Intelligent identification method and device for human body massage area and computer equipment

Abstract

The invention relates to an intelligent recognition method and device for a human body massage area and computer equipment. The method comprises the following steps: acquiring first image data of a human body, and identifying a human body horizontal direction of the human body based on the first image data; based on the human body horizontal direction of the human body, adjusting an image acquisition pose, re-acquiring second image data of the human body, and based on the second image data and a segmentation strategy, generating a massage area segmentation result of the human body; and based on the massage area segmentation result, identifying a target massage area of the human body. By adopting the method, the recognition accuracy of the massage area can be improved.

Description

Technical Field

[0001] This application relates to the field of image recognition technology, and in particular to an intelligent recognition method, apparatus and computer equipment for human massage areas. Background Technology

[0002] Human massage area recognition can assist intelligent massage devices in identifying massage areas in real time, thereby ensuring the accuracy of the massage. However, the core flaw in existing massage area recognition technology lies in its failure to consider the common situation where the relative posture of the camera and the human body is easily skewed during actual use. It can only be designed based on the ideal scenario that "the camera shooting direction is parallel to the length direction of the human torso." Moreover, the human body area boundaries obtained by image segmentation (such as mask extraction) in traditional recognition technologies are mostly axis-aligned rectangles (i.e., rectangles that can only be horizontal or vertical). Even if the human body is tilted in the actual captured image due to posture skew, camera installation deviation, or shooting angle shift, these axis-aligned rectangles still cannot match the actual tilted posture of the human body. They can only force the human body area to be "fitted" into the horizontal or vertical rectangle, resulting in orientational displacement and contour distortion of the human body area. Consequently, the massage path planned by the massage institution cannot be kept parallel and aligned with the actual posture of the human body, leading to problems such as "skewed massage direction," "uneven pressure," "missed areas," and "inadequate massage of key areas." Therefore, how to improve the accuracy of massage area recognition is the current research focus.

[0003] Current technology relies on a complete match between a standard template and the human body to ensure accuracy. This requires photographing large areas of the body, such as the shoulders, back, waist, and even buttocks, demanding that users deliberately expose more body parts. This not only infringes on user privacy but may also lead to users being reluctant to cooperate and complete the entire photoshoot due to their posture or unwillingness to expose too much, resulting in an incomplete shooting area and thus lower accuracy in recognizing massage areas. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, device, and computer equipment for intelligent recognition of human massage areas to address the aforementioned technical problems.

[0005] Firstly, this application provides a method for intelligent recognition of human massage areas, including:

[0006] Acquire first image data of the human body, and based on the first image data, identify the horizontal direction of the human body;

[0007] Based on the horizontal direction of the human body, the image is adjusted to obtain the pose, the second image data of the human body is re-acquired, and the massage area segmentation result of the human body is generated based on the second image data and the segmentation strategy.

[0008] Based on the massage area segmentation results, the target massage area of ​​the human body is identified.

[0009] Optionally, identifying the horizontal direction of the human body based on the first image data includes:

[0010] Based on the first image data, the range of the human body is identified, and based on the range of the human body, the rotation angle of the human body orientation and the orientation of the human body orientation are located by a rotation target detection network.

[0011] Based on the human body orientation rotation angle and the human body orientation position, a human body orientation detection box is marked within the human body range of the first image data, and the human body orientation detection box is used as the horizontal direction of the human body.

[0012] Optionally, adjusting the image pose based on the horizontal orientation of the human body and re-acquiring the second image data of the human body includes:

[0013] Based on the human body orientation detection frame, the amount of shooting angle adjustment of the camera device is identified;

[0014] Based on the shooting angle adjustment, the camera device is rotated, and upon completion of the rotation, the camera device re-acquires the second image data of the human body.

[0015] Optionally, generating the massage region segmentation result of the human body based on the second image data and the segmentation strategy includes:

[0016] Based on the second image data, the boundary information of the massageable area of ​​the human body is identified through a semantic segmentation model;

[0017] The image range in the boundary information of the massageable area is segmented and labeled to obtain the segmentation result of the massage area of ​​the human body.

[0018] Optionally, identifying the target massage area of ​​the human body based on the massage area segmentation result includes:

[0019] Extract sub-first image data from the image range containing the boundary information of the massageable area;

[0020] Based on the sub-first image data, massage boundary information in the sub-first image data is identified through a massage boundary localization strategy;

[0021] Based on the massage boundary information, the target massage area of ​​the human body is selected from the sub-first image data.

[0022] Optionally, the method further includes:

[0023] Acquire massage task information, and based on the massage task information, identify the massage sequence of the human body and the area massage method of the human body;

[0024] Based on the massage sequence of the human body and the regional massage method of the human body, a massage control command sequence of the massage device is generated in the target massage area;

[0025] Based on the massage control command sequence, the massage device is controlled to perform the massage task information on the target massage area of ​​the human body.

[0026] Secondly, this application also provides an intelligent recognition device for human massage areas, comprising:

[0027] The acquisition module is used to acquire first image data of the human body and, based on the first image data, identify the horizontal direction of the human body.

[0028] The generation module is used to adjust the image acquisition pose based on the horizontal direction of the human body, reacquire the second image data of the human body, and generate the massage area segmentation result of the human body based on the second image data and the segmentation strategy.

[0029] The recognition module is used to identify the target massage area of ​​the human body based on the massage area segmentation results.

[0030] Optionally, the acquisition module is specifically used for:

[0031] Based on the first image data, the range of the human body is identified, and based on the range of the human body, the rotation angle of the human body orientation and the orientation of the human body orientation are located by a rotation target detection network.

[0032] Based on the human body orientation rotation angle and the human body orientation position, a human body orientation detection box is marked within the human body range of the first image data, and the human body orientation detection box is used as the horizontal direction of the human body.

[0033] Optionally, the generation module is specifically used for:

[0034] Based on the human body orientation detection frame, the amount of shooting angle adjustment of the camera device is identified;

[0035] Based on the shooting angle adjustment, the camera device is rotated, and upon completion of the rotation, the camera device re-acquires the second image data of the human body.

[0036] Optionally, the generation module is specifically used for:

[0037] Based on the second image data, the boundary information of the massageable area of ​​the human body is identified through a semantic segmentation model;

[0038] The image range in the boundary information of the massageable area is segmented and labeled to obtain the segmentation result of the massage area of ​​the human body.

[0039] Optionally, the identification module is specifically used for:

[0040] Extract sub-first image data from the image range containing the boundary information of the massageable area;

[0041] Based on the sub-first image data, massage boundary information in the sub-first image data is identified through a massage boundary localization strategy;

[0042] Based on the massage boundary information, the target massage area of ​​the human body is selected from the sub-first image data.

[0043] Optionally, the device may also include:

[0044] The task acquisition module is used to acquire massage task information and, based on the massage task information, identify the massage sequence of the human body and the area massage method of the human body.

[0045] The instruction generation module is used to generate a massage control instruction sequence for the massage device in the target massage area based on the massage sequence of the human body and the area massage method of the human body;

[0046] The execution module is used to control the massage device based on the massage control command sequence to perform the massage task information on the target massage area of ​​the human body.

[0047] Thirdly, this application provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of the method described in any one of the first aspects.

[0048] Fourthly, this application provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0049] Fifthly, this application provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the steps of the method described in any one of the first aspects.

[0050] The aforementioned intelligent recognition method, device, and computer equipment for human massage areas acquire first image data of the human body and, based on this first image data, identify the horizontal direction of the human body; based on the horizontal direction of the human body, adjust the image acquisition pose, re-acquire second image data of the human body, and, based on the second image data and a segmentation strategy, generate a massage area segmentation result for the human body; based on the massage area segmentation result, identify the target massage area of ​​the human body. This solution automatically corrects the image acquisition pose of the human body area by detecting the horizontal direction of the human body, ensuring that the corrected image acquisition pose matches the human body area and that subsequent massage area recognition is not affected by pose distortion. Secondly, after correction, this solution directly extracts the contour of a local area of ​​the human body through a segmentation strategy, accurately distinguishing the massageable area of ​​the human body from non-human backgrounds such as clothing wrinkles, bedding, and surrounding debris, eliminating interference from invalid areas. Finally, this solution further corrects and segments the contour to directly generate an effective massage area focusing on this local range, thereby accurately locating the actual massage area of ​​different human bodies and effectively improving the accuracy of massage area recognition. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a flowchart illustrating a method for intelligent recognition of a human massage area in one embodiment;

[0053] Figure 2 This is an example diagram of the detection box corresponding to the horizontal direction of the human body in one embodiment;

[0054] Figure 3 This is an example diagram of the second image data in one embodiment;

[0055] Figure 4 This is a region outline map of a massageable area of ​​the human body in one embodiment;

[0056] Figure 5 This is an example diagram illustrating boundary recognition of massage boundary information in one embodiment;

[0057] Figure 6 This is a flowchart illustrating an example of intelligent recognition of a human massage area in one embodiment;

[0058] Figure 7 This is a structural block diagram of an intelligent recognition device for a human massage area in one embodiment;

[0059] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0060] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0061] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0062] The intelligent recognition method for human massage areas provided in this application embodiment can be applied to a system for intelligent recognition of human massage areas. This system can be used with a terminal, which can be, but is not limited to, various personal computers, laptops, mid-range computers, etc. Specifically, the terminal detects the horizontal direction of the human body to automatically correct the image acquisition pose of the human body area, ensuring that the corrected image acquisition pose matches the human body area and that subsequent massage area recognition is not affected by pose distortion. Secondly, after correction, this solution uses a segmentation strategy to directly extract the contour of a local area of ​​the human body, accurately distinguishing the massageable area from non-human backgrounds such as clothing wrinkles, bedding, and surrounding debris, eliminating interference from invalid areas. Finally, this solution uses further correction and contour segmentation to directly generate an effective massage area focused on that local range, thereby accurately locating the actual massage area of ​​different human bodies and effectively improving the accuracy of massage area recognition.

[0063] In one exemplary embodiment, such as Figure 1 As shown, a method for intelligent recognition of human massage areas is provided. Taking the application of this method to a terminal as an example, the method includes the following steps S101 to S103. Wherein:

[0064] Step S101: Acquire first image data of the human body, and identify the horizontal direction of the human body based on the first image data.

[0065] In this embodiment, the terminal receives the image transmission result from the camera device and obtains the first image data of the human body captured by the camera device. The camera device can be a color camera, high-resolution camera, or other camera / photography device capable of capturing color images. This camera device is positioned directly above the massage area of ​​the human body and can capture images from all angles of the human body. The first image data is a color image including the human body. Then, the terminal identifies the horizontal direction of the human body using the rotation target detection strategy designed in this scheme. This rotation target detection strategy is based on the YOLO-OBB algorithm and is used to detect the tilt posture of the shoulder-back local massage area of ​​the human body. Figure 2 The image shows the detection frame corresponding to the horizontal direction of the human body. The specific detection process will be explained in detail later.

[0066] Step S102: Based on the horizontal direction of the human body, adjust the image acquisition pose, reacquire the second image data of the human body, and generate the massage area segmentation result of the human body based on the second image data and the segmentation strategy.

[0067] In this embodiment, the terminal readjusts the camera angle based on the horizontal direction of the human body and re-acquires the second image data of the human body. This results in the second image data of the human body being as follows: Figure 3 The image shows the first image data, where the shoulder is parallel to the side of the image. Then, based on the second image data, the terminal generates a segmentation result of the massageable area of ​​the human body using a semantic segmentation strategy. This semantic segmentation strategy is based on the YOLO-seg network and is used to identify the contours of the massageable area of ​​the human body. The massageable area refers to the body's overall range, for example... Figure 4 The image shown is a map illustrating the outline of the massageable areas on the human body. The specific identification process will be explained in detail later.

[0068] Step S103: Based on the massage area segmentation results, identify the target massage area of ​​the human body.

[0069] In this embodiment, the terminal identifies the target massage area of ​​the human body based on the massage area segmentation results and employs a massage area recognition strategy. This massage area recognition strategy uses the region contour map obtained above as a reference, defining the horizontal pixel direction of the image as the Y-axis and the vertical pixel direction as the X-axis, and then filters according to these directions. The specific strategy filtering process will be explained in detail later.

[0070] Based on the above scheme, by detecting the horizontal direction of the human body, the image acquisition pose of the human body region is automatically corrected, ensuring that the corrected image acquisition pose matches the human body region and that subsequent massage area recognition is not affected by posture distortion. Secondly, after correction, this scheme uses a segmentation strategy to directly extract the contour of the local human body region, accurately distinguishing the massageable area of ​​the human body from non-human backgrounds such as clothing wrinkles, bedding, and surrounding debris, eliminating invalid areas of interference. Finally, this scheme uses further correction and contour segmentation to directly generate an effective massage area focused on this local range, thereby accurately locating the actual massage area for different human bodies and effectively improving the accuracy of massage area recognition.

[0071] Optionally, based on the first image data, identifying the horizontal direction of the human body includes: based on the first image data, identifying the range of the human body, and based on the range of the human body, locating the rotation angle of the human body orientation and the orientation of the human body through a rotation target detection network; based on the rotation angle of the human body orientation and the orientation of the human body orientation, marking a human body orientation detection box in the range of the human body in the first image data, and using the human body orientation detection box as the horizontal direction of the human body.

[0072] In this embodiment, the terminal identifies the human body's range based on the first image data, and then, based on this range, uses a rotation target detection network to locate the human body's orientation rotation angle and orientation. The human body range refers to the image area containing the torso. This image range identification is coarse; that is, using intelligent recognition technology, the rectangular frame containing the torso in the first image data is selected as the human body range. The human body orientation rotation angle is the angle between the posture direction of the shoulder-back local massage area and the side direction of the first image data. This posture direction is the human body orientation, which corresponds to the direction perpendicular to the line connecting the two shoulders.

[0073] Finally, based on the human body's orientation rotation angle and orientation, the terminal marks a human body orientation detection box within the human body area of ​​the first image data, and uses this human body orientation detection box as the horizontal direction of the human body. The marked orientation detection box is as follows: Figure 2 The green detection box shown.

[0074] Based on the above scheme, the tilt posture of the shoulder-back local massage area of ​​the human body is detected by the YOLO-OBB algorithm, and the relative tilt angle between the camera and the human body is obtained, which improves the accuracy of the recognition of the tilt angle of the human body.

[0075] Optionally, based on the horizontal orientation of the human body, the image is adjusted to obtain the pose, and the second image data of the human body is reacquired, including: based on the human body orientation detection box, identifying the shooting angle adjustment amount of the camera device; based on the shooting angle adjustment amount, performing device rotation processing on the camera device, and when the device rotation processing is completed, reacquiring the second image data of the human body through the camera device.

[0076] In this embodiment, the terminal identifies the shooting angle adjustment amount of the camera device based on the human body orientation detection frame. The adjustment amount of the shooting angle is defined by using the human body's orientation as the left boundary of the shooting angle, and the rotation angle of that human body orientation is used as the adjustment value for the angle change.

[0077] Finally, based on the shooting angle adjustment, the terminal performs device rotation processing on the camera device, and when the device rotation processing is completed, it re-acquires the second image data of the human body through the camera device.

[0078] Based on the above scheme, the orientation of the human body region in the image is automatically corrected according to the included angle, so that the corrected human body region is adapted and aligned with the camera shooting angle, ensuring that the subsequent massage area recognition is not affected by posture distortion, thereby directly avoiding region stretching and offset, making the massage path precisely parallel to the human body, and the processing efficiency is far superior to traditional plane fitting and perspective transformation.

[0079] Optionally, based on the second image data and the segmentation strategy, a segmentation result of the massage area of ​​the human body is generated, including: based on the second image data, identifying the boundary information of the massageable area of ​​the human body through a semantic segmentation model; performing segmentation and labeling processing on the image range in the boundary information of the massageable area to obtain the segmentation result of the massage area of ​​the human body.

[0080] In this embodiment, the terminal identifies the boundary information of the massageable area of ​​the human body based on the second image data and through a semantic segmentation model. The boundary information of the massageable area is as follows: Figure 4 The outline of the human torso is shown.

[0081] Then, the terminal performs segmentation and labeling processing on the image range in the boundary information of the massageable area to obtain the segmentation result of the massageable area of ​​the human body. The segmentation and labeling processing can be performed in a manner such as... Figure 4 The color markings shown can also be used, or other marking methods such as symbol markings, to mark areas in the image using red boxes.

[0082] Based on the above scheme, the YOLO-seg algorithm is used to perform semantic segmentation on the corrected image, directly extracting the contour mask of the local human body region, accurately distinguishing the massageable area of ​​the human body from non-human backgrounds such as clothing folds, bedding, and surrounding clutter, eliminating invalid area interference, making the boundary of the massage area clearer, and having stronger anti-interference ability than template matching and key point partitioning.

[0083] Optionally, based on the massage region segmentation results, the target massage region of the human body is identified, including: extracting sub-first image data from the image range containing the boundary information of the massageable region; based on the sub-first image data, identifying each massage boundary information in the sub-first image data through a massage boundary localization strategy; and based on each massage boundary information, filtering the target massage region of the human body in the sub-first image data.

[0084] In this embodiment, the terminal extracts sub-first image data from the image range containing the boundary information of the massageable area. Then, based on the sub-first image data, the terminal identifies each massage boundary information in the sub-first image data using a massage boundary positioning strategy. Specifically, the massage boundary positioning strategy is executed as follows:

[0085] like Figure 5 As shown, the terminal iterates through the pixel values ​​of the first image data with the maximum X-difference along the Y-axis, where the yellow line represents the interval of the maximum X-difference. Then, the terminal iterates through the pixel value range in the first image data from the pixel value of the maximum X-difference to the pixel value of the minimum X-difference along the Y-axis, where the pixel value of the minimum X-difference is located at... Figure 5 The green line is located in the interval; then, the terminal calculates the upper and lower bounds of X for the minimum X difference along the Y direction, i.e. Figure 5 The range between the two ends of the blue line; finally, the terminal determines the left and right boundaries of the final massage area through the left and right boundaries of the left and right red borders, that is, as shown in the figure. Figure 5 The green box indicates the massage boundary information for the final target massage area.

[0086] Finally, based on the massage boundary information, the terminal filters the target massage area of ​​the human body from the first sub-image data.

[0087] Based on the above scheme, the human body contour mask obtained by YOLO-seg segmentation is fitted to generate a boundary box. The coordinates of the four corner points and the center point of the box are extracted. With the center point as a fixed scaling center, the width (w) and height (h) of the box are proportionally calculated and adjusted according to a preset ratio to obtain the final massage area range that is adapted to the massage device's stroke. Therefore, manual calibration is not required. The width and height of the area are dynamically adjusted by scaling the center ratio, which takes into account both human body fit and device adaptability. This is more flexible and efficient than fixed template scaling.

[0088] Optionally, the method further includes: acquiring massage task information, and based on the massage task information, identifying the massage sequence of the human body and the regional massage method of the human body; based on the massage sequence of the human body and the regional massage method of the human body, generating a massage control command sequence of the massage device in the target massage area; and based on the massage control command sequence, controlling the massage device to execute the massage task information on the massage area of ​​the human body.

[0089] In this embodiment, the terminal acquires massage task information and, based on this information, identifies the massage sequence and area massage methods of the human body. The massage task information refers to the device massage task of the massage equipment, which is a sequence of execution instructions pre-set by the operator. The massage sequence is the order in which different massage areas within the massage area of ​​the human body are massaged. The area massage methods include the massage methods corresponding to different positions of each massage area within the massage sequence. Then, based on the human body's massage sequence and area massage methods, the terminal generates a massage control instruction sequence for the massage equipment in the target massage area. This massage control instruction sequence consists of the order in which different massage areas within the massage area of ​​the human body are massaged, and the massage instructions corresponding to the area massage methods for each massage area. Finally, based on the massage control instruction sequence, the terminal controls the massage equipment to execute the massage task information on the human body's massage area.

[0090] Based on the above solution, the target massage area identified by this solution is used to control the massage device, thereby improving the accuracy of massage on the human body.

[0091] This application also provides an example of intelligent recognition of human massage areas, such as... Figure 6 As shown, the specific processing procedure includes the following steps:

[0092] Step S601: Obtain the first image data of the human body.

[0093] Step S602: Based on the first image data, identify the range of the human body, and based on the range of the human body, locate the rotation angle of the human body and the orientation of the human body through a rotation target detection network.

[0094] Step S603: Based on the human body orientation rotation angle and the human body orientation direction, mark the human body orientation detection box in the human body range of the first image data, and use the human body orientation detection box as the human body horizontal direction.

[0095] Step S604: Based on the human body orientation detection frame, identify the shooting angle adjustment amount of the camera device.

[0096] Step S605: Based on the shooting angle adjustment amount, the camera device is rotated, and when the rotation is completed, the second image data of the human body is re-acquired through the camera device.

[0097] Step S606: Based on the second image data, the boundary information of the massageable area of ​​the human body is identified through a semantic segmentation model.

[0098] Step S607: Perform segmentation and labeling processing on the image range in the boundary information of the massageable area to obtain the segmentation result of the massage area of ​​the human body.

[0099] Step S608: Extract the sub-first image data from the image range containing the boundary information of the massageable area.

[0100] Step S609: Based on the sub-first image data, identify the massage boundary information in the sub-first image data through a massage boundary positioning strategy.

[0101] Step S610: Based on the massage boundary information, select the target massage area of ​​the human body in the sub-first image data.

[0102] Step S611: Obtain massage task information, and based on the massage task information, identify the massage sequence of the human body and the massage method of the human body's regions.

[0103] Step S612: Based on the massage sequence of the human body and the regional massage method of the human body, generate a massage control command sequence of the massage device in the target massage area.

[0104] Step S613: Based on the massage control command sequence, control the massage device to perform massage task information on the target massage area of ​​the human body.

[0105] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0106] Based on the same inventive concept, this application also provides an intelligent recognition device for a human massage area to implement the intelligent recognition method for the human massage area described above. The solution provided by this device is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the intelligent recognition device for a human massage area provided below can be found in the limitations of the intelligent recognition method for a human massage area described above, and will not be repeated here.

[0107] In one exemplary embodiment, such as Figure 7 As shown, an intelligent recognition device for human massage areas is provided, comprising: an acquisition module 710, a generation module 720, and a recognition module 730, wherein:

[0108] The acquisition module 710 is used to acquire first image data of the human body, and based on the first image data, identify the horizontal direction of the human body through a rotation target detection strategy;

[0109] The generation module 720 is used to re-acquire second image data of the human body based on the horizontal direction of the human body, and generate a massage region segmentation result of the human body based on the second image data through a semantic segmentation strategy.

[0110] The recognition module 730 is used to identify the target massage area of ​​the human body based on the massage area segmentation result and through a massage area recognition strategy.

[0111] Optionally, the acquisition module 710 is specifically used for:

[0112] Based on the first image data, the range of the human body is identified, and based on the range of the human body, the rotation angle of the human body orientation and the orientation of the human body orientation are located by a rotation target detection network.

[0113] Based on the human body orientation rotation angle and the human body orientation position, a human body orientation detection box is marked within the human body range of the first image data, and the human body orientation detection box is used as the horizontal direction of the human body.

[0114] Optionally, the generation module 720 is specifically used for:

[0115] Based on the human body orientation detection frame, the amount of shooting angle adjustment of the camera device is identified;

[0116] Based on the shooting angle adjustment, the camera device is rotated, and upon completion of the rotation, the camera device re-acquires the second image data of the human body.

[0117] Optionally, the generation module 720 is specifically used for:

[0118] Based on the second image data, the boundary information of the massageable area of ​​the human body is identified through a semantic segmentation model;

[0119] The image range in the boundary information of the massageable area is segmented and labeled to obtain the segmentation result of the massage area of ​​the human body.

[0120] Optionally, the identification module 730 is specifically used for:

[0121] Extract sub-first image data from the image range containing the boundary information of the massageable area;

[0122] Based on the sub-first image data, massage boundary information in the sub-first image data is identified through a massage boundary localization strategy;

[0123] Based on the massage boundary information, the target massage area of ​​the human body is selected from the sub-first image data.

[0124] Optionally, the device may also include:

[0125] The task acquisition module is used to acquire massage task information and, based on the massage task information, identify the massage sequence of the human body and the area massage method of the human body.

[0126] The instruction generation module is used to generate a massage control instruction sequence for the massage device in the target massage area based on the massage sequence of the human body and the area massage method of the human body;

[0127] The execution module is used to control the massage device based on the massage control command sequence to perform the massage task information on the target massage area of ​​the human body.

[0128] The various modules in the aforementioned intelligent recognition device for the human massage area can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0129] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 8As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements an intelligent recognition method for human massage areas. The display unit is used to form a visually visible image and can be a display screen, projection device, or virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0130] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0131] In one exemplary embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of a beer warehouse inventory optimization method.

[0132] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of a beer warehouse inventory optimization method.

[0133] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of a beer warehouse inventory optimization method.

[0134] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0135] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0136] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0137] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for intelligent recognition of human massage areas, characterized in that, The method includes: Acquire first image data of the human body, and based on the first image data, identify the horizontal direction of the human body; Based on the horizontal direction of the human body, the image is adjusted to obtain the pose, the second image data of the human body is re-acquired, and the massage area segmentation result of the human body is generated based on the second image data and the segmentation strategy. Based on the massage area segmentation results, the target massage area of ​​the human body is identified.

2. The method according to claim 1, characterized in that, The step of identifying the horizontal direction of the human body based on the first image data includes: Based on the first image data, the range of the human body is identified, and based on the range of the human body, the rotation angle of the human body orientation and the orientation of the human body orientation are located by a rotation target detection network. Based on the human body orientation rotation angle and the human body orientation position, a human body orientation detection box is marked within the human body range of the first image data, and the human body orientation detection box is used as the horizontal direction of the human body.

3. The method according to claim 2, characterized in that, The step of adjusting the image to obtain the pose based on the horizontal direction of the human body and then re-acquiring the second image data of the human body includes: Based on the human body orientation detection frame, the amount of shooting angle adjustment of the camera device is identified; Based on the shooting angle adjustment, the camera device is rotated, and upon completion of the rotation, the camera device re-acquires the second image data of the human body.

4. The method according to any one of claims 1 to 3, characterized in that, The step of generating the massage region segmentation result of the human body based on the second image data and the segmentation strategy includes: Based on the second image data, the boundary information of the massageable area of ​​the human body is identified through a semantic segmentation model; The image range in the boundary information of the massageable area is segmented and labeled to obtain the segmentation result of the massage area of ​​the human body.

5. The method according to claim 4, characterized in that, The step of identifying the target massage area of ​​the human body based on the massage area segmentation result includes: Extract sub-first image data from the image range containing the boundary information of the massageable area; Based on the sub-first image data, massage boundary information in the sub-first image data is identified through a massage boundary localization strategy; Based on the massage boundary information, the target massage area of ​​the human body is selected from the sub-first image data.

6. The method according to claim 1, characterized in that, The method further includes: Acquire massage task information, and based on the massage task information, identify the massage sequence of the human body and the area massage method of the human body; Based on the massage sequence of the human body and the regional massage method of the human body, a massage control command sequence of the massage device is generated in the target massage area; Based on the massage control command sequence, the massage device is controlled to perform the massage task information on the target massage area of ​​the human body.

7. An intelligent recognition device for human massage areas, characterized in that, The device includes: The acquisition module is used to acquire first image data of the human body and, based on the first image data, identify the horizontal direction of the human body. The generation module is used to adjust the image acquisition pose based on the horizontal direction of the human body, reacquire the second image data of the human body, and generate the massage area segmentation result of the human body based on the second image data and the segmentation strategy. The recognition module is used to identify the target massage area of ​​the human body based on the massage area segmentation results.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images