Method and system for detecting existence of human body in hotel room based on multi-source perception

Abstract

The embodiment of the invention discloses a method and a system for detecting human body existence in a hotel guest room based on multi-source perception, which are applied to a multi-source sensor and comprise the following steps: collecting multi-source sensing data in the guest room in real time; preprocessing the multi-source sensing data, and removing noise data; fusing the preprocessed multi-source sensing data and carrying out feature extraction; inputting the data features into a pre-trained machine learning model for feature analysis, and obtaining an analysis feature result; and judging whether the feature result is that no person exists in the guest room, if so, sending an instruction to close unnecessary equipment and sending a guest room cleaning notice, otherwise, keeping the current running state and continuing monitoring. According to the embodiment of the invention, by fusing various sensor data, the accuracy and reliability of hotel guest room unmanned detection are improved, the problems of high misjudgment rate, great environmental influence, privacy risk and the like in the existing detection technology are solved, intelligent management of hotel guest rooms is realized, and the hotel service quality and the energy utilization efficiency are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of intelligent detection technology, and in particular to a method and system for detecting human presence in a hotel room based on multi-source perception. Background Art

[0002] In hotel operation and management, it is of great significance to accurately determine whether there are people in the guest room. On the one hand, for room service and energy management, if the hotel can know the vacancy of the guest room in time, it can arrange room cleaning service and efficiently complete the room cleaning work without disturbing the guests; at the same time, it can turn off unnecessary electrical equipment in time, such as lights and air conditioners, to effectively reduce energy consumption and reduce operating costs. On the other hand, from a safety perspective, accurate unmanned detection helps the hotel to promptly discover potential safety hazards after the guests leave the room, such as electrical appliances that are not turned off, fire sources that are not extinguished, etc., to ensure the property safety of the guests and the normal operation of the hotel.

[0003] At present, the common hotel room unmanned detection technologies are mainly the following: Detection technology based on infrared sensors, using infrared rays emitted by the human body to detect the presence of the human body. However, this technology has obvious defects. When the human body is in a stationary state, the infrared sensor may not be able to accurately identify it, and it is easy to make misjudgments; moreover, the infrared sensor is easily disturbed by factors such as ambient temperature and light, and the detection accuracy is greatly reduced in complex environments. Image recognition technology based on cameras, by analyzing the images taken by the camera to determine whether there is someone in the room. However, this technology faces serious privacy issues. Guests usually want to enjoy a high degree of privacy in hotel rooms. The presence of cameras may cause guests' disgust and concerns, resulting in many concerns in the hotel when using this technology. There are also detection technologies based on Bluetooth or Wi-Fi signals, which determine whether there is someone by detecting the signal of the mobile device in the guest room. However, this method is limited by the signal strength and coverage of the device, and cannot be accurately detected when the guest does not carry a device that supports Bluetooth or Wi-Fi, or the device is in flight mode and other special circumstances. Summary of the invention

[0004] To this end, an embodiment of the present invention provides a method and system for detecting human presence in hotel rooms based on multi-source perception, so as to solve the problems of high misjudgment rate, great influence of environment and easy infringement of guest privacy in existing detection technology, making it difficult to realize intelligent management of hotel rooms.

[0005] In order to achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0006] According to a first aspect of an embodiment of the present invention, a method for detecting human presence in a hotel room based on multi-source perception is provided. The method is applied to a multi-source sensor, and the multi-source sensor is composed of a millimeter wave radar sensor, an infrared sensor, a door magnetic sensor, a pressure sensor, and a sound sensor, and specifically includes:

[0007] S1, real-time collection of first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data in the guest room;

[0008] S2, pre-processing the collected first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data respectively, removing noise data, and generating second millimeter-wave radar sensor data, second infrared sensor data, second door magnetic sensor data, second pressure sensor data and second sound sensor data;

[0009] S3, using a data fusion algorithm to fuse the pre-processed second millimeter-wave radar sensor data, the second infrared sensor data, the second door magnetic sensor data, the second pressure sensor data, and the second sound sensor data to generate fused data, and performing feature extraction on the fused data to extract data features;

[0010] S4, inputting the data features into a pre-trained machine learning model for feature analysis to obtain analysis feature results;

[0011] S5. Determine whether the analysis result indicates that the room is empty. If the room is empty, send a command to turn off the lights, television and air conditioning equipment in the room and issue a room cleaning notice. If the room is empty, maintain the current operating state and continue monitoring.

[0012] Furthermore, real-time collection of first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data, and first sound sensor data in the guest room includes:

[0013] Millimeter-wave radar sensing data includes the movement trajectory information, movement speed information, and distance information of objects in the room;

[0014] Infrared sensor data is the infrared information emitted by the human body;

[0015] The door magnetic sensor data is the door's opening and closing power information;

[0016] Pressure sensing data is information about pressure changes;

[0017] The sound sensor data includes the sound intensity and frequency information in the guest room.

[0018] Furthermore, the preprocessing includes denoising and filtering to identify and delete invalid data.

[0019] Furthermore, the Kalman filter algorithm is used to remove noise interference in the millimeter wave radar sensor data to make the data curve smooth;

[0020] The ambient temperature is obtained and the infrared sensing data is adaptively adjusted according to the ambient temperature to compensate for the influence of temperature changes on the detection results.

[0021] Furthermore, the data fusion algorithm is DS evidence theory.

[0022] Furthermore, feature extraction is performed on the fused data to extract data features, including:

[0023] The data features include the frequency, amplitude and duration of human body movements.

[0024] Furthermore, the method further comprises:

[0025] Acquire a certain amount of historical data from a database and extract features from a feature library to train the machine learning model, thereby forming a pre-trained machine learning model;

[0026] The machine learning model is a support vector machine model or a neural network model.

[0027] Furthermore, the method further comprises:

[0028] When a sensor failure or abnormal data fluctuation occurs, an alarm is issued and auxiliary judgment is made through backup sensors or historical data.

[0029] According to a second aspect of an embodiment of the present invention, a human presence detection system in a hotel room based on multi-source sensing is provided, the system comprising:

[0030] A collection module, used for collecting first millimeter wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data in the guest room in real time;

[0031] A preprocessing module is used to preprocess the collected first millimeter-wave radar sensor data, the first infrared sensor data, the first door magnetic sensor data, the first pressure sensor data and the first sound sensor data, remove noise data, and generate second millimeter-wave radar sensor data, second infrared sensor data, second door magnetic sensor data, second pressure sensor data and second sound sensor data;

[0032] a feature extraction module, for fusing the preprocessed second millimeter-wave radar sensor data, the second infrared sensor data, the second door magnetic sensor data, the second pressure sensor data, and the second sound sensor data to generate fused data by using a data fusion algorithm, and performing feature extraction on the fused data to extract data features;

[0033] An analysis module is used to input the data features into a pre-trained machine learning model to perform feature analysis and obtain analysis feature results;

[0034] The decision and control module is used to determine whether the analysis feature result indicates that the room is empty. If the room is empty, a command is sent to turn off the lights, television and air-conditioning equipment in the room and a room cleaning notice is issued. If the room is occupied, the current operating state is maintained and monitoring continues.

[0035] Furthermore, the system further includes a collection module 1, specifically including:

[0036] The multi-source sensor is composed of a millimeter wave radar sensor, an infrared sensor, a door magnetic sensor, a pressure sensor and a sound sensor;

[0037] The millimeter wave radar sensor is installed at the center of the room ceiling to accurately detect the movement, position and static state of the human body;

[0038] The infrared sensors are distributed in every corner of the guest room to assist in detecting the thermal radiation signals of the human body;

[0039] The door magnetic sensor is installed on the guest room door to monitor the door's open and closed state;

[0040] The pressure sensor is installed under the mattress to detect whether there is someone on the bed;

[0041] The sound sensor is used to collect sound signals in the room.

[0042] The embodiments of the present invention have the following advantages:

[0043] The embodiment of the present invention is applied to a multi-source sensor, specifically comprising: S1, real-time collection of first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data in a guest room; S2, pre-processing the collected first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data respectively, removing noise data, and generating second millimeter-wave radar sensor data, second infrared sensor data, second door magnetic sensor data, second pressure sensor data and second sound sensor data; S3, using a data fusion algorithm to calculate the first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data; The method fuses the pre-processed second millimeter-wave radar sensor data, the second infrared sensor data, the second door magnetic sensor data, the second pressure sensor data and the second sound sensor data to generate fused data, performs feature extraction on the fused data, and extracts data features; S4, inputs the data features into the pre-trained machine learning model for feature analysis, and obtains the analysis feature results; S5, determines whether the analysis feature results indicate that the guest room is unoccupied. If the guest room is unoccupied, sends a command to turn off the lights, television and air-conditioning equipment in the room and issues a room cleaning notification. If the guest room is occupied, maintains the current operating state and continues monitoring. The embodiment of the present invention improves the accuracy and reliability of unoccupied detection of hotel guest rooms by fusing multiple sensor data, solves the problems of high misjudgment rate, great environmental impact and privacy risks in existing detection technologies, realizes intelligent management of hotel guest rooms, and improves hotel service quality and energy efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the implementation methods of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the implementation methods or the description of the prior art. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other implementation drawings can be derived from the provided drawings without creative work.

[0045] The structures, proportions, sizes, etc. illustrated in this specification are only used to match the contents disclosed in the specification so as to facilitate understanding and reading by persons familiar with the technology. They are not used to limit the conditions under which the present invention can be implemented, and therefore have no substantial technical significance. Any structural modification, change in proportion or adjustment of size shall still fall within the scope of the technical contents disclosed in the present invention without affecting the effects and purposes that can be achieved by the present invention.

[0046] Figure 1 A schematic diagram of the logical structure of a human presence detection system in a hotel room based on multi-source perception provided by an embodiment of the present invention;

[0047] Figure 2A schematic diagram of a flow chart of a method for detecting human presence in a hotel room based on multi-source sensing provided by an embodiment of the present invention;

[0048] Figure 3 A schematic diagram of the overall architecture of a human presence detection system in a hotel room based on multi-source perception provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The following is a description of the implementation of the present invention by specific embodiments. People familiar with the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. Obviously, the described embodiments are 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.

[0050] In hotel operation and management, it is of great significance to accurately determine whether there is no one in the guest room. On the one hand, for room service and energy management, if the hotel can know the vacant state of the guest room in time, it can arrange room cleaning service and efficiently complete the room tidying work without disturbing the guests; at the same time, it can turn off unnecessary electrical equipment in time, such as lights and air conditioners, to effectively reduce energy consumption and reduce operating costs. On the other hand, from a safety perspective, accurate unmanned detection helps the hotel to promptly discover potential safety hazards after the guests leave the room, such as electrical appliances that are not turned off, fire sources that are not extinguished, etc., to ensure the property safety of the guests and the normal operation of the hotel.

[0051] At present, the common hotel room unmanned detection technologies are mainly the following: Detection technology based on infrared sensors, using infrared rays emitted by the human body to detect the presence of the human body. However, this technology has obvious defects. When the human body is in a stationary state, the infrared sensor may not be able to accurately identify it, and it is easy to make misjudgments; moreover, the infrared sensor is easily disturbed by factors such as ambient temperature and light, and the detection accuracy is greatly reduced in complex environments. Image recognition technology based on cameras, by analyzing the images taken by the camera to determine whether there is someone in the room. However, this technology faces serious privacy issues. Guests usually want to enjoy a high degree of privacy in hotel rooms. The presence of cameras may cause guests' disgust and concerns, resulting in many concerns in the hotel when using this technology. There are also detection technologies based on Bluetooth or Wi-Fi signals, which determine whether there is someone by detecting the signal of the mobile device in the guest room. However, this method is limited by the signal strength and coverage of the device, and cannot be accurately detected when the guest does not carry a device that supports Bluetooth or Wi-Fi, or the device is in flight mode and other special circumstances.

[0052] With the continuous development of Internet of Things technology, sensor technology and data analysis technology, the technical idea of ​​multi-source perception fusion provides a new way to solve the problem of unmanned detection in hotel rooms. By integrating the data of multiple types of sensors and giving full play to the advantages of different sensors, it is possible to more accurately and comprehensively judge the status of people in the guest room and overcome the limitations of single sensor detection technology.

[0053] In order to solve the problems of high misjudgment rate, great environmental impact and easy invasion of guest privacy in the above detection technology, it is difficult to realize intelligent management of hotel rooms.

[0054] refer to Figure 1 The embodiment of the present invention discloses a human presence detection system in a hotel room based on multi-source perception, the system comprising: an acquisition module 1; a preprocessing module 2; a feature extraction module 3; an analysis module 4; and a decision and control module 5.

[0055] Furthermore, the system also includes an acquisition module 1, which specifically includes: the multi-source sensor is composed of a millimeter-wave radar sensor, an infrared sensor, a door magnetic sensor, a pressure sensor and a sound sensor; the millimeter-wave radar sensor is installed at the center of the room ceiling to accurately detect the movement, position and static state of the human body; the infrared sensor is distributed in various corners of the guest room to assist in detecting the thermal radiation signal of the human body; the door magnetic sensor is installed on the guest room door to monitor the switch status of the door; the pressure sensor is installed under the mattress to detect whether there is someone on the bed; the sound sensor is used to collect sound signals in the room.

[0056] According to the design plan, install the millimeter wave radar sensor in the center of the guest room ceiling to ensure that it can fully cover the room space and accurately detect various states of the human body. Distribute the infrared sensors evenly in the four corners of the room and key locations, such as the head of the bed and the bathroom door, to ensure that the human body heat radiation signals in various areas of the room can be effectively detected. Install the door magnetic sensor on the guest room door to ensure that it can accurately sense the door opening and closing actions. Place the pressure sensor under the mattress. Install the sound sensor in a suitable position in the room, such as high on the wall, which can effectively collect sound signals without affecting the normal activities of guests. During the installation process, pay attention to the installation angle and direction of the sensor to avoid obstruction and interference to ensure that the sensor can work properly.

[0057] The acquisition module 1 is connected to the pre-processing module 2 by wire or wireless means to ensure the stability and timeliness of data transmission. The pre-processing module 2 is connected to the analysis module 4 and the decision and control module 5 through a high-speed network to achieve rapid data transmission and processing. According to the hotel's network architecture, the system is connected to the hotel's internal LAN to ensure that the system can exchange data and work collaboratively with other related systems such as the hotel's energy management system and room service system.

[0058] Install the corresponding software programs in the preprocessing module 2, the analysis module 4 and the decision and control module 5. Set parameters for the preprocessing module 2, such as data collection frequency, preprocessing algorithm parameters, etc. Train and optimize the machine learning model for the analysis module 4, use a large amount of historical data and actual scene data to train the model, adjust the model parameters, and improve the accuracy and generalization ability of the model; Set control strategies and rules for the decision and control module 5, and formulate equipment control instructions and service arrangement strategies in the unmanned state according to the hotel's operating needs and service standards.

[0059] After the system is installed and configured, it can be put into daily operation. The acquisition module 1 collects data in the guest room in real time and transmits the data to the data processing module for preprocessing. The preprocessed data is sent to the data analysis module for fusion analysis to determine whether there is anyone in the guest room. The control decision module sends control instructions and service notifications to related systems based on the analysis results. Hotel staff can monitor the system's operating status in real time through the system management interface, view the guest room unmanned detection results and related data reports.

[0060] Regularly inspect and maintain multi-source sensors, including cleaning the sensor surface, checking whether the installation position is loose, testing the sensor's detection performance, etc. Replace aging or damaged sensors in a timely manner according to the sensor's service life and performance changes. Regularly inspect system hardware equipment to ensure the normal operation of hardware such as servers and data processing equipment, and handle hardware failures and network problems in a timely manner.

[0061] According to the actual operation of the hotel and the feedback from guests, the system algorithm and control strategy are continuously optimized. For example, according to the changes in the behavior habits of guests in different seasons and time periods, the parameters of the machine learning model and the data fusion algorithm are adjusted to improve the accuracy and adaptability of unmanned detection. The system software is regularly upgraded to obtain new functions and performance optimizations, such as adding support for new sensor types and improving data analysis algorithms. During the system upgrade process, attention should be paid to data backup and compatibility testing to ensure the stability of the system and the security of the data.

[0062] The embodiment of the present invention also sets up an abnormal situation detection mechanism. When sensor failure, abnormal data fluctuation, etc. occur, an alarm is issued in time, and auxiliary judgment is made through backup sensors or historical data to ensure the accuracy and continuity of unmanned detection. For example, when an infrared sensor fails, the system automatically increases the data collection frequency of other infrared sensors and millimeter wave radar sensors, and uses redundant data for analysis and judgment.

[0063] The system monitors the working status of multi-source sensors in real time, including the sensor's signal strength, battery power, fault alarm and other information. Regularly calibrate and maintain the sensors to ensure the sensor's detection accuracy and stability. When a sensor fails, promptly notify maintenance personnel to replace or repair it.

[0064] Establish a data storage and management mechanism to store data collected by multi-source sensors, processed data, and analysis results for a long time. Use data mining and analysis technology to conduct in-depth mining of historical data, analyze guests' behavior habits, energy consumption patterns, etc., and provide decision support for hotel operations and management. For example, according to guests' check-in habits, optimize the intelligent control strategy of guest room equipment to improve guest satisfaction.

[0065] Different user permissions are set for hotel staff in different positions. Managers have the highest permissions and can fully configure the system, view and analyze data; room service staff can only view room vacancy information and receive service tasks; maintenance staff can only maintain and manage sensor equipment to ensure system security and data confidentiality.

[0066] refer to Figure 3 , the system in the embodiment of the present invention is composed of a control unit 103, a human body sensing unit 204, a network transmission unit 306, the Internet 401, a cloud server unit 501, a door magnetic sensor unit 601, a pressure sensor unit 701 and a sound sensor unit 801;

[0067] The control unit 103 includes a control unit Bluetooth mesh communication module 101 and a power on / off module 102;

[0068] The human body sensing unit 204 includes a human body sensing unit Bluetooth mesh communication module 201, an infrared sensor 202 and a microwave radar sensor 203;

[0069] The network transmission unit 306 includes a WiFi networking module 301, an information interaction module 302, a core processing module 303, a Bluetooth mesh network coordinator 304 and a data storage logic operation 305, wherein the WiFi networking module 301 and the information interaction module 302 communicate bidirectionally, the information interaction module 302 and the core processing module 303 communicate bidirectionally, the core processing module 303 and the Bluetooth mesh network coordinator 304 communicate bidirectionally, and the data storage logic operation 305 communicates bidirectionally with the information interaction module 302 and the core processing module 303 respectively;

[0070] The cloud server unit 501 is composed of a cloud interface 502, a room service system 503, a control decision module 504, a data analysis module 505 and a data processing module 506. The data processing module 506 sends the data to be sent to the room service system 503 through the data analysis module 505 and the control decision module 504 in sequence. The room service system 503 sends the data to the cloud interface 502, and the cloud interface sends the data stored in the cloud to the data processing module 506.

[0071] The door magnetic sensor unit 601 is composed of a Bluetooth mesh communication module and a door magnetic sensor;

[0072] The pressure sensor unit 701 is composed of a Bluetooth mesh communication module and a pressure sensor;

[0073] The sound sensor unit 801 is composed of a Bluetooth mesh communication module and a sound sensor.

[0074] Corresponding to the above-disclosed system for detecting the presence of a human in a hotel room based on multi-source sensing, an embodiment of the present invention further discloses a method for detecting the presence of a human in a hotel room based on multi-source sensing. The following describes in detail a method for detecting the presence of a human in a hotel room based on multi-source sensing disclosed in an embodiment of the present invention in combination with the above-described system for detecting the presence of a human in a hotel room based on multi-source sensing.

[0075] refer to Figure 2 The present invention discloses a method for detecting human presence in a hotel room based on multi-source perception. The method is applied to a multi-source sensor, which is composed of a millimeter wave radar sensor, an infrared sensor, a door magnetic sensor, a pressure sensor and a sound sensor, and specifically includes:

[0076] S1. Collecting first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data in the guest room in real time.

[0077] Multi-source sensors collect various data in the guest room in real time. Millimeter-wave radar sensors continuously monitor the movement trajectory, speed and distance information of objects in the room, and obtain the precise position and activity status of the human body; infrared sensors detect infrared rays emitted by the human body and perceive the approximate position of the human body; door magnetic sensors record the opening and closing actions of the door; pressure sensors detect whether there is a pressure change on the bed; sound sensors capture information such as the sound intensity and frequency in the room.

[0078] S2. Preprocess the collected first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data respectively, remove noise data, and generate second millimeter-wave radar sensor data, second infrared sensor data, second door magnetic sensor data, second pressure sensor data and second sound sensor data.

[0079] The collected raw data is pre-processed by denoising and filtering to remove invalid data caused by sensor errors, environmental interference, etc., to improve the accuracy and reliability of the data. For example, for millimeter-wave radar sensor data, the Kalman filter algorithm is used to remove noise interference and smooth the data curve; for infrared sensor data, adaptive adjustments are made according to the ambient temperature to compensate for the impact of temperature changes on the detection results.

[0080] S3, using a data fusion algorithm to fuse the pre-processed second millimeter-wave radar sensor data, the second infrared sensor data, the second door magnetic sensor data, the second pressure sensor data, and the second sound sensor data to generate fused data, and performing feature extraction on the fused data to extract data features;

[0081] S4. Input the data features into a pre-trained machine learning model for feature analysis to obtain analysis feature results.

[0082] The pre-processed multi-source sensor data is fused using data fusion algorithms such as DS evidence theory. The fused data is subjected to feature extraction, such as the frequency, amplitude, and duration of human motion, and then input into a pre-trained machine learning model (such as support vector machine, neural network, etc.) for analysis. The machine learning model learns the feature patterns of different personnel states based on historical data and feature libraries, thereby determining whether there is currently someone in the room.

[0083] S5. Determine whether the analysis result indicates that the room is empty. If the room is empty, send a command to turn off the lights, television and air conditioning equipment in the room and issue a room cleaning notice. If the room is empty, maintain the current operating state and continue monitoring.

[0084] Make a decision based on the judgment result of the analysis module 4. If it is judged that there is no one in the room, send a command to the energy management system to turn off non-essential electrical equipment in the room, such as lights, televisions, air conditioners, etc.; at the same time, send a notification to the room service system to arrange room cleaning service. If it is judged that there is someone in the room, keep the equipment in normal operation and monitor the changes in the status of the person in real time.

[0085] Furthermore, the first millimeter-wave radar sensor data, the first infrared sensor data, the first door magnetic sensor data, the first pressure sensor data and the first sound sensor data in the guest room are collected in real time, including: the millimeter-wave radar sensor data includes the motion trajectory information, motion speed information and distance information of objects in the room; the infrared sensor data is the infrared information emitted by the human body; the door magnetic sensor data is the opening and closing power information of the door; the pressure sensor data is the pressure change information; the sound sensor data includes the sound intensity and frequency information in the guest room.

[0086] Furthermore, the preprocessing includes denoising and filtering to identify and delete invalid data.

[0087] Furthermore, the Kalman filter algorithm is used to remove noise interference in the millimeter wave radar sensor data to make the data curve smooth; the ambient temperature is obtained and the infrared sensor data is adaptively adjusted according to the ambient temperature to compensate for the influence of temperature changes on the detection results.

[0088] Furthermore, the data fusion algorithm is DS evidence theory.

[0089] Furthermore, feature extraction is performed on the fused data to extract data features, including: the data features include frequency, amplitude and duration of human body movement.

[0090] Furthermore, the method also includes: acquiring a certain amount of historical data from a database and extracting features from a feature library to train the machine learning model to form a pre-trained machine learning model; the machine learning model is a support vector machine model or a neural network model.

[0091] Furthermore, the method also includes: when a sensor failure or abnormal data fluctuation occurs, an alarm is issued and auxiliary judgment is performed through backup sensors or historical data.

[0092] The embodiments of the present invention have the following advantages:

[0093] 1) Improve detection accuracy: By integrating data from multiple sensors, the advantages of different sensors are fully utilized to make up for the shortcomings of a single sensor, effectively improving the accuracy and reliability of guest room vacancy detection and reducing misjudgments and missed judgments.

[0094] 2) Protect guest privacy: The system does not rely on devices such as cameras that may infringe on guest privacy. It uses non-visual sensors such as millimeter-wave radar and infrared sensors to protect guest privacy to the greatest extent while ensuring detection functions.

[0095] 3) Reduce energy consumption: Timely and accurate unattended detection can realize intelligent control of guest room equipment, shut down unnecessary electrical equipment when no one is at home, effectively reduce the hotel's energy consumption, reduce operating costs, and achieve energy conservation and emission reduction goals.

[0096] 4) Improve service quality: According to the unmanned detection results, reasonably arrange room cleaning services, improve service efficiency, reduce disturbance to guests, and improve guests' stay experience and satisfaction.

[0097] 5) Easy to expand and upgrade the system: The system adopts a modular design, which is easy to expand with new sensors or functional modules. With the development of technology and changes in hotel needs, the system can be easily upgraded and optimized to improve the hotel's intelligent management level.

[0098] Although the present invention has been described in detail above by general description and specific embodiments, it is obvious to those skilled in the art that some modifications or improvements can be made to the present invention. Therefore, these modifications or improvements made without departing from the spirit of the present invention all belong to the scope of protection claimed by the present invention.

Claims

1. A method for detecting human presence in hotel rooms based on multi-source perception, characterized in that: The method is applied to a multi-source sensor, which is composed of a millimeter wave radar sensor, an infrared sensor, a door magnetic sensor, a pressure sensor, and a sound sensor, and specifically includes: S1, real-time collection of first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data in the guest room; S2, pre-processing the collected first millimeter-wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data respectively, removing noise data, and generating second millimeter-wave radar sensor data, second infrared sensor data, second door magnetic sensor data, second pressure sensor data and second sound sensor data; S3, using a data fusion algorithm to fuse the pre-processed second millimeter-wave radar sensor data, the second infrared sensor data, the second door magnetic sensor data, the second pressure sensor data, and the second sound sensor data to generate fused data, and performing feature extraction on the fused data to extract data features; S4, inputting the data features into a pre-trained machine learning model for feature analysis to obtain analysis feature results; S5. Determine whether the analysis result indicates that the room is empty. If the room is empty, send a command to turn off the lights, television and air conditioning equipment in the room and issue a room cleaning notice. If the room is empty, maintain the current operating state and continue monitoring.

2. A method for detecting human presence in a hotel room based on multi-source sensing as claimed in claim 1, characterized in that: The first millimeter wave radar sensor data, the first infrared sensor data, the first door magnetic sensor data, the first pressure sensor data and the first sound sensor data in the guest room are collected in real time, including: Millimeter wave radar sensing data includes the movement trajectory information, movement speed information and distance information of objects in the room; Infrared sensor data is the infrared information emitted by the human body; The door magnetic sensor data is the door's opening and closing power information; Pressure sensing data is information about pressure changes; The sound sensor data includes the sound intensity and frequency information in the guest room.

3. A method for detecting human presence in a hotel room based on multi-source sensing as claimed in claim 2, characterized in that: The preprocessing includes denoising and filtering to identify and delete invalid data.

4. A method for detecting human presence in a hotel room based on multi-source sensing as claimed in claim 3, characterized in that: The Kalman filter algorithm is used to remove noise interference in the millimeter-wave radar sensor data to make the data curve smooth; The ambient temperature is obtained and the infrared sensing data is adaptively adjusted according to the ambient temperature to compensate for the influence of temperature changes on the detection results.

5. A method for detecting human presence in a hotel room based on multi-source sensing as claimed in claim 4, characterized in that: The data fusion algorithm is DS evidence theory.

6. A method for detecting human presence in a hotel room based on multi-source sensing as claimed in claim 5, characterized in that: Performing feature extraction on the fused data to extract data features includes: The data features include the frequency, amplitude and duration of human body movements.

7. A method for detecting human presence in a hotel room based on multi-source sensing as claimed in claim 6, characterized in that: The method further comprises: Acquire a certain amount of historical data from a database and extract features from a feature library to train the machine learning model, thereby forming a pre-trained machine learning model; The machine learning model is a support vector machine model or a neural network model.

8. A method for detecting human presence in a hotel room based on multi-source sensing as claimed in claim 7, characterized in that: The method further comprises: When a sensor failure or abnormal data fluctuation occurs, an alarm is issued and auxiliary judgment is made through backup sensors or historical data.

9. A human presence detection system in hotel guest rooms based on multi-source sensing, characterized in that: The system comprises: A collection module, used for collecting first millimeter wave radar sensor data, first infrared sensor data, first door magnetic sensor data, first pressure sensor data and first sound sensor data in the guest room in real time; A preprocessing module is used to preprocess the collected first millimeter-wave radar sensor data, the first infrared sensor data, the first door magnetic sensor data, the first pressure sensor data and the first sound sensor data, remove noise data, and generate second millimeter-wave radar sensor data, second infrared sensor data, second door magnetic sensor data, second pressure sensor data and second sound sensor data; a feature extraction module, for fusing the preprocessed second millimeter-wave radar sensor data, the second infrared sensor data, the second door magnetic sensor data, the second pressure sensor data, and the second sound sensor data to generate fused data by using a data fusion algorithm, and performing feature extraction on the fused data to extract data features; An analysis module is used to input the data features into a pre-trained machine learning model to perform feature analysis and obtain analysis feature results; The decision and control module is used to determine whether the analysis feature result indicates that the room is empty. If the room is empty, a command is sent to turn off the lights, television and air-conditioning equipment in the room and a room cleaning notice is issued. If the room is occupied, the current operating state is maintained and monitoring continues.

10. A human presence detection system in hotel guest rooms based on multi-source sensing as claimed in claim 9, characterized in that: The acquisition module specifically includes: The multi-source sensor consists of millimeter wave radar sensor, infrared sensor, door magnetic sensor, pressure sensor and sound sensor; The millimeter wave radar sensor is installed at the center of the room ceiling to accurately detect the movement, position and static state of the human body; The infrared sensors are distributed in every corner of the guest room to assist in detecting the thermal radiation signals of the human body; The door magnetic sensor is installed on the guest room door to monitor the door's open and closed state; The pressure sensor is installed under the mattress to detect whether there is someone on the bed; The sound sensor is used to collect sound signals in the room.

Images