CN113091223A - Internet of things air purification method and system for machine learning based on placement position - Google Patents
Internet of things air purification method and system for machine learning based on placement position Download PDFInfo
- Publication number
- CN113091223A CN113091223A CN202110246910.8A CN202110246910A CN113091223A CN 113091223 A CN113091223 A CN 113091223A CN 202110246910 A CN202110246910 A CN 202110246910A CN 113091223 A CN113091223 A CN 113091223A
- Authority
- CN
- China
- Prior art keywords
- air
- air purification
- purification
- internet
- movable
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/50—Control or safety arrangements characterised by user interfaces or communication
- F24F11/54—Control or safety arrangements characterised by user interfaces or communication using one central controller connected to several sub-controllers
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/50—Control or safety arrangements characterised by user interfaces or communication
- F24F11/56—Remote control
- F24F11/58—Remote control using Internet communication
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
- F24F11/64—Electronic processing using pre-stored data
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/62—Control or safety arrangements characterised by the type of control or by internal processing, e.g. using fuzzy logic, adaptive control or estimation of values
- F24F11/63—Electronic processing
- F24F11/65—Electronic processing for selecting an operating mode
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F11/00—Control or safety arrangements
- F24F11/70—Control systems characterised by their outputs; Constructional details thereof
- F24F11/72—Control systems characterised by their outputs; Constructional details thereof for controlling the supply of treated air, e.g. its pressure
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F24—HEATING; RANGES; VENTILATING
- F24F—AIR-CONDITIONING; AIR-HUMIDIFICATION; VENTILATION; USE OF AIR CURRENTS FOR SCREENING
- F24F2110/00—Control inputs relating to air properties
- F24F2110/50—Air quality properties
- F24F2110/65—Concentration of specific substances or contaminants
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02B—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO BUILDINGS, e.g. HOUSING, HOUSE APPLIANCES OR RELATED END-USER APPLICATIONS
- Y02B30/00—Energy efficient heating, ventilation or air conditioning [HVAC]
- Y02B30/70—Efficient control or regulation technologies, e.g. for control of refrigerant flow, motor or heating
Abstract
The invention relates to an Internet of things air purification method and system based on placement position for machine learning, which are applied to an air purification terminal with at least three purification modes, and are used for obtaining a depth image of an area where the air purification terminal is located and constructing an image sample with three-dimensional parameters in the area by combining a distance sensor; when entering a learning mode, starting all purification modes for air purification, and acquiring air indexes of an area where an air purification terminal is located; when a certain air index exceeds a set threshold value, marking the air index as a key index, corresponding the key index to an image sample of the area, and performing machine learning to train to obtain a movable area; after exiting from the learning mode, acquiring a moving instruction, and automatically moving the air purification terminal to a moving instruction marking point position to perform air purification of key indexes; the air purifier can adapt to different house types, key purification modes of different use scenes are switched, and more excellent air purification experience is formed.
Description
Technical Field
The invention relates to the technical field of intelligent home furnishing based on the Internet of things, in particular to an Internet of things air purification system for preventing machine learning based on positions.
Background
At present, the air quality, especially the air quality, receives more and more concern in the family that has old man and child, and in order to improve the air quality, intelligent house such as air purifier, robot of sweeping the floor, humidifier, door and window controller become the indispensable domestic appliance of user. In the current market, a plurality of purification functions are integrated, so that one purifier can not only complete PM2.5 purification, but also complete a plurality of purification processes such as oil smoke particle concentration purification and the like; the user can buy a plurality of air purifiers to be arranged at a plurality of places for purification, but the purification system is also controlled manually, and in addition to simple switch and mode selection, an automatic traveling type air purifier in a sweeping robot mode is required in the market; meanwhile, in household use, the regional characteristics are very obvious, for example, most balconies have more cigarette particles, pet houses are full of pet hair, kitchens are polluted by oil smoke particles, air purifiers with multiple purification functions need to be detected and purified all the time, and a lot of items are useless, so that a lot of resources are wasted. Aiming at the air purifier in the prior art, the function that the air purifier automatically moves to a fixed selection place through manual control and autonomously selects a corresponding purification mode can not be achieved.
Disclosure of Invention
Aiming at the existing technologies such as: the prior art does not establish the area that the air purifier can move to through machine learning.
In particular to an Internet of things air purification method for machine learning based on placement position, which is applied to an air purification terminal with at least three purification modes,
entering a learning mode: acquiring a depth image of an area where an air purification terminal is located, and constructing an image sample with three-dimensional parameters of the area by combining a distance sensor;
when entering a learning mode, starting all purification modes for air purification, and acquiring air indexes of an area where an air purification terminal is located;
comparing the air indexes with a set threshold, marking the air indexes as key indexes when a certain air index exceeds the set threshold, corresponding the key indexes with image samples of the area, and performing machine learning to train to obtain a movable area; the air purification terminal exits the learning mode when the air purification terminal acquires the movable area for the third time and the number of the same image samples exceeds 80%;
and after exiting the learning mode, acquiring a moving instruction, and automatically moving the air purification terminal to a moving instruction marking point position to perform air purification of key indexes.
As preferred, when opening all purification modes and carrying out air purification to obtain the air index of the region that air purification terminal belongs to, still obtain weather parameter and carry out influence factor fitting, weather parameter's influence factor includes: air humidity, temperature, return to south index.
Preferably, when the return-to-south index is high in the influence factors for acquiring the weather parameters, the weight of the dehumidification index in the key index is reduced.
Preferably, after the plurality of movable areas are acquired, whether different movable areas are connected or not is judged, and if no connected part exists between the movable areas, a reminding instruction is sent out to indicate that two movable areas which cannot be spanned exist.
Preferably, after the two non-stridable movable sections are connected, confirmation information is transmitted, the confirmation information including whether or not there is an obstacle gate or a step obstacle in the non-stridable movable section, and if there is an obstacle or a step obstacle, the non-stridable movable section is divided into two independent movable sections that are not connected to each other.
The Internet of things air purification system for machine learning based on the placement position is further disclosed, and the purification method is adopted and comprises the air purification terminal with at least three purification modes; the air evolution terminal comprises a control module, a distance sensor, a depth image module, a purification module, a storage module and an Internet of things gateway module;
the control module is used for judging whether to enter a learning mode according to the image sample acquired by the depth image module, and when the learning mode is entered: the depth image module acquires a depth image of an area where the air purification terminal is located, and an image sample with three-dimensional parameters of the area is constructed by combining a distance sensor; the image samples with three-dimensional parameters are stored in the storage module, and the image samples with the three-dimensional parameters are transmitted to the control module;
when entering a learning mode, the purification module starts all purification modes for air purification, acquires an air index of an area where an air purification terminal is located, and transmits the air index to the control module;
the control module compares the air indexes with a set threshold, marks the air indexes as key indexes when a certain air index exceeds the set threshold, corresponds the key indexes with the positions of image samples of the area, and performs machine learning to train the image samples to obtain a movable area; the depth image module exits from the learning mode when the depth image module is acquired into the movable region for the third time and the number of the same image samples exceeds 80%;
after the learning mode exits, the independent internet of things terminal is used for transmitting the acquired movement instruction to the control module, and the air purification terminal automatically moves to the movement instruction marking point position to perform air purification of key indexes.
Preferably, the internet of things gateway module is used for connecting with a mobile terminal through identity authentication and submitting internet weather parameters acquired by the mobile terminal, the internet of things gateway module sends the internet weather parameters to the control module, after all purification modes are started for air purification and air indexes of an area where an air purification terminal is located are acquired, the control module performs influence factor fitting according to the weather parameters, and influence factors of the weather parameters include: air humidity, temperature, return to south index.
Preferably, when the return-to-south index in the acquired influence factors of the weather parameters is high, the control module reduces the weight of the dehumidification index in the key index; and the purification module performs air purification according to the weight.
Preferably, the control module further comprises a map unit for acquiring the movement instruction and the movable area; after a plurality of movable areas are obtained, whether different movable areas are connected or not is judged, if no connected part exists between the movable areas, a reminding instruction is sent, and the reminding instruction is transmitted to a mobile terminal connected through identity authentication by the internet of things gateway, so that the existence of two movable areas which cannot be spanned is indicated.
Preferably, after the two non-stridable movable sections are connected, the map unit sends confirmation information, the confirmation information includes whether the non-stridable movable section has an obstacle door or a step obstacle, if the obstacle door or the step obstacle exists, the map unit does not send the reminding instruction any more, and two non-connected independent movable sections are formed.
The invention has the beneficial effects that: compared with the prior art, the Internet of things air purification method for machine learning based on the placement position is applied to an air purification terminal with at least three purification modes, and enters the learning mode: acquiring a depth image of an area where an air purification terminal is located, and constructing an image sample with three-dimensional parameters of the area by combining a distance sensor; when entering a learning mode, starting all purification modes for air purification, and acquiring air indexes of an area where an air purification terminal is located; comparing the air indexes with a set threshold, marking the air indexes as key indexes when a certain air index exceeds the set threshold, corresponding the key indexes with image samples of the area, and performing machine learning to train to obtain a movable area; the air purification terminal exits the learning mode when the air purification terminal acquires the movable area for the third time and the number of the same image samples exceeds 80%; after exiting from the learning mode, acquiring a moving instruction, and automatically moving the air purification terminal to a moving instruction marking point position to perform air purification of key indexes; the air purifier can adapt to different house types, key purification modes of different use scenes are switched, and more excellent air purification experience is formed.
Drawings
FIG. 1 is a flow chart of a method of the present invention;
FIG. 2 is a system architecture diagram of the present invention;
FIG. 3 is a schematic diagram of the placement of the application example of the present invention in area A;
FIG. 4 is a schematic diagram of the placement of the application example of the present invention in area B;
FIG. 5 is a schematic view of area connectivity in an application example of the present invention.
The main element symbols are as follows:
1. an air purification terminal; 11. a control module; 111. a map unit; 12. a distance sensor; 13. a depth image module; 14. a purification module; 15. a storage module; 16. and the gateway module of the Internet of things.
Detailed Description
In order to more clearly describe the present invention, the present invention will be further described with reference to the accompanying drawings.
In the prior art, as a sub-classification under smart home, the air purifier can also perform the function of internet of things, but the prior effect of internet of things is only used for connecting a mobile terminal of a user, so that simple operations of selecting, switching on and switching off an autonomous purification mode of the user are realized; however, as the air purification terminal with multiple purification modes, the terminal can be intelligentized in gradual machine learning only by frequently interacting with the user, and the terminal brings much time waste and useless interactive experience to the user; although some air purifiers can label the pollution level and the air sample in the area in further research and development, so that the division and the fine control of the pollution level of the area are achieved, only fixed-point purification is available, and the air purifiers occupy very large space when being purchased and respectively placed in the small house type structure at present, and waste to a great extent is caused because the living environment is greatly changed only in certain specific time periods and specific scenes such as cooking (oil fume pollution), smoking (second-hand smoke particles), pet depilation (fluff flying), wall decoration (formaldehyde gas) and the like, and the purification effect of the purifiers overflows; therefore, in response to the series of problems, it is necessary to provide the air cleaner with a function of being able to move; although the combination of the movement form is very simple, under the condition that the environmental parameters are changed all the time and the purification mode is also required to be changed on the premise of movement, the following problems still exist and need to be solved:
1. the existing method cannot realize personalized customization under different environments, for example, the house type of a user A is a 2-house 1-hall, and the house type of a user B is a 3-house 2-hall; on the premise of moving, the air purifier cannot move along a specified route due to changes of house types and environment, the pollution level of machine training in the later period of moving is invalid, and the good purifying effect is required to be achieved, namely, multiple modes are required to be started for detection all the time, so that energy waste and potential safety hazards exist; 2. factors influenced by weather cannot be eliminated, for example, moisture in the air in the south-returning day, during the period, if the air purifier learns, the pollution form of the area is a humid area by default, and a dehumidification label is formed based on the pollution form, and the factors are influenced by accident in the nature and cannot exist for a long time, so that the air purifier after machine learning performs dehumidification in dry weather; if the detection is carried out in real time, although the situation can be avoided to a certain extent, all detection modes need to be standby for a long time, the service life and the energy consumption are very high, and a scheme combining the detection modes and the energy consumption is needed; 3. when the movement is controlled, the movable area of the air purifier cannot be obtained from the internet, and the movable area of the air purifier should be constructed according to the area constructed in the internet of things, so that the construction of the movable area of the air purifier is also a problem to be solved. In order to solve the problems, a novel method for purifying air of the Internet of things based on machine learning is provided.
In particular to an internet of things air purification method based on placement position for machine learning, please refer to fig. 1, which is applied to an air purification terminal with at least three purification modes, wherein the three purification modes can be a dehumidification mode, a PM2.5 anion mode, a smoke adsorption mode, and the like;
because the air purifier cannot know the application environment in advance, when the air purifier is placed at a certain position to start working after being started for the first time, the air purifier enters a learning mode if a new place is found through comparing the depth image with the depth image stored in the storage module; after entering a learning mode, acquiring a depth image of an area where an air purification terminal is located, and constructing an image sample with three-dimensional parameters of the area by combining a distance sensor; because the depth image contains three-dimensional information in the image, an area which can move without obstacles can be constructed by combining the distance sensor; because the family environment of each household is different, the depth image is adopted for machine training and obtaining; in the real use process, the user needs to manually move the air purifier to different areas for machine learning, and one-click operation can be performed until the learning mode is exited.
When entering a learning mode, all the purification modes are required to be started for air purification because the environmental parameters are obtained for the first time, and the air indexes of the area where the air purification terminal is located are obtained;
comparing the air index with a set threshold, marking the air index as a key index when the air index exceeds the set threshold, corresponding the key index with an image sample of a region where the air index is located, performing machine learning to train the air index to obtain a movable region, and marking the pollution level of the region after obtaining a depth image and the air index of the region, so that a customized working mode of the air purifier is controlled, and after the air purifier repeatedly reaches the same region, starting of all detection and purification modes is not required again, so that a real one-key operation is achieved; the air purification terminal exits the learning mode when the air purification terminal acquires the movable area for the third time and the number of the same image samples exceeds 80%; in order to ensure the accuracy of the data of the movable region, after the characteristics of the depth images are compared, if the same samples exceed 80%, the data of the region tend to be real and stable, and a learning model is successfully constructed, and meanwhile, 80% of the data are set to be taken as invalid data to be removed in consideration of the fact that the data of a user are possibly collected in the depth images due to furniture placement change, personnel movement and small object position change; and when the user feels that the parameters are inaccurate, the data can be reset so as to manually carry out the second learning mode.
After exiting from the learning mode, the user obtains a map of the movable area through the mobile phone, then sends out movement formulation, and after the air terminal obtains a movement instruction, the air purification terminal automatically moves to a movement instruction marking point position to carry out air purification of key indexes. At this time, in the movable area, the barrier-free passing of the terminal can be met, the direct purification of key indexes of the terminal can also be met, the option and the mode which need to be purified do not need to be repeatedly detected, and certainly, if a user feels that the purification of other items needs to be carried out, the manual purification mode can be selected.
In this embodiment, opening all purification modes and carrying out air purification to when obtaining the air index in air purification terminal place region, still obtain weather parameter and carry out the influence factor fitting, weather parameter's influence factor includes: air humidity, temperature, return to south index. Since the weather parameters are accidental factors affecting the home environment from nature, such as the south returning in the spring of the south, the living room sitting north to south changes the dry air index to reach a very humid threshold value, misjudgment may exist in the machine learning process, if a mark of a key index is caused, the machine learning process may be ended in the south returning, and dehumidification is also performed under a dry condition, which is obviously a backward air purification mode, and after the weather parameters of the current day are acquired, real air characteristics of most time periods can be obtained by weight allocation, such as the oil smoke air exceeding in the kitchen year, the PM2.5 exceeding in a balcony smoking area, the lint content exceeding in a pet house and the like; and when the return-to-south index is high in the obtained influence factors of the weather parameters, reducing the weight of the dehumidification indexes in key indexes.
In this embodiment, after a plurality of movable regions are acquired, whether different movable regions are connected or not is determined, and if there is no connection portion between the movable regions, a warning instruction is issued to indicate that there are two movable regions that cannot be spanned. After the two movable sections which cannot be spanned are connected, confirmation information is sent, the confirmation information comprises whether the movable sections which cannot be spanned have obstacle doors or step obstacles or not, and if the movable sections which cannot be spanned have the obstacle doors or the step obstacles, the movable sections are divided into two independent movable sections which are not connected with each other. Therefore, if the sliding glass door and the place with high and low steps belong to the area which can not be spanned, the user needs to manually confirm whether the unconnected part is a hard obstacle instead of the area which is not completely collected, and if the unconnected part is the area which is not completely collected, the user manually adjusts the position of the terminal to supplement the movable area so as to achieve the state that the areas are communicated.
Referring to fig. 2, an internet of things air purification system for machine learning based on placement position is also disclosed, and the air purification system adopting the purification method comprises an air purification terminal 1 with at least three purification modes; the air evolution terminal comprises a control module 11, a distance sensor 12, a depth image module 13, a purification module 14, a storage module 15 and an Internet of things gateway module 16;
the control module 11 is configured to determine whether to enter a learning mode according to the image sample obtained by the depth image module 13, and when entering the learning mode: the depth image module 13 acquires a depth image of an area where the air purification terminal is located, and constructs an image sample with three-dimensional parameters in the area where the air purification terminal is located by combining the distance sensor 12; the three-dimensional parameter is stored in a storage module, and an image sample with the three-dimensional parameter is transmitted to a control module 11;
when entering the learning mode, the purification module 14 starts all the purification modes to perform air purification and air index detection, obtains the air index of the area where the air purification terminal is located, and transmits the air index to the control module 11;
the control module 11 compares the air index with a set threshold, marks the air index as a key index when the air index exceeds the set threshold, corresponds the key index with the position of the image sample of the area, and performs machine learning to train the image sample to obtain a movable area; the depth image module exits the learning mode when the depth image module is acquired into the movable region for the third time and the number of the same image samples exceeds 80%;
after exiting from the learning mode, the independent internet of things terminal is used for transmitting the acquired mobile instruction to the control module, and the air purification terminal automatically moves to the mobile instruction marking point position to perform air purification of key indexes.
In this embodiment, the internet of things gateway module 16 is configured to be connected to a mobile terminal that is authenticated, that is, the mobile terminal of the user is connected to the internet of things gateway module in an intranet connection manner, so as to prevent network information of an extranet from being leaked; the mobile terminal is requested to acquire the internet weather parameters, and the internet gateway only receives data and is not directly connected with an external network because the weather parameters are acquired by the mobile terminal; network security is enhanced; thing networking gateway module 16 sends internet weather parameter to control module 11, opens all purification mode and carries out air purification to obtain the air index in air purification terminal place region after, control module carries out the influence factor fitting according to weather parameter, and weather parameter's influence factor includes: air humidity, temperature, return to south index. When the return-to-south index is high in the obtained influence factors of the weather parameters, the control module reduces the weight of the dehumidification index in the key index; the purification module purifies the air according to the weight. The key indicators are, for example: on one side of the hall close to the toilet, the humidity and the total number of air colonies are index items far higher than those of other areas, and then the two items are taken as key indexes, and the weight of the two items can be 1: 1; just if the weather is the return south, the humidity index is far higher than the average level of the whole year and belongs to the weather parameter abnormal influence, the dehumidification is taken as a purification item with equal proportional weight, the air is possibly over-dried, and the weight of the key index of the humidity is adjusted to be 0.6 after the weather parameter is combined; forming a weight ratio of 0.6:1, the adaptive model can be trained more accurately.
And as the use of air purifier, because it constructs except that the movable region is adapted to the real house type picture, therefore should not directly connect the internet and obtain to avoid constituting unnecessary divulgence, also need not to control air purifier and move the work under the condition that the user is not at home at the same time, so its necessity of directly connecting the internet is very low, therefore through the intranet mode, for example the mode of LAN, 900MHz low frequency range directly connecting carries out thing networking connection control can.
In this embodiment, the control module 11 further includes a map unit 111 for acquiring the movement instruction and the movable area; after a plurality of movable areas are obtained, whether different movable areas are connected or not is judged, if no connected part exists between the movable areas, a reminding instruction is sent, and the internet of things gateway transmits the reminding instruction to the mobile terminal connected through identity authentication, so that the existence of two movable areas which cannot be spanned is indicated. After the two movable intervals which cannot be spanned are connected, the map unit sends confirmation information, the confirmation information comprises whether the movable intervals which cannot be spanned have barrier doors or step obstacles or not, if the movable intervals which cannot be spanned have the barriers or the step obstacles, the map unit does not send out a reminding instruction any more, and two independent movable intervals which are not connected with each other are formed.
Application example
Please refer to fig. 3-5; after purchasing the air purification terminal, the user can place the air purification terminal in the corner of the living room A at will; the low-frequency matching and connection of the mobile phone terminal are carried out through a physical key of an internet of things gateway module on the terminal, and the situation that an external network directly obtains terminal parameters is avoided; at the moment, the terminal requests the mobile phone end to acquire the internet weather parameters, the mobile phone end packages and sends the numerical values to the terminal after acquiring the parameters, and the terminal performs item correspondence to acquire weather; after the terminal works for a period of time, obtaining a depth image and an air key index of a corner A, displaying a movable area A1 at the mobile phone end, then placing the movable area at a joint B between a main bed and a living room by a user, obtaining the main bed movable area and the air key index by the terminal, displaying a movable area B1 at the mobile phone end, placing the movable area B at the corner A or the joint B of the living room for the third time, exiting from a learning mode after 80% of image samples are overlapped, indicating whether the A1 is communicated with the B1, requesting the mobile phone end to confirm the joint, if the joint has a step or a sliding door obstacle, dividing the joint into two independent movable areas, if no obstacle exists, adjusting the position of the terminal to C, supplementing the image samples of the joint, and repeating the steps to obtain a connected movable area C1; when any movable area is constructed, a user can click the movable area to operate, the terminal moves to a target area according to a moving instruction, the work of an air purification project is directly carried out by combining key indexes, and air quality detection is not required to be carried out again.
The invention has the advantages that:
1) modeling the purification project of the target area in a mode corresponding to the depth image and the key indexes without detecting again;
2) the movable area can be automatically obtained, and movement and air purification can be carried out according to the movement instruction;
3) and the mobile phone terminal is connected with the intranet in an intranet connection mode, so that data leakage caused by network attack is prevented.
The above disclosure is only for a few specific embodiments of the present invention, but the present invention is not limited thereto, and any variations that can be made by those skilled in the art are intended to fall within the scope of the present invention.
Claims (10)
1. An Internet of things air purification method based on placement position for machine learning is applied to an air purification terminal with at least three purification modes and is characterized in that,
entering a learning mode: acquiring a depth image of an area where an air purification terminal is located, and constructing an image sample with three-dimensional parameters of the area by combining a distance sensor;
when entering a learning mode, starting all purification modes for air purification, and acquiring air indexes of an area where an air purification terminal is located;
comparing the air indexes with a set threshold, marking the air indexes as key indexes when a certain air index exceeds the set threshold, corresponding the key indexes with image samples of the area, and performing machine learning to train to obtain a movable area; the air purification terminal exits the learning mode when the air purification terminal acquires the movable area for the third time and the number of the same image samples exceeds 80%;
and after exiting the learning mode, acquiring a moving instruction, and automatically moving the air purification terminal to a moving instruction marking point position to perform air purification of key indexes.
2. The internet of things air purification method based on placement position for machine learning according to claim 1, wherein when all purification modes are started for air purification and an air index of an area where an air purification terminal is located is obtained, weather parameters are also obtained for influence factor fitting, and influence factors of the weather parameters include: air humidity, temperature, return to south index.
3. The internet of things air purification method based on placement position for machine learning according to claim 2, wherein when the return-to-south index is high in the obtained influence factors of the weather parameters, the weight of the dehumidification index in the key index is reduced.
4. The internet of things air purification method based on placement position for machine learning according to claim 1, wherein after a plurality of movable regions are obtained, whether different movable regions are connected or not is judged, and if no connected part exists between the movable regions, a reminding instruction is sent to indicate that two movable regions which cannot be spanned exist.
5. The method for purifying air in the internet of things based on placement position for machine learning according to claim 4, wherein after the two movable zones which cannot be spanned are connected, confirmation information is sent, wherein the confirmation information comprises whether the movable zones which cannot be spanned have barrier gates or step obstacles, and if the movable zones have the barrier gates or the step obstacles, the movable zones are divided into two independent movable zones which are not connected with each other.
6. An Internet of things air purification system for machine learning based on placement position, which adopts the method of any one of claims 1-5, and is characterized by comprising an air purification terminal with at least three purification modes; the air evolution terminal comprises a control module, a distance sensor, a depth image module, a purification module, a storage module and an Internet of things gateway module;
the control module is used for judging whether to enter a learning mode according to the image sample acquired by the depth image module, and when the learning mode is entered: the depth image module acquires a depth image of an area where the air purification terminal is located, and an image sample with three-dimensional parameters of the area is constructed by combining a distance sensor; the image samples with three-dimensional parameters are stored in the storage module, and the image samples with the three-dimensional parameters are transmitted to the control module;
when entering a learning mode, the purification module starts all purification modes for air purification, acquires an air index of an area where an air purification terminal is located, and transmits the air index to the control module;
the control module compares the air indexes with a set threshold, marks the air indexes as key indexes when a certain air index exceeds the set threshold, corresponds the key indexes with the positions of image samples of the area, and performs machine learning to train the image samples to obtain a movable area; the depth image module exits from the learning mode when the depth image module is acquired into the movable region for the third time and the number of the same image samples exceeds 80%;
after the learning mode exits, the independent internet of things terminal is used for transmitting the acquired movement instruction to the control module, and the air purification terminal automatically moves to the movement instruction marking point position to perform air purification of key indexes.
7. The internet of things air purification system for machine learning based on placement position as claimed in claim 6, wherein the internet of things gateway module is used for being connected with a mobile terminal through identity authentication and submitting internet weather parameters acquired by the mobile terminal, the internet of things gateway module sends the internet weather parameters to the control module, after all purification modes are started for air purification and air indexes of an area where the air purification terminal is located are acquired, the control module performs influence factor fitting according to the weather parameters, and influence factors of the weather parameters include: air humidity, temperature, return to south index.
8. The internet of things air purification system for machine learning based on placement according to claim 7, wherein the control module reduces the weight of a dehumidification index in a key index when the return-to-south index in the acquired influence factors of the weather parameters is high; and the purification module performs air purification according to the weight.
9. The internet of things air purification system for machine learning based on placement location of claim 6, wherein the control module further comprises a map unit for obtaining the movement instructions and the movable area; after a plurality of movable areas are obtained, whether different movable areas are connected or not is judged, if no connected part exists between the movable areas, a reminding instruction is sent, and the reminding instruction is transmitted to a mobile terminal connected through identity authentication by the internet of things gateway, so that the existence of two movable areas which cannot be spanned is indicated.
10. The internet of things air purification system for machine learning based on placement position according to claim 9, wherein after the two movable zones which cannot be spanned are connected, the map unit sends confirmation information, the confirmation information includes whether an obstacle door or a step obstacle exists in the movable zones which cannot be spanned, if the obstacle door or the step obstacle exists, the map unit does not send out the reminding instruction, and two independent movable zones which are not connected with each other are formed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110246910.8A CN113091223B (en) | 2021-03-05 | 2021-03-05 | Internet of things air purification method and system for machine learning based on placement position |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110246910.8A CN113091223B (en) | 2021-03-05 | 2021-03-05 | Internet of things air purification method and system for machine learning based on placement position |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113091223A true CN113091223A (en) | 2021-07-09 |
| CN113091223B CN113091223B (en) | 2022-10-04 |
Family
ID=76666976
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110246910.8A Active CN113091223B (en) | 2021-03-05 | 2021-03-05 | Internet of things air purification method and system for machine learning based on placement position |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113091223B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113834182A (en) * | 2021-11-26 | 2021-12-24 | 深圳市嘉达美智能家居有限公司 | Intelligent household equipment control method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0893308A2 (en) * | 1997-07-22 | 1999-01-27 | Kabushiki Kaisha Equos Research | Device mounted in vehicle |
| KR20090088187A (en) * | 2008-02-14 | 2009-08-19 | 주식회사 우주엘텍 | Air cleaner |
| CN104180486A (en) * | 2014-08-14 | 2014-12-03 | 南通博云物联网技术有限公司 | Intelligent home system and control method |
| CN107842952A (en) * | 2017-11-08 | 2018-03-27 | 蚌埠市瑞风净化设备工程有限责任公司 | A kind of intelligent movable air clarifier |
| KR102151548B1 (en) * | 2019-06-05 | 2020-09-03 | 김민호 | Air cleaner |
| CN111795447A (en) * | 2020-07-17 | 2020-10-20 | 韶关市诚湃新能源科技有限公司 | Air purifier based on big data and air quality analysis method |
-
2021
- 2021-03-05 CN CN202110246910.8A patent/CN113091223B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP0893308A2 (en) * | 1997-07-22 | 1999-01-27 | Kabushiki Kaisha Equos Research | Device mounted in vehicle |
| KR20090088187A (en) * | 2008-02-14 | 2009-08-19 | 주식회사 우주엘텍 | Air cleaner |
| CN104180486A (en) * | 2014-08-14 | 2014-12-03 | 南通博云物联网技术有限公司 | Intelligent home system and control method |
| CN107842952A (en) * | 2017-11-08 | 2018-03-27 | 蚌埠市瑞风净化设备工程有限责任公司 | A kind of intelligent movable air clarifier |
| KR102151548B1 (en) * | 2019-06-05 | 2020-09-03 | 김민호 | Air cleaner |
| CN111795447A (en) * | 2020-07-17 | 2020-10-20 | 韶关市诚湃新能源科技有限公司 | Air purifier based on big data and air quality analysis method |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113834182A (en) * | 2021-11-26 | 2021-12-24 | 深圳市嘉达美智能家居有限公司 | Intelligent household equipment control method |
| CN113834182B (en) * | 2021-11-26 | 2022-02-08 | 深圳市嘉达美智能家居有限公司 | Intelligent household equipment control method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113091223B (en) | 2022-10-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN206710818U (en) | A kind of intelligent home control system based on ZigBee and cloud computing | |
| US20170154517A1 (en) | Air quality alert system and method | |
| CN104596580A (en) | Environmental monitoring system based on wireless home-appliance network | |
| KR101972227B1 (en) | Smart home controlling apparatus based intellect learning and method thereof | |
| CN108829074A (en) | A kind of artificial intelligence domestic environment system control method based on big data | |
| CN205137798U (en) | Energy -conserving temperature control device | |
| CN104360607A (en) | Intelligent appliance system and intelligent appliance control method | |
| CN106950847A (en) | A kind of intelligent home control system based on ZigBee and cloud computing | |
| CN107461860B (en) | A kind of intelligent fresh air purification system and its control method | |
| CN203338177U (en) | Indoor temperature and humidity adjusting system based on Internet of things | |
| CN103389710B (en) | A kind of environment of internet of things control method based on context aware and equipment | |
| CN208186713U (en) | A kind of domestic environment intelligence regulation and control system | |
| CN205606846U (en) | Long -range air purification system based on internet | |
| CN108843176B (en) | Control method and control device of intelligent window, air purifier and water dispenser | |
| KR20200092807A (en) | smart window | |
| CN201417392Y (en) | Intelligent home system based on LonWorks | |
| CN113091223B (en) | Internet of things air purification method and system for machine learning based on placement position | |
| KR20180138100A (en) | Indoor environmental quality monitoring sensor device | |
| CN205351662U (en) | Air -conditioning control system | |
| CN205540013U (en) | Smart home system | |
| CN205579790U (en) | Movable intelligent air purifier and system | |
| CN207700958U (en) | A kind of Multifunctional intelligent window | |
| CN209086697U (en) | Artificial intelligence domestic environment management system based on big data | |
| CN201637032U (en) | Fresh air conditioner control system | |
| KR102546739B1 (en) | Cleaning system and method for indoor environment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220830 Address after: Room 618, Xize Building, No. 8 Luonan Avenue, Luoshe Town, Huishan District, Wuxi City, Jiangsu Province 214000 Applicant after: Jiangsu Huixin Intellectual Property Service Co.,Ltd. Address before: 312000 room 421, building 4, Shaoxing Sanjiang Supply Chain Management Co., Ltd., renjiafan village, Lanting street, Keqiao District, Shaoxing City, Zhejiang Province Applicant before: Shaoxing zhimingcao Technology Co.,Ltd. |
|
| TA01 | Transfer of patent application right | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |