Toothbrush head test method, apparatus, and system, medium, and program product

Abstract

The present disclosure provides a toothbrush head test method, apparatus, and system, a storage medium, and a program product. The toothbrush head test method comprises: acquiring an image to be processed, said image comprising a bristle planting region of a toothbrush head to be tested, and the bristle planting region comprising a plurality of bristle planting holes; performing region segmentation on the bristle planting region of the image to obtain a plurality of brush head partitions comprised in the bristle planting region, the plurality of brush head partitions corresponding to different cleaning effects; determining the bristle planting area of all of the bristle planting holes comprised in each brush head partition by means of performing image detection on each brush head partition; and on the basis of the bristle planting area corresponding to each brush head partition, determining a test result corresponding to the brush head. The cleaning effect of the toothbrush head can be objectively evaluated, thereby improving the reliability and guidance of brush head testing.

Description

Methods, devices, systems, media, and procedures for testing toothbrush heads.

[0001] Cross-referencing

[0002] This disclosure claims priority to Chinese Patent Application No. 202410667712.2, filed on May 27, 2024, entitled "Method and Apparatus for Detecting Toothbrush Heads", the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to the field of image processing technology, specifically to a method, apparatus, system, storage medium, and computer program product for detecting toothbrush heads. Background Technology

[0004] Toothbrushes are the most commonly used oral hygiene tools. A toothbrush's cleaning power is a crucial indicator of its oral care capabilities; effective cleaning can prevent or alleviate oral diseases. However, current technologies for testing toothbrush cleaning power primarily rely on human trials, which depend heavily on the tester's subjective experience and lack objective and effective evaluation standards. This results in inconsistent toothbrush cleaning effectiveness, impacting people's oral health. Summary of the Invention

[0005] This disclosure provides a method, apparatus, system, storage medium, and computer program product for detecting toothbrush heads.

[0006] In a first aspect, embodiments of this disclosure provide a method for detecting a toothbrush head, comprising:

[0007] Acquire an image to be processed, the image to be processed including the bristle implantation area of ​​the toothbrush head to be detected, the bristle implantation area including multiple bristle implantation holes;

[0008] The hair-planting area of ​​the image to be processed is segmented to obtain multiple brush head partitions included in the hair-planting area, wherein the cleaning effects of the multiple brush head partitions are different.

[0009] By performing image detection on each brush head partition, the hair implantation area of ​​all hair implantation holes included in each brush head partition is determined.

[0010] The detection result corresponding to each brush head is determined based on the bristle area corresponding to each brush head partition.

[0011] In some implementations, the hair-planting region of the image to be processed is segmented to obtain multiple brush head partitions comprising the hair-planting region, including:

[0012] Image detection is performed on the image to be processed to determine the region image corresponding to the bristle-planting area of ​​the brush head;

[0013] The region image is input into a pre-trained region segmentation network to obtain the multiple brush head partitions output by the region segmentation network.

[0014] In some implementations, the hair implantation area of ​​all hair implantation holes included in each brush head partition is determined by performing image detection on each brush head partition, including:

[0015] For any brush head partition, perform image detection on the brush head partition to obtain the hair implantation holes included in the brush head partition;

[0016] The bristle implantation holes in the brush head partition are classified to obtain the number of bristle implantation holes and the area of ​​a single bristle implantation hole for each category.

[0017] For any category, determine the total area of ​​the hair implantation holes corresponding to that category based on the number of hair implantation holes and the area of ​​a single hair implantation hole;

[0018] The hair implantation area is obtained by summing the total area of ​​the hair implantation holes corresponding to all categories.

[0019] In some embodiments, the plurality of brush head partitions include a first region located at the front and rear ends of the brush head, a second region located on both sides of the brush head, and a third region located in the middle of the brush head.

[0020] In some implementations, the detection result corresponding to the brush head is determined based on the bristle area corresponding to each brush head partition, including:

[0021] The total bristle area of ​​the brush head is obtained by summing the bristle areas corresponding to each brush head section.

[0022] The detection result corresponding to the brush head is determined based on the ratio of the bristle area of ​​each brush head zone to the total bristle area, the ratio of the bristle area of ​​the second region to the sum of the bristle areas of other brush head zones, and the ratio of the preset pressure to the total bristle area.

[0023] In some implementations, acquiring the image to be processed includes:

[0024] Acquire an initial image of the bristle-planting area of ​​the brush head;

[0025] The initial image is segmented to obtain the image to be processed.

[0026] Secondly, embodiments of this disclosure provide a toothbrush head detection device, comprising:

[0027] The image acquisition module is configured to acquire an image to be processed, the image to be processed including the bristle-planting area of ​​the toothbrush head to be detected, the bristle-planting area including multiple bristle-planting holes;

[0028] The region segmentation module is configured to segment the hair-planting region of the image to be processed to obtain multiple brush head partitions included in the hair-planting region, wherein the multiple brush head partitions correspond to different cleaning effects.

[0029] The area determination module is configured to determine the hair implantation area of ​​all hair implantation holes included in each brush head partition by performing image detection on each brush head partition.

[0030] The result determination module is configured to determine the detection result corresponding to the brush head based on the bristle area corresponding to each brush head partition.

[0031] In some implementations, the region segmentation module is configured to:

[0032] Image detection is performed on the image to be processed to determine the region image corresponding to the bristle-planting area of ​​the brush head;

[0033] The region image is input into a pre-trained region segmentation network to obtain the multiple brush head partitions output by the region segmentation network.

[0034] In some implementations, the area determination module is configured to:

[0035] For any brush head partition, perform image detection on the brush head partition to obtain the hair implantation holes included in the brush head partition;

[0036] The bristle implantation holes in the brush head partition are classified to obtain the number of bristle implantation holes and the area of ​​a single bristle implantation hole for each category.

[0037] For any category, determine the total area of ​​the hair implantation holes corresponding to that category based on the number of hair implantation holes and the area of ​​a single hair implantation hole;

[0038] The hair implantation area is obtained by summing the total area of ​​the hair implantation holes corresponding to all categories.

[0039] In some embodiments, the plurality of brush head partitions include a first region located at the front and rear ends of the brush head, a second region located on both sides of the brush head, and a third region located in the middle of the brush head.

[0040] In some implementations, the result determination module is configured to:

[0041] The total bristle area of ​​the brush head is obtained by summing the bristle areas corresponding to each brush head section.

[0042] The detection result corresponding to the brush head is determined based on the ratio of the bristle area of ​​each brush head zone to the total bristle area, the ratio of the bristle area of ​​the second region to the sum of the bristle areas of other brush head zones, and the ratio of the preset pressure to the total bristle area.

[0043] In some implementations, the image acquisition module is configured to:

[0044] Acquire an initial image of the bristle-planting area of ​​the brush head;

[0045] The initial image is segmented to obtain the image to be processed.

[0046] Thirdly, embodiments of this disclosure provide a toothbrush head detection system, comprising:

[0047] Image acquisition device, used to acquire images of the toothbrush head to be inspected;

[0048] Processor; and

[0049] The memory stores computer instructions that cause the processor to perform the method described in any of the above embodiments.

[0050] Fourthly, embodiments of this disclosure provide a storage medium storing computer instructions for causing a computer to execute the methods described in any of the above embodiments.

[0051] Fifthly, this disclosure provides a computer program product that implements the methods described in any of the above embodiments when running. Attached Figure Description

[0052] To more clearly illustrate the technical solutions in the specific embodiments of this disclosure or the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0053] Figure 1 is a schematic diagram of the application scenario of the detection method of some embodiments of this disclosure.

[0054] Figure 2 is a structural block diagram of a detection system according to some embodiments of the present disclosure.

[0055] Figure 3 is a flowchart of a detection method according to some embodiments of this disclosure.

[0056] Figure 4 is a schematic diagram of the detection method according to some embodiments of this disclosure.

[0057] Figure 5 is a schematic diagram of the detection method according to some embodiments of this disclosure.

[0058] Figure 6 is a schematic diagram of the detection method according to some embodiments of this disclosure.

[0059] Figure 7 is a flowchart of a detection method according to some embodiments of this disclosure.

[0060] Figure 8 is a flowchart of a detection method according to some embodiments of this disclosure.

[0061] Figure 9 is a flowchart of a detection method according to some embodiments of this disclosure.

[0062] Figure 10 is a schematic diagram of the principle of the detection method according to some embodiments of this disclosure.

[0063] Figure 11 is a flowchart of a detection method according to some embodiments of this disclosure.

[0064] Figure 12 is a structural block diagram of a detection device according to some embodiments of the present disclosure.

[0065] Figure 13 is a structural block diagram of a detection system according to some embodiments of the present disclosure. Detailed Implementation

[0066] The technical solutions of this disclosure will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this disclosure, not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without inventive effort are within the scope of protection of this disclosure. Furthermore, the technical features involved in the different embodiments of this disclosure described below can be combined with each other as long as they do not conflict with each other.

[0067] Toothbrushes are the most commonly used oral hygiene tools, and with the development of electronic technology, electric toothbrushes are increasingly replacing ordinary toothbrushes, becoming the preferred dental cleaning tool. However, regardless of whether it's a regular or electric toothbrush, the cleaning effect on teeth is an important indicator of its cleaning power. Highly effective cleaning can thoroughly clean teeth and reduce the risk of oral diseases.

[0068] The brush head is a crucial component affecting the cleaning power and comfort of a toothbrush. Brush heads are generally formed by embedding flexible bristles into a rigid material. The material, distribution, length, softness, hardness, and diameter of the bristles all affect the oral cleaning effect and comfort of the toothbrush. Therefore, evaluating and testing the cleaning effect of toothbrush heads is a necessary means to improve the oral care capabilities of toothbrushes.

[0069] In related technologies, there are no objective testing and evaluation methods for toothbrush heads. Most testing work only focuses on the toothbrush as a whole, and generally adopts human efficacy tests. For example, a certain number of test subjects are selected to brush their teeth with the toothbrush under test during the test period. The cleaning effect on teeth is observed and combined with the test subjects' subjective feelings to determine the cleaning ability of the toothbrush.

[0070] On the one hand, such human efficacy testing is costly and time-consuming, and heavily reliant on the tester's subjective experience, making it unsuitable for the product development stage. Furthermore, manufacturers lack objective testing standards during the R&D phase, leading to suboptimal product performance. On the other hand, this testing is conducted on the toothbrush as a whole. For electric toothbrushes, the brush head and handle are typically separate and replaceable. If users want to replace the brush head, the lack of industry standards for objectively evaluating its cleaning effectiveness leaves them without clear and effective guidance when choosing a new one.

[0071] Based on the deficiencies in the aforementioned related technologies, this disclosure provides a method, apparatus, electronic device, storage medium, and computer program product for testing toothbrush heads. The aim is to provide an automated testing process that objectively quantifies the cleaning effect of toothbrush heads. By dividing the bristle-planting area of ​​the brush head into zones and quantifying the cleaning power of the pressure head based on the bristle-planting area of ​​each zone, the oral care capability of the brush head can be tested objectively and efficiently.

[0072] Figure 1 illustrates the contact process between the toothbrush head and the human mouth during brushing. As shown in Figure 1, the contact positions of the bristles at different locations on the brush head vary during brushing. For example, bristles on the sides of the toothbrush primarily contact the tooth surfaces and gum line, while bristles in the middle primarily contact the gingival sulcus. Furthermore, the effects of bristles at different locations on the brush head also differ. For instance, bristles at the front and back of the toothbrush mainly clean plaque and interdental spaces, while bristles in the middle provide deeper cleaning for whitening, and bristles on the sides massage and care for the gums.

[0073] In practical use, different bristles with varying numbers, lengths, and softness / rigidity can be incorporated into different parts of the brush head to achieve different cleaning effects on teeth. For example, some bristles may be designed to effectively remove food debris adhering to the tooth surface, while others may be designed to effectively remove residue between teeth without insufficient cleaning power due to excessive bristle bending. Still others may be designed to effectively remove plaque and other tartar buildup. Therefore, different bristle configurations can be strategically placed in different parts of the toothbrush head for targeted cleaning.

[0074] As can be seen, in actual oral cleaning, the bristles in different sections of the toothbrush head have different functions and roles. Therefore, the rationality of the brush head's zoning is one of the important indicators of its cleaning effect. If the brush head's zoning is not set up reasonably, it will not only fail to effectively help users clean their mouths, but will also reduce the user's brushing comfort and cleaning effect, resulting in a poor user experience.

[0075] Therefore, in this embodiment, the cleaning effect of the toothbrush head is objectively quantified and expressed by performing zoned testing on the brush head and based on the bristle area of ​​different zones and the area ratio between them. This yields target parameters reflecting the oral care capabilities of the brush head as the test results. These target parameters can serve as an industry standard, providing users with intuitive and efficient guidance when selecting toothbrush products, and also providing objective guidance for manufacturers during the R&D stage, effectively mitigating risks upfront and creating an effective closed loop for product iteration.

[0076] Figure 2 shows a structural block diagram of the detection system in some embodiments of the present disclosure. Referring to Figure 2, the detection system of the present disclosure includes a brush head 100 of a toothbrush to be detected, an image acquisition device 200, and a controller 300.

[0077] Brush head 100 refers to a brush head whose oral care effect needs to be tested and evaluated. Brush head 100 can be a regular toothbrush head or an electric toothbrush head, and this disclosure does not limit it. Brush head 100 includes a body and bristles embedded in the body. The body is generally made of plastic or rubber, and the bristles are generally made of flexible materials such as nylon, polyester, and plant fibers.

[0078] The image acquisition device 200 is used to acquire images or video streams from the brush head 100. Frames from the acquired images or video streams can be used as the images to be processed as described in this disclosure. The image acquisition device 200 may include one or more industrial cameras, depth cameras, infrared cameras, etc., and this disclosure does not limit it.

[0079] The controller 300 refers to the processing core of the detection system, which may include a processor and memory. The processor can be of any type and has one or more processing cores. It can perform single-threaded or multi-threaded operations, used to parse instructions to perform operations such as acquiring data, performing logical operations, and sending out processing results.

[0080] The memory may include a non-volatile computer-readable storage medium, such as at least one disk storage device, flash memory device, distributed storage device remotely located relative to the processor, or other non-volatile solid-state storage device. The memory may have a program storage area for storing non-volatile software programs, non-volatile computer-executable programs, and modules, which the processor can call to cause the processor to execute one or more method steps. The memory may also include a volatile random access storage medium, or a storage portion such as a hard disk, as a data storage area for storing the processing results and data output by the processor.

[0081] In this embodiment, the memory stores computer-readable instructions that can be executed by the processor. When these computer-readable instructions are executed, the processor can perform the detection method described in the following embodiment. Furthermore, the memory can also store image data acquired by the image acquisition device 200.

[0082] Based on the foregoing, Figure 3 shows a flowchart of the detection method in some embodiments of this disclosure. The embodiments of this disclosure will be described below with reference to Figure 3.

[0083] As shown in Figure 3, in some embodiments, the toothbrush head detection method of this disclosure includes:

[0084] S310. Obtain the image to be processed.

[0085] In this embodiment of the disclosure, the image to be processed includes the bristle-implanted area of ​​the toothbrush head to be detected. It is understood that a toothbrush head generally consists of a body and bristles implanted on the body. For example, as shown in Figure 4, multiple bristle-implanting holes are formed on the brush head body, and a tuft of bristles can be implanted into each hole, thereby forming the brush head structure. In this embodiment of the disclosure, the area on the body used for bristle implantation is defined as the bristle-implanted area.

[0086] It is understandable that the shape, area, and distribution of the bristle implantation holes on different toothbrush heads vary, which will result in differences in the cleaning effect of the brush head after bristle implantation. The detection method of this disclosure is to perform zone detection on the bristle implantation area of ​​the brush head to achieve objective quantification of the cleaning effect of the brush head.

[0087] Referring to Figure 2, in some embodiments, an image acquisition device 200 can be used to acquire an image of the brush head 100 to obtain an image to be processed. For example, in one example, the brush head 100 can be fixed on a bracket, and then the image acquisition device 200 can be used to acquire an image of the brush head 100 from the front, and then the acquired image can be used as the image to be processed of the brush head 100.

[0088] In some embodiments, the brush head can be processed before acquiring the image to be processed from the brush head 100. For example, the bristles on the brush head 100 can be removed, leaving only the brush head body. The image acquisition device 200 then acquires the bristle holes on the brush head body to obtain the image to be processed. For example, in the example of Figure 4, because the bristles on the brush head 100 are removed, the shape and distribution of the bristle holes can be clearly displayed on the image to be processed, reducing image noise interference caused by the bristles.

[0089] In some implementations, considering that the original image acquired by the image acquisition device 200 may include not only the brush head 100 but also the background, in order to improve the image detection effect, the acquired original image can be preprocessed. For example, the original image can be segmented to remove the background area and retain only the brush head area in the image. The preprocessed brush head image can then be used as the image to be processed.

[0090] S320. Perform region segmentation on the hair-planting area of ​​the image to be processed to obtain multiple brush head partitions included in the hair-planting area.

[0091] In this embodiment of the disclosure, after obtaining the image of the brush head to be processed, image detection technology can be used to segment the bristle-planting area of ​​the brush head, dividing the bristle-planting area into multiple brush head partitions. As mentioned above, the bristle-planting area of ​​the brush head refers to the area on the brush head body used to implant bristles, for example, the area where the bristle-planting holes are distributed in Figure 4 is the bristle-planting area of ​​the brush head.

[0092] As can be understood from the above, the bristles at different positions on the brush head have different functions and roles. Therefore, in this embodiment of the disclosure, the bristle planting area can be divided into multiple functional zones based on the different zoning functions. These functional zones are also known as the brush head zoning as described in this disclosure.

[0093] For example, Figure 5 shows an example of dividing the bristle-planting area of ​​the brush head into sections. The following explanation is based on the example in Figure 5.

[0094] As shown in Figure 5, in some implementations, the bristle-embedded area of ​​the brush head can be divided into the following three brush head zones:

[0095] 1. First Area

[0096] This refers to the green areas at the front and rear ends shown in Figure 5. The bristles in this area are generally long bristles with the lowest stiffness. During brushing, the bristles in the first area are used to remove plaque attached to the tooth surface, clean the gingival sulcus, and clean between teeth, mainly playing a role in cleaning teeth.

[0097] 2. Second Area

[0098] This refers to the purple areas on the left and right sides shown in Figure 5. The bristles in this area are generally short bristles with high stiffness. During brushing, the bristles in the second area massage the gums, mainly playing a role in gum care and improving user comfort.

[0099] 3. Third Region

[0100] This refers to the orange area in the middle of Figure 5. The bristles in this area are generally the second shortest bristles with the highest stiffness. During brushing, the bristles in the third area are used to deeply clean the teeth and remove stubborn deposits such as stains, mainly to whiten the teeth.

[0101] It should be noted that the embodiment shown in Figure 5 is merely a guiding example of the testing method of this disclosure and does not limit this disclosure. Moreover, the method of dividing the bristle-planting area shown in Figure 5 was derived by the inventors through extensive experiments and theoretical research, and is a process of creative labor. In other embodiments, the zoning method for the bristle-planting area of ​​the brush head can also take other forms. For example, the bristle-planting area can be divided into more or fewer brush head partitions, which will not be elaborated upon in this disclosure.

[0102] In this embodiment of the disclosure, after obtaining the image of the brush head to be detected through the aforementioned S310, the bristle-planting area of ​​the brush head 100 can be detected and segmented by image detection technology, thereby dividing the bristle-planting area into three brush head partitions as shown in Figure 5.

[0103] In some implementations, a region segmentation network based on a deep neural network (DNN) can be used to segment the region and obtain multiple brush head partitions. The following implementations of this disclosure will describe this, but will not be detailed here.

[0104] For example, taking the image to be processed as shown in Figure 4, after segmenting the image according to the partitioning criteria shown in Figure 5, the resulting three brush head partitions are shown in Figure 6. In the example in Figure 6, the green areas at the top and bottom are the first region, the purple areas on the left and right sides are the second region, and the orange area in the center is the third region. It can be seen that each region includes multiple hair implantation holes.

[0105] S330. By performing image detection on each brush head partition, determine the hair implantation area of ​​all hair implantation holes included in each brush head partition.

[0106] As shown in Figure 6, after dividing the bristle implantation area into regions, each brush head partition includes one or more bristle implantation holes. The number and shape of the bristle implantation holes in different brush head partitions will also be different, and even the shapes of the bristle implantation holes in the same brush head partition will be different.

[0107] In this embodiment of the disclosure, it is necessary to calculate the bristle implantation area of ​​the bristle implantation holes in each brush head section. It can be understood that for a given brush head section, the larger the bristle implantation area of ​​the bristle implantation holes within that section, the more bristles are implanted in that section, and thus the higher the cleaning effect provided by the bristles in that section. Therefore, the bristle implantation area in each brush head section can be determined separately, and the bristle implantation area can be used as a quantitative indicator to objectively evaluate the cleaning effect of the brush head.

[0108] In some implementations, image processing tools such as ImageJ can be used to detect the hair implantation holes in each brush head partition based on image detection technology, thereby determining the hair implantation area of ​​the hair implantation holes included in the brush head partition.

[0109] In some implementations, when detecting the hair implantation area of ​​the hair implantation holes included in the brush head partition, image detection technology can be used to segment the image region of all hair implantation holes, and then area detection can be used to determine the hair implantation area of ​​all segmented hair implantation holes.

[0110] In other implementations, considering the complexity of the area detection algorithm, in order to improve computational efficiency, the hair implantation holes can be classified after the image region of the hair implantation holes included in the brush head partition is segmented.

[0111] For example, in the example in Figure 6, the entire brush head includes only one type of hair implantation hole, that is, all hair implantation holes have the same shape and size. Therefore, it is only necessary to calculate the number of hair implantation holes included in each brush head section and the area of ​​any hair implantation hole. By multiplying the area of ​​a single hair implantation hole by the number of hair implantation holes included in the brush head section, the hair implantation area of ​​each brush head section can be obtained.

[0112] For example, in the example in Figure 5, each brush head section includes multiple types of hair implantation holes. That is, the shape and size of the hair implantation holes in the brush head section are not completely consistent. Therefore, after classifying the hair implantation holes, we can obtain the number of hair implantation holes in each type and the area of ​​a single hair implantation hole in each type. Then, by multiplying the area of ​​a single hair implantation hole in each type by the number of hair implantation holes in that type, and finally summing all the categories, we can obtain the hair implantation area of ​​the hair implantation holes contained in the brush head section.

[0113] The process of calculating the hair implantation area of ​​each hair implantation hole in each brush division is described in the following embodiments.

[0114] S340. Determine the test result corresponding to each brush head based on the bristle area corresponding to each brush head section.

[0115] Through the aforementioned method and process, the bristle area corresponding to each brush head section can be determined. As mentioned above, since different brush head sections have different cleaning functions, after quantifying the bristle area of ​​each brush head section, the target parameter representing the cleaning effect of the brush head can be obtained by fusing and analyzing the bristle areas of each brush head section. This target parameter is the test result corresponding to the brush head to be tested.

[0116] For example, in one instance, suppose the bristle area of ​​the first region of the brush head to be tested is C, the bristle area of ​​the second region is G, and the bristle area of ​​the third region is W. Thus, the total bristle area of ​​the entire bristle-planting region is Z = C + G + W.

[0117] In some implementations, data fusion can be performed based on the ratio of the bristle area (C, G, W) of each brush head partition to the total bristle area (Z) to obtain the target parameter M, which is the detection result of the brush head.

[0118] In other implementations, based on the above examples, the ratio of the sum of the bristle areas of the second region G and other brush head partitions (C+W) can be further combined with the ratio of the bristle areas (C, G, W) of each brush head partition to the total bristle area (Z) to obtain the target parameter M, which is the detection result of the brush head.

[0119] In some other implementations, based on the above examples, the pressure P of the brush head under a preset pressure can be further combined with the ratio of the bristle area (C, G, W) of each of the aforementioned brush head partitions to the total bristle area (Z), and the ratio of the bristle area of ​​the second region G to the sum of the bristle areas of other brush head partitions (C+W) to obtain the target parameter M. The target parameter M is the detection result of the brush head.

[0120] The process of calculating the target parameter M in the above-described various implementation methods will be described in detail below.

[0121] It is worth noting that in this disclosed testing method, the bristle-embedded areas of the brush head are tested in separate zones. The bristle area of ​​different zones is used as a quantitative indicator to objectively quantify and express the cleaning ability of different brush head zones, resulting in target parameters reflecting the brush head's oral care capabilities as the test results. This provides an automated process for objectively evaluating the oral care effect of brush heads, eliminating the need for human efficacy testing, standardizing the brush head testing process, and improving testing efficiency. Furthermore, the test results can serve as an industry standard, providing users with intuitive and efficient guidance for product selection, and also provide objective guidance for manufacturers during the R&D phase, enabling targeted improvements to the brush head structure, effectively mitigating risks upfront, and creating an effective closed loop for product iteration.

[0122] As shown in Figure 7, in some embodiments, the detection method of this disclosure, in order to acquire the image to be processed, includes:

[0123] S311. Acquire the initial image of the bristle-embedded area of ​​the brush head.

[0124] S312. Perform background segmentation on the initial image to obtain the image to be processed.

[0125] In conjunction with the detection system shown in Figure 1, in some embodiments, an initial image can first be obtained by using an image acquisition device 200 to acquire an image of the brush head 100.

[0126] It is understandable that the initial image may contain not only the brush head, but also irrelevant background noise. In order to reduce the interference of background noise on subsequent image detection, the initial image can be segmented to obtain an image to be processed that retains only the foreground brush head.

[0127] In some implementations, an image segmentation model can be used to segment the background of the initial image. The image segmentation model can be, for example, SAM (Segment Anything Model), FCN (Fully Convolutional Network), SegNet, etc., and this disclosure does not limit it.

[0128] As can be seen from the above, in this embodiment of the disclosure, the image to be processed is obtained by background segmentation of the original image, thereby reducing background noise interference and providing a data foundation for subsequent image detection.

[0129] In some implementations, when performing region segmentation on the hair-planting area of ​​the image to be processed, the partitioning rules for each brush head partition can be predefined. The partitioning rules can include the position and range of each brush head partition. Then, the image to be processed is segmented based on the partitioning rules to obtain each brush head partition.

[0130] For example, taking the scenario shown in Figure 5 as an example, the brush head partition includes three partitions: the first region, the second region, and the third region. The partitioning rules for each brush head partition can be as follows:

[0131] The first region can be determined, for example, by cutting a specific proportion of the toothbrush head along the length of the toothbrush handle. For instance, assuming a toothbrush head is 3 cm long along the handle, the first region could be the front and back 10% of the toothbrush head, or 0.3 cm. Of course, other proportions are also possible, such as selecting the front 10% and back 5% as the first region respectively. The selected proportion can be adjusted adaptively as needed.

[0132] The second region can be determined by cutting a specific portion of the toothbrush head along the length direction of the toothbrush handle and / or along the width direction of the toothbrush handle. In the aforementioned embodiment, the front and rear 10% of the toothbrush head has been determined as the first region. At the same time, the front and rear 10% of the brush head can be cut off along the length direction of the toothbrush handle as a candidate region for the second region. Then, in the candidate region, a specific portion of the toothbrush head can be cut along the width direction of the toothbrush handle as the second region. For example, 15% of both sides of the toothbrush head can be cut along the width direction of the toothbrush handle as the second region.

[0133] The third region can be determined by cutting off the portion of the toothbrush head that is in the center in both the length and width directions. For example, in the previous embodiment, the front and back 10% of the toothbrush head has been determined as the first region. At the same time, the front and back 10% of the brush head along the length direction of the toothbrush handle can be cut off as the candidate region for the second region. Then, in the candidate region, the area of ​​the brush head excluding the second region can be used as the third region.

[0134] Of course, the above partitioning rules for the brush head partition are merely illustrative examples. In other embodiments, other forms of partitioning rules can be defined for each partition, and this disclosure does not impose any restrictions on this.

[0135] After defining the above partitioning rules, when performing image detection on the image to be processed, image segmentation can be performed based on the above partitioning rules to separate each brush head partition. For example, as shown in Figures 5 and 6, the same brush head partition is represented by a color block of the same color, and different brush head partitions are represented by color blocks of different colors.

[0136] In the above embodiments, partitioning rules are predefined for toothbrush head partitioning, so that during image processing, image segmentation only needs to be performed according to the predefined partitioning rules. In some embodiments, considering the diverse bristle layouts of different brush heads, in order to further improve the accuracy and effect of brush head partitioning, this disclosure embodiment can use a region segmentation network based on a DNN network model for region segmentation, as described below with reference to Figure 8.

[0137] As shown in Figure 8, in some embodiments, the detection method of this disclosure, in the process of region segmentation of the hair-planted area, includes:

[0138] S321. Perform image detection on the image to be processed to determine the region image corresponding to the bristle implantation area of ​​the brush head.

[0139] S322. Input the region image into a pre-trained region segmentation network to obtain multiple brush head partitions output by the region segmentation network.

[0140] It is understood that the image to be processed is an image containing the entire brush head 100. The detection method of this embodiment only involves partitioning the bristle-planting area of ​​the brush head, so the image area outside the bristle-planting area can be removed.

[0141] For example, in some implementations, image segmentation techniques can be used to process the image to be processed, segmenting the bristle-planting region of the brush head 100 to obtain the corresponding region image. Specifically, an image segmentation model can be used to segment the bristle-planting region. The image segmentation model can be SAM, FCN, SegNet, etc., and this disclosure does not limit it. For example, in one example, the bristle-planting region image obtained by performing image segmentation on the bristle-planting region of the brush head is shown in Figure 4.

[0142] After obtaining the image of the hair-implanting area, a deep neural network (DNN) can be used to segment the area image, thereby obtaining multiple brush head partitions.

[0143] In some implementations, a region segmentation network can be pre-built. This network can employ a SAM-based architecture, comprising an input layer, intermediate layers, and an output layer. The input layer is one or more convolutional layers used to extract high-dimensional image features. The intermediate layers may include activation layers, pooling layers, etc., to enhance the network's non-linearity and aid in better learning and training. The output layer may include fully connected layers, with the number of nodes matching the number of brush head partitions. These layers establish fully connected relationships between image features and output the final partitioning result.

[0144] The network structure of the region segmentation network can be understood and fully implemented by those skilled in the art based on the above and in combination with relevant technologies, and will not be described in detail here.

[0145] After constructing the region segmentation network, it can be trained using a training sample set. The training sample set can include multiple sample data points, each containing a brush head image and pre-annotated partition labels for the brush head image. These partition labels can be obtained manually. Therefore, when training the network using the training sample set, taking any single sample data point as an example, inputting it into the region segmentation network will yield the network's predicted output. Then, a pre-constructed loss function is used to calculate the loss between the output and the partition labels. The network parameters are then optimized and adjusted using backpropagation. This iterative process continues until the region segmentation network converges, thus completing the network training process.

[0146] The network training process for region segmentation networks can be understood and fully implemented by those skilled in the art based on the above and in combination with relevant technologies, and will not be elaborated further in this disclosure.

[0147] After the network training is complete, the region segmentation network can be used to partition the hair-planting area. Specifically, the previously obtained hair-planting area image is input into the region segmentation network, which then predicts multiple brush head partitions.

[0148] In some implementations, multiple brush head partitions output by the region segmentation network can be marked with selection boxes of different colors. For example, in the examples of Figures 5 and 6, selection boxes of the same color correspond to the same type of brush head partition. In the examples of Figures 5 and 6, the brush head partitions are divided into a green first region, a purple second region, and an orange third region. The partitioning of the first, second, and third regions is based on the aforementioned criteria and will not be repeated here.

[0149] As can be seen from the above, in this embodiment of the invention, the use of a deep neural network to segment the bristle-planting area of ​​the brush head is more efficient than manual segmentation and does not require human intervention, thus achieving a fully automated detection process.

[0150] As shown in Figure 9, in some embodiments, the detection method of this disclosure, the process of determining the bristle area of ​​each brush head partition, includes:

[0151] S331. For any brush head partition, perform image detection on the brush head partition to obtain the hair implantation pores included in the brush head partition.

[0152] S332. Classify the bristle holes included in the brush head partition to obtain the number of bristle holes and the area of ​​a single bristle hole for each category.

[0153] S334. For any category, determine the total area of ​​the hair implantation holes corresponding to that category based on the number of hair implantation holes and the area of ​​a single hair implantation hole.

[0154] S334. Sum the total area of ​​the hair implantation holes corresponding to all categories to obtain the hair implantation area.

[0155] In some alternative embodiments, the implantation area can be obtained by statistically analyzing the bristles in the implantation holes. For example, the number of bristles in a specific implantation hole can be obtained, and the bristle area of ​​the implantation hole can be determined based on the arrangement density of the bristles. In some embodiments, the bristles arranged in the same implantation hole belong to the same region, that is, the same type of bristles are arranged in the same implantation hole. For example, for a certain implantation hole, all the bristles in that hole are used as bristles in a first region. By obtaining the number of bristles in that implantation hole, the implantation area of ​​the implantation hole used as the first region can be determined based on the arrangement density of the bristles. Furthermore, the region corresponding to the implantation hole can be determined based on the bristles in the implantation hole. For example, a mapping relationship between regions and bristle characteristics can be established. For example, if the stiffness of the bristles is within range A, the bristles are considered to be used in the first region. If the stiffness of the bristles is within range B (which differs from range A), the bristles are considered to be used in the second region. Similarly, if the stiffness of the bristles is within range C, the bristles are considered to be used in the third region, and so on. Therefore, by sampling and measuring the bristles in the bristle holes, the region corresponding to the bristle hole providing the corresponding function can be determined according to mapping rules. The area of ​​the region can then be determined based on factors such as bristle density. In some embodiments, the bristles arranged in the same bristle hole belong to different regions; that is, different types of bristles are arranged in the same bristle hole. For example, for a certain bristle hole, some of the bristles in that hole are used as bristles in the first region, and another part is used as bristles in the second region. In this case, by obtaining the bristle type and quantity of the bristles in the bristle hole respectively, and based on the bristle arrangement density of the bristles in the bristle hole, the bristle area of ​​the corresponding region (e.g., the first region and the second region) can be determined respectively. By determining the bristle characteristics of the bristle holes and calculating the bristle area based on the number of bristles and the bristle density, the total area of ​​the bristle area of ​​the brush head and the bristle area of ​​each brush head section can be obtained.

[0156] In one example, after dividing the bristle-planting area of ​​the brush head into regions, the resulting three brush head partitions can be shown in Figure 10.

[0157] For ease of explanation, the process of calculating the bristle area of ​​each brush head section will be explained below using the orange third area as an example.

[0158] First, image detection can be performed on the third region to detect all the hair implantation holes included in the third region. As shown in Figure 10, the shapes and sizes of the hair implantation holes in the third region are not exactly the same, so it is necessary to classify the hair implantation holes.

[0159] For example, in some implementations, an image classification network can be pre-trained and used to classify the hair implantation holes in the third region. Hair implantation holes with the same shape and size are classified into the same type, while hair implantation holes with different shapes or sizes are classified into different types. For example, as shown in Figure 10, the hair implantation holes in the third region include two types: type 4 and type 5. Type 4 has 12 hair implantation holes, and type 5 has 3 hair implantation holes.

[0160] Furthermore, in this embodiment, the image classification network can further perform area detection for each type of hair implantation hole, thereby determining the area of ​​a single hair implantation hole for each type. Since the image classification network only needs to detect the area of ​​a single hair implantation hole for each type, it is more efficient than detecting the area of ​​all hair implantation holes.

[0161] In the embodiment shown in Figure 10, the detected bristle pore information for each brush head is shown in Table 1 below:

[0162] Table 1

[0163] In the example in Table 1, taking the third region as an example, its corresponding hair implantation area S3 is the sum of the areas of all types of hair implantation holes, expressed as S3 = 12*f + 3*g. Similarly, the hair implantation area S1 of the first region is expressed as S1 = 4*a + 2*b + 2*c, and the hair implantation area S2 of the second region is expressed as S2 = 4*d + 2*e.

[0164] It can be understood that the hair implantation area S1 in the first region represents the total area of ​​all hair implantation holes in the first region, the hair implantation area S2 in the second region represents the total area of ​​all hair implantation holes in the second region, and the hair implantation area S3 in the third region represents the total area of ​​all hair implantation holes in the third region.

[0165] As can be seen from the above, in this embodiment of the disclosure, the efficiency of calculating the hair implantation area is improved by classifying the hair implantation holes based on their shape and size.

[0166] In this embodiment of the disclosure, after obtaining the bristle area of ​​each brush head partition, the target parameter M of the brush head can be calculated based on the bristle area of ​​each brush head partition, and the target parameter M is the detection result of the brush head.

[0167] In some implementations, the ratio of the bristle area of ​​each brush head partition to the total bristle area can be calculated separately, and then the ratios of each brush head partition can be combined to obtain the target parameter M. For example, in one example, the bristle area and area ratio of different brush head partitions are shown in Table 2 below:

[0168] Table 2

[0169] In some implementations, the target parameter M is represented as:

[0170] In formula (1), This represents the first ratio of the hair-planted area in the first region to the total hair-planted area. This represents the second ratio of the hair-planted area in the second region to the total hair-planted area. This represents the third ratio of the hair-planted area in the third region to the total hair-planted area. f[] represents the fusion function, used to perform fusion calculations on the first, second, and third ratios.

[0171] In some implementations, preset coefficient values ​​can be pre-set for the first ratio, the second ratio, and the third ratio. Then, the first ratio, the second ratio, and the third ratio are fused according to the preset coefficient values ​​to obtain the target parameter M, which can be expressed as:

[0172] In formula (2), α represents the coefficient value corresponding to the first ratio, β represents the coefficient value corresponding to the second ratio, and ω represents the coefficient value corresponding to the third ratio. In this embodiment, the specific values ​​of α, β, and ω are not limited. Those skilled in the art can select the corresponding coefficient values ​​according to the specific application scenario, and this disclosure will not elaborate further. Combining formulas (1) and (2), it can be understood that by partitioning and quantifying the different functional areas of the brush head, the target parameter M obtained can reflect the rationality of the brush head partitioning and the cleaning effect of the brush head.

[0173] In some implementations, in addition to the above, the target parameter M can further consider user comfort and the proportion of special functions. Comfort refers to the user's feeling of comfort when brushing their teeth using the brush head, which can be represented by the pressure of the brush head at a preset pressure level. Special functions are relative to the basic functions; the basic function of the brush head is oral cleaning, while special functions, such as gum massage, are provided. The process of calculating the target parameter M is explained below with reference to Figure 11.

[0174] As shown in Figure 11, in some embodiments, the detection method of this disclosure, the process of determining the brush head detection result, includes:

[0175] S341. Sum the bristle areas corresponding to each brush head section to obtain the total bristle area of ​​the brush head.

[0176] S342. Determine the corresponding test result for each brush head based on the ratio of the bristle area of ​​each brush head zone to the total bristle area, the ratio of the bristle area of ​​the second region to the sum of the bristle areas of other brush head zones, and the ratio of the preset pressure to the total bristle area.

[0177] It is worth noting that the force applied to a toothbrush by a user while brushing is generally relatively constant. Under the same pressure, the bristle area of ​​the brush head varies, resulting in different pressures applied to the user's oral cavity, and consequently, differences in brushing comfort. Therefore, in some embodiments of this disclosure, the pressure parameter of the brush head under a preset pressure can be further integrated into the target parameter M to quantify the comfort of the brush head.

[0178] Furthermore, as mentioned above, the primary function of the bristles in the first and second regions is to clean the teeth, and their effect constitutes the basic function of the brush head. The primary function of the bristles in the second region is to care for and massage the gums; therefore, the effect produced by the bristles in the second region can be defined as the special function of the brush head. Based on this, in some embodiments of this disclosure, the ratio of the special function to the basic function of the brush head can be further integrated into the target parameter M to quantify the function of the brush head.

[0179] Specifically, in some implementations, the target parameter M is represented as:

[0180] In formula (3), This represents the first ratio of the hair-planted area in the first region to the total hair-planted area. This represents the second ratio of the hair-planted area in the second region to the total hair-planted area. This represents the third ratio of the hair-planted area in the third region to the total hair-planted area.

[0181] This represents the ratio of the hair-planting area in the second region to the sum of the hair-planting areas in the first and third regions. Based on the aforementioned implementation methods, it can be understood that the second region is a special-effect region, while the first and third regions are basic-effect regions. This reflects the ratio of the brush head's special functions to its basic functions.

[0182] F represents the preset pressure, and its specific value can be given according to the needs of the scenario. For example, in one example, by collecting data on the force applied by a large number of users when brushing their teeth, F can be set to 200N. Z is the total bristle area of ​​the brush head, that is, the maximum contact area between the brush head and the user's mouth. According to the pressure formula, ... This indicates the pressure applied by the brush head to the user's mouth, reflecting the comfort level of the brush head.

[0183] f[] represents the fusion function, used to perform fusion calculations on various parameters. In some implementations, preset coefficient values ​​can be set for each parameter in advance, and then the parameters in formula (3) can be fused according to the preset coefficient values ​​to obtain the target parameter M, which can be expressed as:

[0184] In formula (4), α, β, ω, θ, In this embodiment of the disclosure, the coefficient values ​​are represented as α, β, ω, θ. The specific value of is not limited, and those skilled in the art can select the appropriate coefficient value according to the specific application scenario. This disclosure will not elaborate on this further.

[0185] Combining formulas (3) and (4), it can be understood that by further integrating the ratio of special effects to basic effects, as well as the pressure applied to the user's oral cavity by the brush head during brushing, the proportion of special effects and comfort of the brush head can be further reflected, establishing an objective evaluation standard that is closer to real-world usage conditions, and providing objective data and theoretical support for the standardized testing of brush heads.

[0186] As described above, in this embodiment, by performing zoned testing on the bristle-embedded areas of the brush head, and using the bristle-embedded area of ​​different brush head zones as a quantitative indicator, the cleaning ability of different brush head zones is objectively quantified and expressed, resulting in target parameters reflecting the brush head's oral care capabilities as the brush head's test results. This provides an automated process for objectively evaluating the oral care effect of brush heads, eliminating the need for human efficacy testing, standardizing the brush head testing process, and improving testing efficiency. Furthermore, the test results can serve as an industry-wide standard, providing users with intuitive and efficient guidance for product selection, and also provide objective guidance for manufacturers during the R&D phase, enabling targeted improvements to the brush head structure, effectively mitigating risks upfront, and creating an effective closed loop for product iteration.

[0187] In some embodiments, this disclosure provides a toothbrush head detection device, as shown in FIG12, the detection device comprising:

[0188] Image acquisition module 10 is configured to acquire an image to be processed, the image to be processed including the bristle implantation area of ​​the toothbrush head to be detected, the bristle implantation area including multiple bristle implantation holes;

[0189] The region segmentation module 20 is configured to segment the hair-planting region of the image to be processed to obtain multiple brush head partitions included in the hair-planting region, wherein the multiple brush head partitions correspond to different cleaning effects.

[0190] The area determination module 30 is configured to determine the hair implantation area of ​​all hair implantation holes included in each brush head partition by performing image detection on each brush head partition.

[0191] The result determination module 40 is configured to determine the detection result corresponding to the brush head based on the bristle area corresponding to each brush head partition.

[0192] In some implementations, the region segmentation module 20 is configured to:

[0193] Image detection is performed on the image to be processed to determine the region image corresponding to the bristle-planting area of ​​the brush head;

[0194] The region image is input into a pre-trained region segmentation network to obtain the multiple brush head partitions output by the region segmentation network.

[0195] In some embodiments, the area determination module 30 is configured to:

[0196] For any brush head partition, perform image detection on the brush head partition to obtain the hair implantation holes included in the brush head partition;

[0197] The bristle implantation holes in the brush head partition are classified to obtain the number of bristle implantation holes and the area of ​​a single bristle implantation hole for each category.

[0198] For any category, determine the total area of ​​the hair implantation holes corresponding to that category based on the number of hair implantation holes and the area of ​​a single hair implantation hole;

[0199] The hair implantation area is obtained by summing the total area of ​​the hair implantation holes corresponding to all categories.

[0200] In some embodiments, the plurality of brush head partitions include a first region located at the front and rear ends of the brush head, a second region located on both sides of the brush head, and a third region located in the middle of the brush head.

[0201] In some implementations, the result determination module 40 is configured to:

[0202] The total bristle area of ​​the brush head is obtained by summing the bristle areas corresponding to each brush head section.

[0203] The detection result corresponding to the brush head is determined based on the ratio of the bristle area of ​​each brush head zone to the total bristle area, the ratio of the bristle area of ​​the second region to the sum of the bristle areas of other brush head zones, and the ratio of the preset pressure to the total bristle area.

[0204] In some embodiments, the image acquisition module 10 is configured to:

[0205] Acquire an initial image of the bristle-planting area of ​​the brush head;

[0206] The initial image is segmented to obtain the image to be processed.

[0207] In some embodiments, this disclosure provides a toothbrush head detection system, comprising:

[0208] Image acquisition device, used to acquire images of the toothbrush head to be inspected;

[0209] Processor; and

[0210] The memory stores computer instructions that cause the processor to perform the method described in any of the above embodiments.

[0211] The detection system can be implemented as shown in Figure 2 above, and will not be described in detail here.

[0212] In some embodiments, this disclosure provides a storage medium storing computer instructions for causing a computer to perform the methods described in any of the above embodiments.

[0213] In some embodiments, this disclosure provides a computer program product that implements the methods described in any of the above embodiments when running.

[0214] Specifically, Figure 13 shows a schematic diagram of the structure of a system 600 suitable for implementing the method of this disclosure. The corresponding functions of the controller and storage medium described above can be realized through the structure shown in Figure 13.

[0215] As shown in Figure 13, system 600 includes a processor 601, which can perform various appropriate actions and processes according to a program stored in memory 602 or a program loaded into memory 602 from storage section 608. Memory 602 also stores various programs and data required for the operation of system 600. Processor 601 and memory 602 are connected to each other via bus 604. Input / output (I / O) interface 605 is also connected to bus 604.

[0216] The following components are connected to I / O interface 605: an input section 606 including a keyboard, mouse, etc.; an output section 607 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 608 including a hard disk, etc.; and a communication section 609 including a network interface card such as a LAN card, modem, etc. The communication section 609 performs communication processing via a network such as the Internet. A drive 610 is also connected to I / O interface 605 as needed. A removable medium 611, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 610 as needed so that computer programs read from it can be installed into storage section 608 as needed.

[0217] In particular, according to embodiments of this disclosure, the above-described method process can be implemented as a computer software program. For example, embodiments of this disclosure include a computer program product comprising a computer program tangibly embodied on a machine-readable medium, the computer program containing program code for performing the above-described methods. In such embodiments, the computer program can be downloaded and installed from a network via communication section 609, and / or installed from removable medium 611.

[0218] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0219] Obviously, the above embodiments are merely examples for clear illustration and are not intended to limit the embodiments. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all embodiments here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this disclosure.

Claims

1. A method for detecting a toothbrush head, comprising: obtaining a to-be-processed image, the to-be-processed image comprising a bristle planting area of a toothbrush head to be detected, the bristle planting area comprising a plurality of bristle planting holes; performing region segmentation on the bristle planting area of the to-be-processed image to obtain a plurality of brush head sub-regions included in the bristle planting area, wherein the plurality of brush head sub-regions correspond to different cleaning effects; determining, by performing image detection on each brush head sub-region, a bristle planting area of all bristle planting holes included in each brush head sub-region; determining, according to the bristle planting area corresponding to each brush head sub-region, a detection result corresponding to the brush head.

2. The method of claim 1, wherein, The performing of the region segmentation on the bristle planting area of the to-be-processed image to obtain the plurality of brush head sub-regions included in the bristle planting area comprises: performing image detection on the to-be-processed image to determine a region image corresponding to the bristle planting area of the brush head; inputting the region image into a pre-trained region segmentation network to obtain the plurality of brush head sub-regions output by the region segmentation network.

3. The method of claim 1, wherein, The determining of the bristle planting area of all bristle planting holes included in each brush head sub-region by performing image detection on each brush head sub-region comprises: for any one brush head sub-region, performing image detection on the brush head sub-region to obtain bristle planting holes included in the brush head sub-region; classifying the bristle planting holes included in the brush head sub-region to obtain a number of bristle planting holes corresponding to each category and an area of a single bristle planting hole; for any one category, determining a total area of bristle planting holes corresponding to the category according to the number of bristle planting holes and the area of a single bristle planting hole corresponding to the category; summing up the total areas of bristle planting holes corresponding to all categories to obtain the bristle planting area. 4.The method of any one of claims 1 to 3, wherein the plurality of brush head sub-regions comprise a first region located at a front end and a tail end of the brush head, a second region located at two sides of the brush head, and a third region located at a middle part of the brush head.

5. The method of claim 4, wherein, The determining of the detection result corresponding to the brush head according to the bristle planting area corresponding to each brush head sub-region comprises: summing up the bristle planting area corresponding to each brush head sub-region to obtain a total bristle planting area of the brush head; determining the detection result corresponding to the brush head according to a ratio of the bristle planting area of each brush head sub-region to the total bristle planting area, a ratio of the bristle planting area of the second region to a sum of the bristle planting areas of the other brush head sub-regions, and a ratio of a preset pressure to the total bristle planting area.

6. The method of claim 1, wherein, The obtaining of the to-be-processed image comprises: collecting an initial image of the bristle planting area of the brush head; performing background segmentation on the initial image to obtain the to-be-processed image. 7.A device for detecting a toothbrush head, comprising: an image obtaining module configured to obtain a to-be-processed image, the to-be-processed image comprising a bristle planting area of a toothbrush head to be detected, the bristle planting area comprising a plurality of bristle planting holes; a region segmentation module configured to perform region segmentation on the bristle planting area of the to-be-processed image to obtain a plurality of brush head sub-regions included in the bristle planting area, wherein the plurality of brush head sub-regions correspond to different cleaning effects; an area determining module configured to determine, by performing image detection on each brush head sub-region, a bristle planting area of all bristle planting holes included in each brush head sub-region. The result determination module is configured to determine the detection result corresponding to the brush head according to the bristle area corresponding to each brush head partition.

8. The detection device of claim 7, wherein, The region segmentation module is configured to: perform image detection on the image to be processed to determine a region image corresponding to the bristle region of the brush head; input the region image into a pre-trained region segmentation network to obtain the plurality of brush head partitions output by the region segmentation network.

9. The detection device of claim 7, wherein, The area determination module is configured to: for any one brush head partition, perform image detection on the brush head partition to obtain bristle holes included in the brush head partition; classify the bristle holes included in the brush head partition to obtain the number of bristle holes corresponding to each category and the area of a single bristle hole; for any one category, determine the total bristle hole area corresponding to the category according to the number of bristle holes corresponding to the category and the area of a single bristle hole; sum the total bristle hole areas corresponding to all categories to obtain the bristle area.

10. The detection device of claim 7, wherein, The plurality of brush head partitions include a first region located at the front and tail ends of the brush head, a second region located at the two sides of the brush head, and a third region located at the middle of the brush head.

11. The detection device of claim 7, wherein, The result determination module is configured to: sum the bristle area corresponding to each brush head partition to obtain the total bristle area of the brush head; determine the detection result corresponding to the brush head according to the ratio of the bristle area of each brush head partition to the total bristle area, the ratio of the bristle area of the second region to the sum of the bristle areas of the other brush head partitions, and the ratio of the preset pressure to the total bristle area.

12. The detection device of claim 7, wherein, The image acquisition module is configured to: acquire an initial image of the bristle region of the brush head; perform background segmentation on the initial image to obtain the image to be processed.

13. A toothbrush head detection system, comprising: an image acquisition device configured to acquire an image of a toothbrush head to be detected; a processor; and a memory storing computer instructions for causing the processor to execute the method according to any one of claims 1 to 6.

14. A storage medium storing computer instructions for causing a computer to execute the method according to any one of claims 1 to 6.

15. A computer program product, which, when executed, implements the method according to any one of claims 1 to 6.

Images