormal region is recognized as abnormal in all frame images. If C j is close to 1, it indicates that the abnormal region is recognized as abnormal in most frame images, suggesting that it has high spatial consistency. On the contrary, if C j is close to 0, it indicates that the abnormal region is recognized as normal in most frame images, suggesting that its spatial consistency is low.

[0145] Step B4, combining the time series analysis results and the spatial consistency analysis results, conduct a comprehensive analysis of the abnormal situation.

[0146] Integrate the time series analysis results with the spatial consistency analysis results. The time series analysis results provide trend information on the change of the abnormal situation over time, while the spatial consistency analysis results provide consistency information on the abnormal region at different time points. By combining these two aspects of information, the characteristics of the abnormal situation can be understood more comprehensively.

[0147] The comprehensive analysis can be achieved in the following ways:

[0148] Match the abnormal trends identified in the time series analysis with the abnormal regions identified in the spatial consistency analysis. For example, if the time series analysis shows that the abnormal situation increases over time, and the spatial consistency analysis shows that a certain region has high consistency, it can be inferred that this region may be the source or key area of the abnormality.

[0149] Determine the priority of the abnormal regions according to the results of the time series analysis and the spatial consistency analysis. For example, those regions that persist over time and are highly consistent spatially may need to be given priority attention and treatment.

[0150] Classify the abnormal situation according to the results of the comprehensive analysis. For example, classify the abnormal situation into acute, chronic, local, extensive, etc. types to facilitate the adoption of targeted treatment measures.

[0151] The results of the comprehensive analysis will be used to generate a final detection report, which will include a comprehensive description of the abnormal situation, possible causes, recommended follow-up actions, etc. For example, if the comprehensive analysis shows that there is chronic inflammation on the scalp, the report may recommend that the user consult a doctor and undergo long-term treatment.

[0152] Furthermore, referring to Figure 2 , Figure 2This is a schematic flowchart of the second exemplary embodiment of the scalp abnormality dynamic detection method of the present application. In the second exemplary embodiment of the present application, step S30: The step of correcting the position of the detection target in each frame of the image sequence according to the pixel offset further includes steps S51 to S52:

[0153] Step S51, calculate the error data between the corrected position and the initial position of the detection target in each frame of the image sequence;

[0154] Specifically, first obtain the initial position, the position before correction, and the position after correction of the detection target in each frame of the image. The initial position refers to the true position of the detection target in the image without any movement or transformation. The position before correction refers to the offset position of the detection target caused by the movement of the head massage device. The position after correction refers to the predicted position of the detection target after applying the correction algorithm.

[0155] The calculation of the error data can be expressed by the following formula:

[0156] Let P original，i be the initial position of the detection target in the i-th frame of the image, P before，i be the position before correction, and P after，i be the position after correction. Then the error vector E i of the i-th frame of the image can be expressed as:

[0157] E i =P after,i -P original,i

[0158] The magnitude of the error can be calculated by the norm of the error vector, such as the Euclidean norm:

[0159]

[0160] where E ix and E iy are the components of the error vector in the x-axis and y-axis directions respectively.

[0161] By calculating the error data of each frame of the image, an error sequence can be obtained, which reflects the performance of the correction algorithm in the entire image sequence.

[0162] Step S52, evaluate the correction effect of the detection target in the image sequence according to the error data.

[0163] Specifically, perform statistical analysis on the error data. Let E mean be the mean value of the errors of all frames of the image, E std be the standard deviation. If E mean is close to zero and E stdis small, indicating that the calibration algorithm can accurately calibrate the detection target to its initial position, and the calibration effect is good. On the contrary, if E mean is large or E std is large, it indicates that there are biases or inconsistencies in the calibration algorithm, and further adjustment and optimization may be required.

[0164] In addition, the present application also provides a device for dynamically detecting scalp abnormalities, and the device for dynamically detecting scalp abnormalities includes:

[0165] An acquisition module 10, configured to acquire the attitude change data of the head massage device while collecting an image sequence of the head massage device moving on the head;

[0166] A calculation module 20, configured to calculate the pixel offset between adjacent image frames of the image sequence by using the attitude change data;

[0167] A calibration module 30, configured to perform position calibration on the detection target of each frame of the image sequence according to the pixel offset;

[0168] An analysis module 40, configured to perform abnormality analysis on the detection target in the image sequence, mark the abnormal area based on the analysis result, and generate a detection report.

[0169] The device for dynamically detecting scalp abnormalities provided by the present application adopts the method for dynamically detecting scalp abnormalities in the above embodiment, aiming to solve the technical problem of insufficient recognition accuracy in detecting the scalp condition when the head massage device moves rapidly. Compared with the prior art, the beneficial effects of the device for dynamically detecting scalp abnormalities provided by the present application are the same as those of the method for dynamically detecting scalp abnormalities provided by the above embodiment, and other technical features in the device for dynamically detecting scalp abnormalities are the same as those disclosed in the method of the above embodiment, and will not be elaborated here.

[0170] The present application provides a device for dynamically detecting scalp abnormalities, and the device for dynamically detecting scalp abnormalities includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the method for dynamically detecting scalp abnormalities in the first embodiment above.

[0171] The scalp abnormal dynamic detection device in the embodiments of the present application may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description: tablet computers), PMPs (Portable Media Player), vehicle-mounted terminals (such as vehicle-mounted navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. Figure 4 The scalp abnormal dynamic detection device shown is merely an example and should not impose any limitations on the functions and usage scope of the embodiments of the present application.

[0172] As Figure 4 shown, the scalp abnormal dynamic detection device may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM: Read Only Memory) 1002 or the program loaded from the storage device 1003 into the random access memory (RAM: Random Access Memory) 1004. In the RAM 1004, various programs and data required for the operation of the scalp abnormal dynamic detection device are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems may be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 can allow the scalp abnormal dynamic detection device to communicate with other devices wirelessly or wiredly to exchange data. Although the scalp abnormal dynamic detection device with various systems is shown in the figure, it should be understood that it is not required to implement or have all the shown systems. More or fewer systems may be alternatively implemented or had.

[0173] In particular, according to the embodiments disclosed in the present application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed in the present application include a computer program product that includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by a processing device 1001, the above-mentioned functions defined in the methods of the embodiments disclosed in the present application are executed.

[0174] The scalp abnormality dynamic detection device provided by the present application adopts the scalp abnormality dynamic detection method in the above-mentioned embodiment, aiming to solve the technical problem of insufficient recognition accuracy in detecting the scalp condition when the head massage device moves rapidly. Compared with the prior art, the beneficial effects of the scalp abnormality dynamic detection device provided by the present application are the same as those of the scalp abnormality dynamic detection method provided by the above-mentioned embodiment, and other technical features in the scalp abnormality dynamic detection device are the same as the features disclosed in the method of the previous embodiment, which will not be elaborated here.

[0175] It should be understood that the various parts disclosed in the present application can be implemented by hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0176] As described above, the above are only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, and all of them should be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0177] The present application provides a computer-readable storage medium having computer-readable program instructions (i.e., computer programs) stored thereon, and the computer-readable program instructions are used to execute the scalp abnormality dynamic detection method in the above-mentioned embodiment.

[0178] The computer-readable storage medium provided by this application can be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems or devices, or any combination of the above. More specific examples of computer-readable storage media can include, but are not limited to: electrical connections with one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or flash memory, optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this embodiment, the computer-readable storage medium can be any tangible medium that contains or stores a program, and this program can be used by or in combination with an instruction execution system or device. The program code contained on the computer-readable storage medium can be transmitted by any appropriate medium, including but not limited to: wires, optical cables, RF (radio frequency), etc., or any suitable combination of the above.

[0179] The above computer-readable storage medium can be included in the scalp abnormality dynamic detection device; or it can exist independently without being assembled into the scalp abnormality dynamic detection device.

[0180] The computer program code for performing the operations of this application can be written in one or more programming languages or combinations thereof. The above programming languages include object-oriented programming languages - such as Java, Smalltalk, C++, and also include conventional procedural programming languages - such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, executed as an independent software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer through any type of network - including a local area network (LAN) or a wide area network (WAN) - or, it can be connected to an external computer (for example, by using an Internet service provider to connect through the Internet).

[0181] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a segment of a program, or a part of code that contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, and combinations of blocks in the block diagram and / or flowchart, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0182] The modules described in the embodiments of the present application can be implemented in software or in hardware. In some cases, the name of the module does not constitute a limitation on the unit itself.

[0183] The readable storage medium provided by the present application is a computer-readable storage medium that stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned scalp abnormality dynamic detection method, aiming to solve the technical problem of insufficient recognition accuracy in detecting scalp conditions when the head massage device is moving rapidly. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided by the present application are the same as those of the scalp abnormality dynamic detection method provided in the above embodiments, and will not be elaborated here.

[0184] The present application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the scalp abnormality dynamic detection method as described above.

[0185] The computer program product provided by the present application aims to solve the technical problem of insufficient recognition accuracy in detecting scalp conditions when the head massage device is moving rapidly. Compared with the prior art, the beneficial effects of the computer program product provided by the present application are the same as those of the scalp abnormality dynamic detection method provided in the above embodiments, and will not be elaborated here.

[0186] Compared with the prior art, the scalp abnormality dynamic detection method, device, equipment, medium and computer product proposed in the embodiments of the present application extract the business feature information of the target service, perform data standardization processing on the business feature information to obtain standard feature data, perform hashing processing on the standard feature data to obtain unique feature data, perform numerical processing and splicing processing on the unique feature data to obtain the first business feature value, accumulate the first business feature values of the target service to obtain the target business feature value, and finally compare the target business feature value with the feature value set to obtain the scalp abnormality dynamic detection result. It is more efficient, flexible and reliable than the traditional method of generating a unique key value or a continuous serial number for each business to identify duplicate services. Based on the solution of the present application, by transforming the services in complex scenarios through a series of simple transformations, it is finally transformed into a comparison of two numbers, making the comparison process very intuitive and efficient. The system only needs to simply compare whether these two values are equal to quickly determine whether two services are exactly the same.

[0187] It should be noted that in this article, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements not only includes those elements, but also includes other elements not expressly listed, or also includes elements inherent to such process, method, article or system. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or system including the element.

[0188] The serial numbers of the embodiments of the present application above are only for description and do not represent the superiority or inferiority of the embodiments.

[0189] Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general hardware platform. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. The computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above and includes several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, a controlled terminal, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0190] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application accordingly. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A method for dynamic detection of scalp abnormality, characterized in that: Applied to head massage equipment, the scalp abnormality dynamic detection method includes: While collecting a sequence of images of the head massage device when it moves on the head, acquiring posture change data of the head massage device; Calculating pixel offsets of adjacent image frames of the image sequence using the posture change data; According to the pixel offset, position correction is performed on the detection target of each frame image in the image sequence; An abnormality analysis is performed on the detection target in the image sequence, abnormal areas are marked based on the analysis results, and a detection report is generated.

2. The method for dynamic detection of scalp abnormality according to claim 1, characterized in that: The step of calculating the pixel offsets of adjacent image frames of the image sequence using the posture change data comprises: Calculate the posture matrix of the head massage device when collecting each frame of image according to the posture change data; Obtaining a translation vector between adjacent image frames by extracting transformation information in the posture matrix of adjacent image frames; Convert the translation vector to a pixel offset in the image coordinate system.

3. The scalp abnormality dynamic detection method according to claim 1, characterized in that: The step of performing position correction on the detection target of each frame image in the image sequence according to the pixel offset comprises: Using image processing technology to identify the detection target of each frame of the image sequence; For each frame of the image in the image sequence, calculating a cumulative offset from the first frame to the current frame according to the pixel offset; The accumulated offset is used to perform geometric transformation on the detection target in each frame image, so as to move the detection target to a correct position.

4. The scalp abnormality dynamic detection method according to claim 1, characterized in that: After the step of performing position correction on the detection target of each frame image in the image sequence according to the pixel offset, the following steps are included: Calculating error data between the corrected position and the initial position of the detection target in each frame of the image sequence; The correction effect of the detection target in the image sequence is evaluated according to the error data.

5. The method for dynamic detection of scalp abnormality according to claim 1, characterized in that: The step of performing abnormal analysis on the detection target in the image sequence, marking the abnormal area based on the analysis result and generating a detection report comprises: Acquire statistical features of the detection target by analyzing multiple frames of images; Comparing the statistical characteristics of the detection target with a preset threshold to determine whether there is an abnormality; If an abnormality is detected, the abnormal area is marked on the image and a detection report is generated based on the detection results.

6. The method for dynamic detection of scalp abnormality according to claim 1, characterized in that: The step of performing abnormal analysis on the detection target in the image sequence further comprises: Obtaining an abnormality recognition result of the detection target for each frame of the image sequence; Performing time series analysis on the abnormality recognition results of the image sequence to detect the changing trend of the abnormal condition; Analyzing the consistency of abnormality recognition results in the same area in different frame images of the image sequence to obtain a spatial consistency analysis result; Combine the time series analysis results and the spatial consistency analysis results to conduct a comprehensive analysis of the abnormal situation.

7. The method for dynamic detection of scalp abnormality according to claim 1, characterized in that: The step of acquiring posture change data of the head massage device while collecting an image sequence of the head massage device moving on the head comprises: When the head massage device moves on the head, scalp images are continuously collected at a preset frequency to obtain an image sequence, and posture change data of the head massage device is obtained at the same frequency as the frequency of collecting the scalp images.

8. A scalp abnormality dynamic detection device, characterized in that: The device comprises: an acquisition module, for acquiring posture change data of the head massage device while acquiring an image sequence when the head massage device moves on the head; A calculation module, used to calculate pixel offsets of adjacent image frames of the image sequence using the posture change data; A correction module, used for performing position correction on the detection target of each frame image in the image sequence according to the pixel offset; The analysis module is used to perform abnormal analysis on the detection targets in the image sequence, mark abnormal areas based on the analysis results and generate a detection report.

9. A scalp abnormality dynamic detection device, characterized in that: The device comprises: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the scalp abnormality dynamic detection method according to any one of claims 1 to 7.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the scalp abnormality dynamic detection method according to any one of claims 1 to 7 are implemented.

Images