Automatic hairdressing method, device and system

Abstract

The invention relates to the field of automation, in particular to an automatic haircut method, an automatic haircut device and an automatic haircut system. The rebounding amplitude and the rebounding direction of the hair combed by the hair comb during cutting can be determined according to the hair characteristics during haircut, so that whether the risk of excessive cutting of the hair exists when the hair clipper moves according to a set path or not is judged accordingly, if yes, adaptive yaw adjustment is conducted on the set path, and if not, the rebounding amplitude and the rebounding direction are judged. Therefore, the hair clipper is controlled to move according to the adjusted path to avoid excessive cutting; according to the method and the device, the hair property characteristics can be accurately monitored, so that the rebound characteristics of the combed hair are accurately estimated, the haircut route of the haircut device is subjected to targeted yaw adjustment, the problem that the haircut device excessively cuts the hair due to the rebound of the hair of a user can be effectively avoided, and the user experience is improved. And a good hairdressing effect is ensured.

Description

Technical Field

[0001] This invention relates to the field of automation, and in particular to an automated haircutting method, apparatus and system. Background Technology

[0002] Automatic haircutting equipment is a device that integrates automation technology, designed to complete the haircutting process without human operation, and is gradually being used more and more widely; Automatic hair-cutting devices typically require the use of a comb and clippers (such as electric clippers) during the haircut. The comb is used to lift the hair, and as the device is moved to the hair, the comb moves synchronously, allowing the lifted hair to enter the clipper's blades and be cut. However, due to variations in hair parameters (such as thickness and length) and direction of hair strands at different head locations, the lifted hair may spring back to varying degrees after the comb moves. This can cause some hair strands to spring back excessively before entering the blades, resulting in over-cutting (e.g., instead of cutting 1cm, 2cm is cut short), potentially leading to localized hair loss and affecting the haircut's outcome. Summary of the Invention

[0003] Therefore, it is necessary to provide an automatic haircutting method, device, and system to address the aforementioned problems.

[0004] The present invention is implemented as follows: an automatic haircutting method, the method comprising: S1: Before starting the haircut, capture an image of the area of ​​hair to be trimmed; S2: Based on the acquired images, identify the hair characteristics and natural direction of each sub-region of the hair area; S3: After the haircut begins, monitor the position of the hair comb to determine the rebound amplitude and direction of the combed hair based on the hair characteristics at the location of the hair comb and the natural orientation of the hair. S4: Monitor the real-time position of the hair clipper and generate the first path for the hair clipper to move from the real-time position to the combed hair; S5: Based on the determined rebound amplitude and rebound direction, determine whether there is a risk of excessive cutting of hair when the hair clipper moves along the first path. If not, do not adjust the first path. S6: If so, the first path is adaptively yawed according to the rebound amplitude and rebound direction to obtain the second path. The hair clipper is controlled to move along the second path to avoid excessive cutting of the hair.

[0005] In one embodiment, the present invention provides an automatic hair-cutting device, the device comprising: The image acquisition module is used to acquire images of the area of ​​hair to be trimmed before starting the haircut; The first processing module is used to identify the hair features and natural orientation of each sub-region of the hair region based on the acquired image. The second processing module is used to monitor the position of the hair comb after the haircut begins, and then determine the rebound amplitude and rebound direction of the combed hair based on the hair characteristics at the location of the hair comb and the natural orientation of the hair. The third processing module is used to monitor the real-time position of the hair clipper and generate the first path for the hair clipper to move from the real-time position to the combed hair. The judgment module is used to determine whether there is a risk of excessive cutting of the hair when the hair clipper moves along the first path based on the determined rebound amplitude and rebound direction. If not, the first path is not adjusted. The fourth processing module is used to adaptively adjust the yaw of the first path according to the rebound amplitude and rebound direction to obtain the second path, and control the hair clipper to move according to the second path, thereby avoiding excessive cutting of the hair.

[0006] In one embodiment, the present invention provides an automated haircutting system, the system comprising: A vision device used to acquire images of the hair region; A hair-cutting device, connected to a vision device, for performing the described automatic hair-cutting method.

[0007] This invention provides an automatic haircutting method, apparatus, and system. The method includes: before starting the haircut, acquiring an image of the hair area to be trimmed; identifying hair characteristics and natural hair orientation in each sub-region of the hair area based on the acquired image; after starting the haircut, monitoring the position of the hair comb to determine the rebound amplitude and direction of the combed hair based on the hair characteristics and natural hair orientation at the location of the hair comb; monitoring the real-time position of the hair clipper and generating a first path for the hair clipper to move from the real-time position to the combed hair; determining whether there is a risk of over-cutting the hair if the hair clipper moves along the first path based on the determined rebound amplitude and direction; if not, not adjusting the first path; if so, adaptively yawing the first path based on the rebound amplitude and direction to obtain a second path, and controlling the hair clipper to move along the second path. This application utilizes a method to accurately monitor hair characteristics. By analyzing images of the user's hair area, the characteristics of the hair at various locations can be determined. This allows the system to determine the rebound amplitude and direction of the hair combed by the clipper during cutting, thus assessing whether there is a risk of over-cutting the hair as it moves along a predetermined path. If so, the predetermined path is adaptively adjusted to control the clipper's movement along the adjusted path, preventing over-cutting. Furthermore, this application enables accurate monitoring of hair properties, allowing for accurate prediction of the rebound characteristics of the combed hair. This, in turn, allows for targeted yaw adjustments to the clipper's cutting path, effectively preventing over-cutting due to hair rebound and ensuring a good haircutting result. Attached Figure Description

[0008] Figure 1 A flowchart of an automated haircutting method provided in one embodiment; Figure 2 This is an application environment diagram of an automated haircutting method provided in one embodiment; Figure 3 A flowchart of the automatic haircutting method provided in one embodiment; Figure 4 This is a block diagram of the internal structure of a hair-cutting device in one embodiment. Detailed Implementation

[0009] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0010] It is understood that the terms "first," "second," etc., used in this invention may be used to describe various elements herein, but unless specifically stated otherwise, these elements are not limited by these terms. These terms are used only to distinguish one element from another. For example, without departing from the scope of this invention, a first script may be referred to as a second script, and similarly, a second script may be referred to as a first script.

[0011] like Figure 1 As shown, in one embodiment, an automated haircutting method is proposed, the method comprising: S1: Before starting the haircut, capture an image of the area of ​​hair to be trimmed; S2: Based on the acquired images, identify the hair characteristics and natural direction of each sub-region of the hair area; S3: After the haircut begins, monitor the position of the hair comb to determine the rebound amplitude and direction of the combed hair based on the hair characteristics at the location of the hair comb and the natural orientation of the hair. S4: Monitor the real-time position of the hair clipper and generate the first path for the hair clipper to move from the real-time position to the combed hair; S5: Based on the determined rebound amplitude and rebound direction, determine whether there is a risk of excessive cutting of hair when the hair clipper moves along the first path. If not, do not adjust the first path. S6: If so, the first path is adaptively yawed according to the rebound amplitude and rebound direction to obtain the second path. The hair clipper is controlled to move along the second path to avoid excessive cutting of the hair.

[0012] In this embodiment, as Figure 2 As shown, this method is executed in a hair-cutting device; the hair-cutting device is a fully automatic hair-cutting device, which is equipped with a processor, hair clippers and hair combs; the hair-cutting device is also connected to several vision devices, which can then be used to acquire images of the head area.

[0013] In this embodiment, the vision device consists of multiple high-definition cameras and a depth sensor. When the user uses the hair-cutting device, it performs a 360° scan around the head, covering the entire hair area, including the forehead, top of the head, sides, and back of the head, capturing images without blind spots. The acquisition parameters are set as follows: the camera resolution is set to 1920×1080, the frame rate is 30fps, and the depth sensor ranging range is 0-10cm (adapted to hair length). During acquisition, one frame is captured every 5°, resulting in a total of 72 frames of two-dimensional images and corresponding depth images, forming three-dimensional point cloud data of the head hair. Image preprocessing includes noise reduction (using Gaussian filtering algorithm), grayscale conversion, edge enhancement, removal of scalp reflection and ambient light interference, extraction of the boundary between the hair area (the area with hair) and other areas, removal of non-hair pixels (such as clothing and skin), and retention of valid hair area data.

[0014] In this embodiment, after determining the second path, the processor of the hair-cutting device sends a path switching command to the hair clipper, controlling the hair clipper to smoothly transition from the first path to the second path, and providing real-time feedback on the movement position to ensure that the cutting length of the hair accurately matches the corresponding target length when moving along the second path, avoiding over-cutting. Simultaneously, the position of the hair comb is adaptively adjusted to maintain the combing state in conjunction with the cutting action of the hair clipper.

[0015] In this embodiment, after step S6, the hair cutting in one position is completed. Then, the hair cutting device controls the hair comb to move and comb up the hair in another position, and then steps S3 to S6 are executed again. The above steps are repeated until the hair cutting is completed.

[0016] In this application, the hair characteristics of the user's hair at various locations can be determined based on monitoring images of the user's hair area. This allows the system to determine the rebound amplitude and direction of the hair combed up during haircutting based on these hair characteristics. This enables the system to assess whether there is a risk of over-cutting the hair as the clipper moves along a predetermined path. If so, the predetermined path is adaptively adjusted to control the clipper to move along the adjusted path, thus avoiding over-cutting. This application also enables accurate monitoring of hair properties, allowing for accurate prediction of the rebound characteristics of the combed hair. This, in turn, allows for targeted adjustment of the clipper's cutting path, effectively preventing over-cutting due to hair rebound and ensuring a good haircutting result.

[0017] In a preferred embodiment, the acquired images include two-dimensional images and corresponding depth images, thereby obtaining three-dimensional point cloud data of the hair region; based on the acquired images, the hair features of each sub-region of the hair region and the natural orientation of the hair are identified, including: Based on the 3D point cloud data of the head, the hair region is divided into several sub-regions; For each sub-region, calculate the corresponding hair length L and hair density. Hair diameter d; For each sub-region, determine the natural growth direction vector of each individual hair within that sub-region, and then determine the natural orientation of the hair in that sub-region based on the direction vectors of all hairs.

[0018] Between steps S2 and S3, the following is also included: The system obtains the user's target hairstyle, retrieves the hairstyle model of the target hairstyle from the hairstyle library, and adaptively adjusts the size of the invention model according to the size of the user's head area to fit the user's head shape. The hairstyle model is compared with the point cloud model of the user's hair area to determine the target length that needs to be cut at each position of the hair area.

[0019] In this embodiment, human hair features exhibit different characteristics in different regions, while the features of hair in the same region tend to be similar. Based on this hair distribution characteristic and the collected three-dimensional point cloud data of the head, the hair region is divided into several sub-regions, including the forehead region, the top of the head region, the left temporal region, the right temporal region, the upper occipital region, and the lower occipital region. The area of ​​each sub-region does not exceed 10cm² and the hair features within the sub-region are roughly the same, ensuring the accuracy of feature recognition.

[0020] In this embodiment, for each sub-region, the hair length L and hair density are... Both the hair diameter d and the average value of the corresponding sub-region can be taken as the average value of all hairs in the sub-region identified from the detection image. In order to identify more hair features, the hair comb can be used to lift the hair during image acquisition to collect the features of the obscured hair, thereby improving the accuracy of detection. In this embodiment, the hair-cutting device may also include a display screen, which can display various hairstyles before the haircut, allowing the user to select the desired hairstyle. The hair-cutting device also stores three-dimensional hairstyle models of various target hairstyles. Once the target hairstyle is determined, the corresponding hairstyle model can be retrieved, and the size of the hairstyle model can be adjusted according to the size of the user's hair area to make the hairstyle model fit the hair area. Furthermore, the hair-cutting device can generate a three-dimensional point cloud model of the hair area based on the collected three-dimensional point cloud data. The hairstyle model and the point cloud model can then be superimposed, so that each position on the point cloud model corresponds one-to-one with the corresponding position on the hairstyle model. The hairstyle model is compared with the point cloud model of the user's hair area to determine the target length to be cut at each position of the hair area. That is, for each position on the point cloud model, the hair length at that position (i.e., the hair length of the sub-region where that position is located) is retrieved as the first length, and the hair length at the corresponding position on the hairstyle model (the hair length at each position on the hairstyle model is known) is retrieved as the second length. The target length at that position is obtained by subtracting the second length from the first length.

[0021] In this embodiment, determining the natural growth direction vector of each individual hair within the sub-region, and then determining the natural orientation of the hair in the sub-region based on the direction vectors of all hairs, includes: Determine the direction vector of n hairs in this sub-region , … … ;in, Let i be the direction vector of the i-th hair. Let be the vector angle of the i-th hair (obtained through the built-in depth image recognition algorithm), and n be the number of hairs in the sub-region. The natural orientation of the hair in this sub-region is calculated using the following formula: in, This represents the natural orientation of the hair in this sub-region.

[0022] As a preferred embodiment, determining the bounce amplitude and bounce direction of the combed hair based on the hair characteristics at the location of the hair comb and the natural orientation of the hair includes: Monitor the contact pressure between the hair comb teeth and the hair to determine whether the hair comb has lifted the hair; When combing your hair with a hairbrush, determine the position where the hairbrush will pick up the hair; Identify the sub-region where the location is located and retrieve the hair features of that sub-region; The rebound amplitude of the combed hair is calculated based on the contact pressure and the hair characteristics in the sub-region; The direction of the rebound of the combed hair is calculated based on the direction in which the hair is combed and the contact pressure.

[0023] In this embodiment, a pressure sensor is integrated on the teeth of the hair comb, and a pressure threshold, such as 50Pa, is pre-input into the hair-cutting device (this value is determined by repeated experiments). When the hair-cutting device detects that the pressure on the hair comb teeth exceeds the pressure threshold, it can be considered that the hair comb has effectively combed the hair. In this embodiment, the bounce amplitude of the combed hair is calculated using the following formula: in, The rebound range, The proportionality coefficient can be taken as 0.02, and P is the contact pressure; The direction of hair bounce after being combed can be calculated using the following formula: in, The direction of the rebound. The direction in which the hair is combed. This refers to the maximum working pressure of the comb teeth (i.e., the maximum pressure the comb teeth have historically exerted when combing hair).

[0024] As a preferred embodiment, generating a first path for the barber to move from its real-time position toward the combed hair includes: Retrieve the target length to be cut corresponding to the combed hair; Starting from the top of the combed hair, determine the target cutting segment of the target length on the hair, where the end point of the target cutting segment is the cutting point of the hair; Generate fitting lines for each cutting point, and take the midpoint of the fitting line as the target point; Generate all paths from the real-time location of the barber to the target point; The generated paths are filtered out to find paths that do not pass through other hairs, and the shortest path among the filtered paths is selected as the first path.

[0025] In this embodiment, the position of the midpoint of the hair clipper blade (position, i.e., spatial coordinates) can be used as the position of the hair clipper. The combed hair is often arranged in a row with multiple strands. A cutting point can be determined for each strand. Since the hair lengths in the same sub-region are roughly the same, the fitting line of the cutting point is roughly a straight line segment. The midpoint of this straight line segment is taken as the target point. The method of determining the cutting point of each strand is to determine the length of the topmost point of the strand (the end not connected to the head, i.e., the starting point of the strand) as a strand of the target length. The end point of this strand is the cutting point.

[0026] As a preferred embodiment, for each piece of hair that is combed up, the vertex coordinates of that hair are determined as the first coordinate; The coordinates of the intersection point between the clipper blade and the hair when the clipper moves along the first path are determined as the second coordinates; Determine the path direction in which the clipper is positioned on the first path when it moves along the first path to the hair. Based on the determined first coordinates, second coordinates, and path direction, calculate the cutting length of the hair when the hair clipper moves to the hair along the first path; Determine whether the cutting length of hair exceeding a set proportion in the combed hair exceeds the corresponding target length. If so, there is a risk of over-cutting the hair; otherwise, there is no risk of over-cutting the hair.

[0027] In this embodiment, a spatial rectangular coordinate system is established in the field of view of the vision device, which can determine the spatial coordinates of each point in the field of view, thereby determining the coordinates of the tip of each combed hair strand. In this embodiment, the cutting length and target length are compared for each combed hair strand to determine whether there is a risk of over-cutting. If a set proportion (e.g., 20%) or more of the combed hair strands have a risk of over-cutting, it is determined that the current action of moving the hair clipper along the first path has a risk of over-cutting, and then a second path is determined for path adjustment. For each combed hair strand, the corresponding second coordinate can be determined by generating a hair clipper model in the monitoring field of view and simulating the model's movement along the first path. When the blade of the model moves to the hair strand, the coordinates of the intersection point between the hair strand and the blade can be determined, and at this time, the position of the hair clipper model on the first path can be determined. Taking the tangent of the first path at this position yields the corresponding path direction. If the first coordinate is A and the second coordinate is B, the formula for calculating the cutting length can be expressed as: in, For cutting length, This represents the path direction of the first path at the combed hair strands; In this embodiment, a reference direction vector is pre-set in the field of view of the vision device. The path direction, rebound direction, combing direction, and natural orientation of the hair in this application are all direction angles formed by the corresponding direction vector and the reference direction vector. For example, the path direction represents the direction angle between the path direction vector and the reference direction vector.

[0028] As a preferred embodiment, adaptive yaw adjustment of the first path based on the rebound amplitude and rebound direction includes: Determine the offset direction that will move the clipper away from the direction of rebound; The basic offset is determined based on the rebound amplitude and the angle between the rebound direction and the direction angle of the first path. The compensation offset is determined based on the difference between the cutting length and the target length. The target offset is obtained by adding the basic offset to the compensation offset. Determine the offset reference point for the first path, and offset each point on the first path segment after the offset reference point by the target offset along the offset direction to obtain the second path.

[0029] In this embodiment, the offset direction ; The target offset is calculated using the following formula: in, The target offset. This is the distance adjustment factor. The target length; To compensate for the offset, Based on the offset.

[0030] In this embodiment, the offset reference point is a point located on the first path and a distance set in front of the target point. The set distance can be 3cm, 5cm or other distances, and is not limited here.

[0031] like Figure 3 As shown, in one embodiment, an automatic hair-cutting device is provided, the device comprising: The image acquisition module is used to acquire images of the area of ​​hair to be trimmed before starting the haircut; The first processing module is used to identify the hair features and natural orientation of each sub-region of the hair region based on the acquired image. The second processing module is used to monitor the position of the hair comb after the haircut begins, and then determine the rebound amplitude and rebound direction of the combed hair based on the hair characteristics at the location of the hair comb and the natural orientation of the hair. The third processing module is used to monitor the real-time position of the hair clipper and generate the first path for the hair clipper to move from the real-time position to the combed hair. The judgment module is used to determine whether there is a risk of excessive cutting of the hair when the hair clipper moves along the first path based on the determined rebound amplitude and rebound direction. If not, the first path is not adjusted. The fourth processing module is used to adaptively adjust the yaw of the first path according to the rebound amplitude and rebound direction to obtain the second path, and control the hair clipper to move according to the second path, thereby avoiding excessive cutting of the hair.

[0032] For details on how each module in the automatic hair-cutting device provided in this application implements its respective function, please refer to the foregoing. Figure 1The description of the illustrated embodiment will not be repeated here.

[0033] like Figure 2 As shown, in one embodiment, an automated haircutting system is proposed, the system comprising: A vision device used to acquire images of the hair region; A hair-cutting device, connected to a vision device, for performing the described automatic hair-cutting method.

[0034] In this application, the hair-cutting device and the vision device work together to determine the hair characteristics of the user's hair at various locations based on monitoring images of the user's hair area. This allows the device to determine the rebound amplitude and direction of the hair combed up during the haircut based on these hair characteristics. Based on this, it can determine whether there is a risk of over-cutting the hair as the clipper moves along a predetermined path. If so, the predetermined path is adaptively adjusted to control the clipper to move along the adjusted path, thus avoiding over-cutting. This application enables accurate monitoring of hair properties, thereby accurately predicting the rebound characteristics of the combed hair. This allows for targeted adjustment of the clipper's cutting path, effectively preventing over-cutting due to hair rebound and ensuring a good haircutting effect.

[0035] Figure 4 An internal structural diagram of a hair-cutting device in one embodiment is shown. Figure 4 As shown, the hair-cutting device includes a processor, a memory, a network interface, an input device, and a display screen connected via a system bus. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and may also store a computer program. When executed by the processor, this computer program enables the processor to implement the automatic hair-cutting method provided in this embodiment of the invention. The internal memory may also store a computer program, which, when executed by the processor, enables the processor to execute the automatic hair-cutting method provided in this embodiment of the invention. The display screen of the hair-cutting device can be a liquid crystal display or an e-ink display. The input device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad mounted on the outer casing of the hair-cutting device, or an external keyboard, touchpad, or mouse, etc.

[0036] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the hairdressing device to which the present invention is applied. Specific hairdressing devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0037] In one embodiment, the automatic hair-cutting device provided by this invention can be implemented as a computer program, which can be implemented in the form of, for example... Figure 4 The hair-cutting device shown operates on this device. The device's memory can store the various program modules that make up the automatic hair-cutting apparatus, for example, Figure 2 The diagram shows an image acquisition module, a first processing module, a second processing module, a third processing module, a judgment module, and a fourth processing module. The computer program comprised of these modules causes the processor to execute the steps of the automatic haircutting methods described in the various embodiments of the present invention.

[0038] For example, Figure 4 The hair-cutting equipment shown can be used as follows Figure 2 The image acquisition module of the automatic haircutting device shown executes step S1; the haircutting device can execute step S2 through the first processing module; the haircutting device can execute step S3 through the second processing module; the haircutting device can execute step S4 through the third processing module; the haircutting device can execute step S5 through the judgment module; and the haircutting device can execute step S6 through the fourth processing module.

[0039] In one embodiment, a hair-cutting device is provided, the hair-cutting device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to perform the following steps: S1: Before starting the haircut, capture an image of the area of ​​hair to be trimmed; S2: Based on the acquired images, identify the hair characteristics and natural direction of each sub-region of the hair area; S3: After the haircut begins, monitor the position of the hair comb to determine the rebound amplitude and direction of the combed hair based on the hair characteristics at the location of the hair comb and the natural orientation of the hair. S4: Monitor the real-time position of the hair clipper and generate the first path for the hair clipper to move from the real-time position to the combed hair; S5: Based on the determined rebound amplitude and rebound direction, determine whether there is a risk of excessive cutting of hair when the hair clipper moves along the first path. If not, do not adjust the first path. S6: If so, the first path is adaptively yawed according to the rebound amplitude and rebound direction to obtain the second path. The hair clipper is controlled to move along the second path to avoid excessive cutting of the hair.

[0040] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, causes the processor to perform the following steps: S1: Before starting the haircut, capture an image of the area of ​​hair to be trimmed; S2: Based on the acquired images, identify the hair characteristics and natural direction of each sub-region of the hair area; S3: After the haircut begins, monitor the position of the hair comb to determine the rebound amplitude and direction of the combed hair based on the hair characteristics at the location of the hair comb and the natural orientation of the hair. S4: Monitor the real-time position of the hair clipper and generate the first path for the hair clipper to move from the real-time position to the combed hair; S5: Based on the determined rebound amplitude and rebound direction, determine whether there is a risk of excessive cutting of hair when the hair clipper moves along the first path. If not, do not adjust the first path. S6: If so, the first path is adaptively yawed according to the rebound amplitude and rebound direction to obtain the second path. The hair clipper is controlled to move along the second path to avoid excessive cutting of the hair.

[0041] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention 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 various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps. 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 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, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0042] 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 specification.

[0043] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. An automatic hair cutting method characterized by, The method comprises: S1: before starting haircut, collecting images of the hair region to be trimmed; S2: identifying the hair features and natural orientation of each sub-region of the hair region according to the collected images; S3: after starting haircut, monitoring the position of the haircut comb, so as to determine the rebound amplitude and rebound direction of the combed hair according to the hair features and natural orientation of the position where the haircut comb is located; S4: monitoring the real-time position of the haircut device, and generating a first path of the haircut device moving from the real-time position to the combed hair; S5: judging whether there is a risk of excessive cutting of the hair if the haircut device moves according to the first path according to the determined rebound amplitude and rebound direction, and if not, the first path is not adjusted; S6: if yes, the first path is adaptively adjusted according to the rebound amplitude and rebound direction to obtain a second path, and the haircut device is controlled to move according to the second path, so as to avoid excessive cutting of the hair.

2. The method of claim 1, wherein, The collected images include two-dimensional images and corresponding depth images, so as to obtain three-dimensional point cloud data of the hair region; Identifying the hair features and natural orientation of each sub-region of the hair region according to the collected images comprises: dividing the hair region into a plurality of sub-regions based on the three-dimensional point cloud data of the head; For each sub-region, the corresponding hair length L, hair density , hair diameter d are calculated; for each sub-region, determining the natural growth direction vector of each single hair in the sub-region, and then determining the natural orientation of the hair in the sub-region according to the direction vectors of all the hairs.

3. The method of claim 2, wherein, Between step S2 and step S3, it further comprises: obtaining a target hairstyle of the user, so as to retrieve a hairstyle model of the target hairstyle from a hairstyle library, and adaptively adjusting the size of the hairstyle model according to the size of the head region of the user to adapt to the head shape of the user; comparing the hairstyle model with the point cloud model of the hair region of the user, and then determining the target length of cutting required at each position of the hair region.

4. The method of claim 3, wherein, Determining the rebound amplitude and rebound direction of the combed hair according to the hair features and natural orientation of the position where the haircut comb is located comprises: monitoring the contact pressure between the comb teeth of the haircut comb and the hair to determine whether the haircut comb combs the hair; when the haircut comb combs the hair, determining the position of the combed hair; determining the sub-region where the position is located, and retrieving the hair features of the sub-region; calculating the rebound amplitude of the combed hair according to the contact pressure and the hair features in the sub-region; calculating the rebound direction of the combed hair according to the combed direction of the hair and the contact pressure.

5. The method of claim 4, wherein, Generating a first path of the haircut device moving from the real-time position to the combed hair comprises: retrieving the target length of cutting required for the combed hair; determining a target cutting section of the target length on the hair with the top end of the combed hair as the starting point, wherein the end point of the target cutting section is the cutting point of the hair; generating a fitting line of each cutting point, and taking the midpoint of the fitting line as a target point; generating all paths from the real-time position of the haircut device to the target point; filtering out the paths that do not pass through other hairs from the generated paths, and selecting the shortest path as the first path from the filtered paths.

6. The method of claim 5, wherein, Judging whether there is a risk of excessive cutting of the hair if the haircut device moves according to the first path according to the determined rebound amplitude and rebound direction comprises: For each of the combed hairs, determining a vertex coordinate of the hair as a first coordinate; determining a coordinate of an intersection between the hair and a blade of the hair cutter as a second coordinate when the hair cutter moves along a first path to the hair; determining a path direction of a position of the hair cutter on the first path when the hair cutter moves along the first path to the hair; calculating a cutting length of the hair when the hair cutter moves along the first path to the hair according to the determined first coordinate, second coordinate and path direction; judging whether more than a set proportion of the combed hairs have a cutting length greater than a corresponding target length, and if so, there is a risk of over-cutting the hair, otherwise, there is no risk of over-cutting the hair.

7. The method of claim 6, wherein, The adaptive yaw adjustment of the first path according to the rebound amplitude and rebound direction comprises: determining an offset direction away from the rebound direction for the hair cutter; determining a basic offset according to the rebound amplitude and an included angle between the rebound direction and a direction angle of the first path, and determining a compensation offset according to a difference between the cutting length and the target length, so as to obtain a target offset by adding the basic offset and the compensation offset; determining an offset reference point of the first path, and offsetting each point on a section of the first path after the offset reference point by the target offset along the offset direction to obtain a second path.

8. An automatic hair cutting apparatus characterized by comprising: The device comprises: an image acquisition module configured to acquire an image of a hair region to be trimmed before starting the haircut; a first processing module configured to identify hair features and natural orientations of each sub-region of the hair region according to the acquired image; a second processing module configured to monitor a position of a haircut comb after starting the haircut, so as to determine a rebound amplitude and a rebound direction of combed hairs according to hair features and natural orientations of the position where the haircut comb is located; a third processing module configured to monitor a real-time position of the hair cutter and generate a first path of the hair cutter moving from the real-time position to the combed hairs; a judging module configured to judge whether there is a risk of over-cutting the hair when the hair cutter moves along the first path according to the determined rebound amplitude and rebound direction, and if not, the first path is not adjusted; a fourth processing module configured to, if so, adaptively yaw adjust the first path according to the rebound amplitude and rebound direction to obtain a second path, and control the hair cutter to move along the second path, so as to avoid over-cutting the hair.

9. An automatic hair cutting system characterized by, The system comprises: a visual device configured to acquire an image of a hair region; a haircut device connected with the visual device and configured to perform the automatic haircut method according to any one of claims 1-7.

Images