Method and system for estimating gymnasium for nursing home based on image recognition

Abstract

The invention discloses a method and a system for estimating a gymnasium for a nursing home based on image recognition. The method for estimating the gymnasium for the nursing home based on image recognition comprises the following steps: acquiring user request information; analyzing the user request information to obtain to-be-obtained data information; according to the to-be-obtained data information, a camera device in a gymnasium and all intelligent fitness equipment capable of sending use data are called to work, an image information set in the gymnasium is obtained through the camera device, and the use information, transmitted by all the intelligent fitness equipment, of the intelligent fitness equipment is obtained through all the intelligent fitness equipment; acquiring gymnasium historical use data information and gymnasium reservation information; and according to the image information group, the use information of the intelligent fitness equipment transmitted by the intelligent fitness equipment, the historical use data information of the gymnasium and the gymnasium reservation information, obtaining to-be-obtained data information. According to the application, the user can know the specific condition of the gymnasium without going out.

Description

Technical Field

[0001] The present application relates to the field of image recognition technology, and in particular to a method for estimating a gym for a nursing home based on image recognition and a system for estimating a gym for a nursing home based on image recognition. Background Art

[0002] Existing large-scale nursing homes occupy a huge area. For example, Taikang Yanyuan Nursing Home in Beijing covers an area of ​​about 140,000 square meters and can provide about 3,000 nursing units. Assuming two elderly people in one household, the number will be around 6,000 people. The nursing home is equipped with a gym, cinema, multiple restaurants, massage, community hospital and other facilities for the elderly in the nursing home to use. The large area provides the elderly in the nursing home with enough space and facilities for activities. However, it also causes the elderly to need more exercise when exercising in the nursing home. For example, it may take an ordinary person 10 minutes to walk from their residence to the restaurant, while the elderly (especially those with underlying diseases) may need 20 minutes or more to walk the same distance from their residence to the hotel. However, if they go to the site and find that there is no space or other situations, it will not only disappoint the elderly but also cause physical burden on them.

[0003] Taking the gym as an example, in the prior art, one can inquire whether there are vacancies in various facilities by calling. However, the shortcomings of this method are:

[0004] (1) The facility owner needs to arrange for someone to provide telephone service, which wastes manpower;

[0005] (2) The number of consultation calls that can be accommodated by general facilities is limited. When there are many people consulting, the phone line may be busy for a long time;

[0006] (3) Elderly people may be hard of hearing or have speech problems, which may lead to poor communication. This not only fails to solve the problem, but also prolongs the telephone communication time and delays other telephone calls.

[0007] (4) Even if the call is connected and communication is made with the service provider, the service provider may not understand the specific situation or be unable to express it in detail. For example, when asking about the number of people swimming in the gym's swimming pool, the service provider may only be able to say general statements such as "a lot" or "a little".

[0008] Therefore, it is hoped that a technical solution can be provided to solve or at least alleviate the above-mentioned deficiencies of the prior art. Summary of the invention

[0009] The purpose of the present invention is to provide a method for estimating the use of a gym in a nursing home based on image recognition to solve at least one of the above-mentioned technical problems.

[0010] The present invention provides the following scheme:

[0011] According to one aspect of the present invention, a method for estimating a gym for a nursing home based on image recognition is provided, and the method for estimating a gym for a nursing home based on image recognition comprises:

[0012] The data center obtains user request information;

[0013] The data center analyzes the user request information to obtain the data information to be obtained;

[0014] The data center mobilizes the camera device in the gym and each smart fitness device capable of sending usage data to work according to the data information to be obtained, wherein the image information group in the gym is obtained by the camera device, and the usage information of each smart fitness device transmitted by each smart fitness device is obtained by each smart fitness device;

[0015] The data center obtains the gym's historical usage data and gym reservation information;

[0016] The data center obtains the data information to be obtained according to the image information group, the usage information of the smart fitness device transmitted by each smart fitness device, the historical usage data information of the gym, and the reservation information of the gym.

[0017] Optionally, the data information to be acquired includes current personnel quantity information, current equipment usage information, future personnel quantity prediction information, and future equipment usage information;

[0018] The obtaining of estimated information according to the image information group, the usage information of the smart fitness device transmitted by each smart fitness device, the historical usage data information of the gym, and the gym reservation information includes:

[0019] Acquire current number of personnel information according to the image information group;

[0020] Acquire current device usage information according to usage information of each smart fitness device transmitted by the smart fitness device;

[0021] Acquire future personnel quantity prediction information according to the image information group, usage information of each smart fitness device transmitted by the smart fitness device, historical usage data information of the gym, and gym reservation information;

[0022] The future equipment usage information is acquired according to the image information group, the usage information of each smart fitness device transmitted by the smart fitness device, the historical usage data information of the gym and the gym reservation information.

[0023] Optionally, the image information group includes a plurality of images, and one camera device provides one image;

[0024] The acquiring of the number of people information according to the image information group comprises:

[0025] For each image in the image information group, the following operations are performed: the human-shaped target in the image is identified by the YOLO neural network;

[0026] Acquire a related camera device relationship database, wherein the related camera device relationship database includes a plurality of camera device coordinate relationship sub-libraries, each camera device coordinate relationship sub-library stores at least part of pixel points of one camera device and at least part of pixel points of another camera device, and a pixel point in one camera device corresponds to a pixel point in another camera device, and two pixel points having a corresponding relationship are called overlapping pixel points;

[0027] Obtaining the position of each pixel point in the image occupied by each human-shaped target in the image taken by each camera device;

[0028] Deduplication processing is performed on each human-shaped target according to the position of each pixel point in the image occupied by each human-shaped target and the related camera device relationship database;

[0029] The number of humanoid targets after deduplication processing is obtained as the current number of personnel information.

[0030] Optionally, the acquiring of future personnel quantity prediction information according to the image information group, usage information of the smart fitness device transmitted by each smart fitness device, historical usage data information of the gym, and gym reservation information includes:

[0031] Respectively identify the identity information of the human-shaped target in each image information group;

[0032] According to the acquired identity information of each humanoid target and the historical usage data information of the gym, the probability of the humanoid target leaving the gym within a preset future time is acquired; when the probability is less than a preset threshold, it is determined that the person is still in the gym within the preset future time; when the probability is greater than the preset threshold, it is determined that the person leaves the gym within the preset future time;

[0033] Obtain the number of people who will enter the gym within a preset time in the future according to the gym reservation information;

[0034] The sum of the number of people who are still in the gym within a preset future time and the number of people who enter the gym within the preset future time constitutes future personnel quantity prediction information.

[0035] Optionally, the acquiring future equipment usage information according to the image information group, usage information of each smart fitness device transmitted by the smart fitness device, historical usage data information of the gym, and reservation information of the gym includes:

[0036] Respectively identify the identity information of the human-shaped target in each image information group;

[0037] According to the identity information of each humanoid target obtained, obtain the smart fitness equipment that each humanoid target may use and the smart fitness equipment that has never been used from the historical usage data information of the gym;

[0038] Determine, based on the historical gym usage data information, which smart fitness equipment each humanoid target may use and which has been used in the current training;

[0039] The future equipment usage information of the people in the current gym is obtained based on the smart fitness equipment that has been used by each human-shaped target and the smart fitness equipment that has never been used.

[0040] Optionally, the acquiring future equipment usage information according to the image information group, usage information of each smart fitness device transmitted by the smart fitness device, historical gym usage data information, and gym reservation information further includes:

[0041] Acquire personnel information of personnel who will enter the gym within a preset time in the future according to the gym reservation information, and the personnel information is called reservation personnel information;

[0042] Get the gym's historical usage data for each reservation person;

[0043] Obtain the fitness equipment that the reservation person may use based on the gym's historical usage data information of each reservation person; wherein,

[0044] The fitness equipment that may be used by the reservation personnel and the future equipment usage information of the personnel currently in the gym constitute the future equipment usage information.

[0045] The present application also provides a nursing home gym estimation system based on image recognition, which includes a data center, various cameras located in the gym, and at least one client for generating user request information. The data center, the camera devices, and the client cooperate to implement the nursing home gym estimation method based on image recognition as described above.

[0046] The image recognition-based nursing home gym estimation method of the present application can provide users with various data information to be obtained according to the user's request, so that users can know the specific situation of the gym without going out, thereby reducing the cost of the elderly returning empty-handed. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a flowchart of a method for estimating the use of a fitness room in a nursing home based on image recognition in one embodiment of the present application. DETAILED DESCRIPTION

[0048] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] like Figure 1 The image recognition-based method for estimating the use of a gym in a nursing home includes:

[0050] Step 1: The data center obtains the user request information;

[0051] Step 2: The data center analyzes the user request information to obtain the data information to be obtained;

[0052] Step 3: The data center mobilizes the camera device in the gym and each smart fitness device capable of sending usage data to work according to the data information to be obtained, wherein the image information group in the gym is obtained by the camera device, and the usage information of each smart fitness device transmitted by each smart fitness device is obtained by each smart fitness device;

[0053] Step 4: The data center obtains the gym's historical usage data and gym reservation information;

[0054] Step 5: The data center obtains the data information to be obtained according to the image information group, the usage information of each smart fitness device transmitted by the smart fitness device, the historical usage data information of the gym, and the reservation information of the gym.

[0055] The image recognition-based nursing home gym estimation method of the present application can provide users with various data information to be obtained according to the user's request, so that users can know the specific situation of the gym without going out, thereby reducing the cost of the elderly returning empty-handed.

[0056] In this embodiment, the data information to be acquired includes current personnel quantity information, current equipment usage information, future personnel quantity prediction information, and future equipment usage information;

[0057] The obtaining of estimated information according to the image information group, the usage information of the smart fitness device transmitted by each smart fitness device, the historical usage data information of the gym, and the gym reservation information includes:

[0058] Acquire current number of personnel information according to the image information group;

[0059] Acquire current device usage information according to usage information of each smart fitness device transmitted by the smart fitness device;

[0060] Acquire future personnel quantity prediction information according to the image information group, usage information of each smart fitness device transmitted by the smart fitness device, historical usage data information of the gym, and gym reservation information;

[0061] The future equipment usage information is acquired according to the image information group, the usage information of each smart fitness device transmitted by the smart fitness device, the historical usage data information of the gym and the gym reservation information.

[0062] In this embodiment, the image information group includes a plurality of images, and one camera device provides one image;

[0063] The acquiring of the number of people information according to the image information group comprises:

[0064] For each image in the image information group, the following operations are performed: the human-shaped target in the image is identified by the YOLO neural network;

[0065] Acquire a related camera device relationship database, wherein the related camera device relationship database includes a plurality of camera device coordinate relationship sub-libraries, each camera device coordinate relationship sub-library stores at least part of pixel points of one camera device and at least part of pixel points of another camera device, and a pixel point in one camera device corresponds to a pixel point in another camera device, and two pixel points having a corresponding relationship are called overlapping pixel points;

[0066] Obtaining the position of each pixel point in the image occupied by each human-shaped target in the image taken by each camera device;

[0067] Deduplication processing is performed on each human-shaped target according to the position of each pixel point in the image occupied by each human-shaped target and the related camera device relationship database;

[0068] The number of humanoid targets after deduplication processing is obtained as the current number of personnel information.

[0069] In this way, relatively accurate information on the current number of people can be obtained. It can be understood that other existing image deduplication methods can also be applied to the solution of the present application.

[0070] In this embodiment, the obtaining of future personnel quantity prediction information according to the image information group, the usage information of the smart fitness device transmitted by each smart fitness device, the historical usage data information of the gym, and the gym reservation information includes:

[0071] Respectively identify the identity information of the human-shaped target in each image information group;

[0072] According to the identity information of each humanoid target and the historical usage data of the gym, the probability of the humanoid target leaving the gym within a preset future time is obtained; when the probability is less than a preset threshold, it is determined that the person is still in the gym within the preset future time; when the probability is greater than the preset threshold, it is determined that the person leaves the gym within the preset future time;

[0073] Obtain the number of people who will enter the gym within a preset time in the future according to the gym reservation information;

[0074] The sum of the number of people who are still in the gym within a preset future time and the number of people who enter the gym within the preset future time constitutes future personnel quantity prediction information.

[0075] In this embodiment, the probability of the humanoid target leaving the gym within a preset future time can be obtained by the following method based on the identity information of each humanoid target and the historical usage data information of the gym:

[0076] Obtain the historical data information of the gym. In this embodiment, the historical data information of the gym includes the historical data information of everyone who has ever been to the gym. Since the gym in this application is a gym used in a nursing home, a real-name registration system is usually adopted. Therefore, a gym file can be established for each elderly person in the nursing home. These gym files are stored in the historical data information of the gym. The gym file may include the time used for each fitness session, the fitness equipment commonly used for each fitness session, and other content.

[0077] In this embodiment, the identity information of the humanoid target can be obtained by face recognition. Since the gym of this application is opened for the elderly in nursing homes, a file can be established in advance for each elderly person by entering the face image, so that the identity information of each humanoid target can be obtained by comparing the image similarity.

[0078] In this embodiment, the probability of the humanoid target leaving the gym within a preset future time is obtained based on the acquired identity information of each humanoid target and the gym's historical usage data information, including:

[0079] Obtain the current time of the humanoid target entering the gym (it is understandable that this information can be obtained through face recognition using a photo taken by a camera when entering the gym, that is, the humanoid target is obtained by face recognition, and the time is obtained by using the clock of the camera. In addition, it can also be obtained by registration, for example, the gym can be entered by face scanning or by card scanning, and the time can also be obtained);

[0080] After obtaining the time of entering the gym, the average exercise time of the humanoid target is obtained through the historical gym usage data information. It can be understood that the average exercise time can be obtained by dividing the sum of the time used for each exercise by the average number of exercises.

[0081] Get the current time, and get the current exercise time based on the current time of the humanoid target entering the gym and the current time. If the current exercise time is close to the average time or exceeds the average time, 1 point is obtained, otherwise 0 point is obtained. It can be understood that a threshold can be set for the current exercise time to be close to the average time, such as within 10 minutes or within 20 minutes.

[0082] It is understandable that, in one embodiment, the current period time can also be obtained, and the time used for each exercise related to the current date and time in the gym's historical usage data information is selected according to the current date and time to obtain the average exercise time. For example, through the current period time, it is found that it is currently Sunday, and the user's usual exercise time is different from the exercise time on weekends. Therefore, only the time used for each exercise on Sunday in the gym's historical usage data information is selected to obtain the average exercise time.

[0083] In one embodiment, only the current exercise time may be used to determine the probability if it is close to or exceeds the average time, that is, if it is close to or exceeds the average time, it is considered to be greater than the preset threshold, and if it is far from the average time, it is considered to be less than the average time.

[0084] In another embodiment, other probability judgments can be added. For example, each time the human-shaped target has 5 exercise machines, and 2 exercise machines have just been exercised, then each unexercised machine will get 1 point, otherwise 0 points. At this time, a preset threshold can be set, such as 3 points. If the score of the unexercised machine is close to the average time of the current exercise time or exceeds the average time score and exceeds the preset threshold, then the probability is considered to be greater than the preset threshold, and it is judged that the person is still in the gym within the future preset time.

[0085] In this embodiment, after determining which people in the gym will still be in the gym, the number of people entering the gym within a future preset time can be obtained through the future appointment time. The future preset time can be set by yourself, for example, 30 minutes, 1 hour, etc.

[0086] In this embodiment, obtaining future equipment usage information according to the image information group, usage information of each smart fitness device transmitted by the smart fitness device, historical gym usage data information, and gym reservation information includes:

[0087] Respectively identify the identity information of the human-shaped target in each image information group;

[0088] According to the identity information of each human-shaped target obtained, the smart fitness devices that each human-shaped target may use and the smart fitness devices that have never been used are obtained from the historical usage data information of the gym; for example, if in the historical usage data information of the gym, a person has never used a certain device, then the device is a smart fitness device that has never been used. It is understandable that a threshold value can also be set. For example, if a person has come to the gym more than 20 times but has used the device less than 2 times, it is considered to be a smart fitness device that has never been used.

[0089] According to the historical usage data information of the gym, the smart fitness equipment that each human-shaped target may use and which has been used in the current training is determined; through the image information obtained by each camera device, it is easy to obtain the smart fitness equipment used by each human-shaped target after entering the gym this time. Usually, the used fitness equipment will not be used again. Therefore, in this way, the usage of each fitness equipment by the people currently in the gym can be known.

[0090] According to the smart fitness equipment that each human-shaped target has used and the smart fitness equipment that has never been used, the future equipment usage information of the people in the current gym is obtained. For example, suppose there are two users, A and B, in the gym, and there are 10 kinds of fitness equipment in the gym, of which A will never use and B will never use, then these two kinds of fitness equipment are idle fitness equipment. If one of them is not used by A but will be used by B, but it is found through image recognition obtained by the camera device that B has used this fitness equipment this time, then this fitness equipment is also idle fitness equipment. In addition, if the number of fitness equipment of this type is large and exceeds the number of users who will use it but have not used it yet, then this fitness equipment is still idle fitness equipment.

[0091] If a type of fitness equipment is the kind of fitness equipment that A uses every time he comes and has not been used at the current stage, then the fitness equipment is fitness equipment to be used.

[0092] If a type of fitness equipment is currently occupied, then the fitness equipment is considered to be fitness equipment in use.

[0093] It is understandable that most fitness equipment, especially smart fitness equipment, such as treadmills, can obtain information about what is being used, and therefore can directly send the information to the data center for statistics, while some non-intelligent fitness equipment, such as dumbbells, can be identified through image recognition.

[0094] In this embodiment, the acquiring of future equipment usage information according to the image information group, the usage information of each smart fitness device transmitted by the smart fitness device, the historical usage data information of the gym, and the gym reservation information further includes:

[0095] Acquire personnel information of personnel who will enter the gym within a preset time in the future according to the gym reservation information, and the personnel information is called reservation personnel information;

[0096] Get the gym's historical usage data for each reservation person;

[0097] Obtain the fitness equipment that the reservation person may use based on the gym's historical usage data information of each reservation person; wherein,

[0098] The fitness equipment that may be used by the reservation personnel and the future equipment usage information of the personnel currently in the gym constitute the future equipment usage information.

[0099] In this embodiment, each person who has made an appointment will theoretically have historical gym usage data information, and through the historical gym usage data information, their usual fitness habits can also be obtained, that is, what kind of fitness equipment they generally use, so as to know the fitness equipment that the person who has made an appointment may use.

[0100] The present application also provides a nursing home gym estimation system based on image recognition, which includes a data center, various cameras located in the gym, and at least one client for generating user request information. The data center, the camera devices, and the client cooperate to implement the nursing home gym estimation method based on image recognition as described above.

[0101] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for estimating the number of gyms used in nursing homes based on image recognition, characterized in that: The method for estimating a gym for a nursing home based on image recognition includes: The data center obtains user request information; The data center analyzes the user request information to obtain the data information to be obtained; The data center mobilizes the camera device in the gym and each smart fitness device capable of sending usage data to work according to the data information to be obtained, wherein the image information group in the gym is obtained by the camera device, and the usage information of each smart fitness device transmitted by each smart fitness device is obtained by each smart fitness device; The data center obtains the gym's historical usage data and gym reservation information; The data center obtains the data information to be obtained according to the image information group, the usage information of the smart fitness device transmitted by each smart fitness device, the historical usage data information of the gym, and the reservation information of the gym.

2. The method for estimating the use of a gym in a nursing home based on image recognition as claimed in claim 1, characterized in that: The data information to be acquired includes current personnel quantity information, current equipment usage information, future personnel quantity prediction information, and future equipment usage information; The obtaining of estimated information according to the image information group, the usage information of the smart fitness device transmitted by each smart fitness device, the historical usage data information of the gym, and the gym reservation information includes: Acquire current number of personnel information according to the image information group; Acquire current device usage information according to usage information of each smart fitness device transmitted by the smart fitness device; Acquire future personnel quantity prediction information according to the image information group, usage information of each smart fitness device transmitted by the smart fitness device, historical usage data information of the gym, and gym reservation information; The future equipment usage information is acquired according to the image information group, the usage information of each smart fitness device transmitted by the smart fitness device, the historical usage data information of the gym and the gym reservation information.

3. The method for estimating the use of a gym in a nursing home based on image recognition as claimed in claim 2, characterized in that: The image information group includes a plurality of images, and one camera device provides one image; The acquiring of the number of people information according to the image information group comprises: For each image in the image information group, the following operations are performed: the human-shaped target in the image is identified by the YOLO neural network; Acquire a related camera device relationship database, wherein the related camera device relationship database includes a plurality of camera device coordinate relationship sub-libraries, each camera device coordinate relationship sub-library stores at least part of pixel points of one camera device and at least part of pixel points of another camera device, and a pixel point in one camera device corresponds to a pixel point in another camera device, and two pixel points having a corresponding relationship are called overlapping pixel points; Obtaining the position of each pixel point in the image occupied by each human-shaped target in the image taken by each camera device; Deduplication processing is performed on each human-shaped target according to the position of each pixel point in the image occupied by each human-shaped target and the related camera device relationship database; The number of humanoid targets after deduplication processing is obtained as the current number of personnel information.

4. The method for estimating the use of gym facilities in a nursing home as claimed in claim 3, characterized in that: The obtaining of future personnel quantity prediction information according to the image information group, the usage information of the smart fitness equipment transmitted by each smart fitness equipment, the historical usage data information of the gym and the gym reservation information comprises: Respectively identify the identity information of the human-shaped target in each image information group; According to the identity information of each humanoid target and the historical usage data of the gym, the probability of the humanoid target leaving the gym within a preset future time is obtained; when the probability is greater than a preset threshold, it is determined that the person is still in the gym within the preset future time; when the probability is greater than the preset threshold, it is determined that the person leaves the gym within the preset future time; Obtain the number of people who will enter the gym within a preset time in the future according to the gym reservation information; The sum of the number of people who are still in the gym within a preset future time and the number of people who enter the gym within the preset future time constitutes future personnel quantity prediction information.

5. The method for estimating the use of gym facilities in a nursing home as claimed in claim 4, characterized in that: The acquiring of future equipment usage information according to the image information group, usage information of each smart fitness equipment transmitted by the smart fitness equipment, historical gym usage data information, and gym reservation information comprises: Respectively identify the identity information of the human-shaped target in each image information group; According to the identity information of each humanoid target obtained, obtain the smart fitness equipment that each humanoid target may use and the smart fitness equipment that has never been used from the historical usage data information of the gym; Determine, based on the historical gym usage data information, which smart fitness equipment each humanoid target may use and which has been used in the current training; The future equipment usage information of the people in the current gym is obtained based on the smart fitness equipment that has been used by each human-shaped target and the smart fitness equipment that has never been used.

6. The method for estimating the use of gym facilities in a nursing home as claimed in claim 5, characterized in that: The acquiring of future equipment usage information according to the image information group, usage information of the smart fitness equipment transmitted by each smart fitness equipment, historical usage data information of the gym and gym reservation information further includes: According to the gym reservation information, obtain the personnel information of the personnel who will enter the gym within a preset time in the future, and the personnel information is called the reservation personnel information; Get the gym's historical usage data for each reservation person; Obtain the fitness equipment that the reservation person may use based on the gym's historical usage data information of each reservation person; wherein, The fitness equipment that may be used by the reservation personnel and the future equipment usage information of the personnel currently in the gym constitute the future equipment usage information.

7. A fitness gym estimation system for nursing homes based on image recognition, characterized in that: The image recognition-based fitness center estimation system for nursing homes includes a data center, various cameras located in the fitness center, and at least one client for generating user request information. The data center, the cameras, and the client cooperate to implement the image recognition-based fitness center estimation method for nursing homes as described in any one of claims 1 to 6.

Images