Detection and evaluation method for gesture recognition control function of intelligent extractor hood

Abstract

The invention discloses a method for detecting and evaluating a gesture recognition control function of an intelligent extractor hood, which comprises the following steps of: controlling a bionic hand motion control module to simulate a gesture action of a human hand in front of the extractor hood through an upper computer, and acquiring the pressure of an air inlet of the extractor hood through an instrument data acquisition module; and then the upper computer automatically counts the recognition rate of the gesture recognition control function of the range hood according to the change of the pressure after each gesture action, and forms a detection report according to a template. According to the invention, through automatic statistics of the recognition rate which is one of the most important indexes of the gesture recognition control function of the intelligent extractor hood, a solution for automatically detecting and evaluating the gesture recognition control function of the intelligent extractor hood is provided for a third-party laboratory or an enterprise internal laboratory, so that the detection workload of testers can be reduced; the detection efficiency is improved and the practicability is high.

Description

Technical Field

[0001] The invention relates to a detection and evaluation method for a gesture recognition control function of an intelligent range hood. Background Art

[0002] With the continuous development and progress of science and technology and the innovation of human-computer communication methods, people's pursuit of smart homes and convenient lifestyles is increasing, which has also promoted the development of kitchen appliances such as range hoods towards intelligence and humanization. As an advanced human-computer interaction method, gesture recognition system is the focus of research by experts and scholars in the field of human-computer interaction. It is also the hottest field at present, and has a wide range of application scenarios. Gesture control means that the human hand does not need to directly touch the machine, but makes corresponding posture changes in the air, and then controls the machine to make corresponding actions. The use of gesture control of range hoods enables users to control the operation of range hoods through simple gestures, thereby improving the convenience of use and user experience. By applying gesture recognition technology to range hoods, users can control the switch, wind speed gear and other functions of the range hood by waving their hands. By controlling the range hood through gestures, there is no need to touch the buttons, which is convenient and hygienic, avoiding dirtying the control panel of the range hood, meeting the user's expectation of effective control of the range hood when the hands are greasy or inconvenient, and providing convenience and cleanliness of use.

[0003] The gesture recognition rate of a range hood is one of the most important indicators for measuring the intelligence of a range hood and its user experience. Currently, the industry focuses on the realization of gesture recognition control functions in products, and there are few detection and evaluation devices and methods. Therefore, it is very necessary to design a set of devices and methods to detect and evaluate the accuracy of gesture recognition control functions of smart range hoods. Summary of the invention

[0004] The object of the present invention is to provide a detection and evaluation method for the gesture recognition control function of an intelligent range hood.

[0005] The object of the present invention is achieved through the following technical scheme: a detection and evaluation method for the gesture recognition control function of an intelligent range hood, characterized in that it controls the bionic hand motion control module through a host computer to simulate the gesture movements of a human hand in front of the range hood, collects the pressure of the air inlet of the range hood through an instrument data acquisition module, and then the host computer automatically counts the recognition rate of the gesture recognition control function of the range hood according to the change of the pressure after each gesture movement, and forms a detection report according to a template.

[0006] The host computer is provided with a user parameter configuration interface, which supports user-configured parameters including the bionic hand waving direction such as up and down or left and right, the number of waving cycles, the waving time interval, the waving speed, and the distance between the bionic hand and the air inlet of the range hood. After the user completes the parameter setting and saves it, the host computer reconfigures the control logic of the bionic hand motion control module according to the above user settings, thereby conveniently and efficiently meeting the diverse requirements of the product's gesture control mode and working logic, and can adapt to different range hoods.

[0007] The present invention configures the distance parameter between the initial position of the bionic hand and the air inlet of the range hood, and then calculates the distance between the bionic hand and the air inlet of the range hood after the movement based on the distance parameter, thereby realizing program control of the distance.

[0008] The specific steps of automatically counting the recognition rate of the range hood gesture recognition control function according to the change of the pressure after each gesture action by the host computer are as follows:

[0009] 1) Set initial conditions, including:

[0010] Initial air inlet pressure P0; initial normal opening times x=0; initial normal closing times y=0; total initial opening times Nx=0; total initial closing times Ny=0; number of test cycles N;

[0011] 2) Test Control

[0012] 2.1) Start operation

[0013] a) controlling the bionic hand to swing out according to the user's setting of the swing direction of the bionic hand;

[0014] b) After waiting for a certain period of time, the current pressure value P1 is read. If it exceeds the tolerance range of P0, it is determined that the opening control operation is successful, the normal opening times x=x+1, and the total opening operation times Nx=Nx+1; if the range hood is not successfully opened, the bionic hand is controlled to reset to the initial position of the opening action, and the above opening operation is repeated;

[0015] 2.2) Close operation

[0016] a) Control the bionic hand to wave back;

[0017] b) After waiting for a certain period of time, the current pressure value P2 is read. If it exceeds the tolerance range of P1, it is determined that the closing control operation is successful, the normal closing number y=y+1, and the total number of closing operations Ny=Ny+1; if the range hood is not successfully closed, the bionic hand is controlled to reset to the initial position of the closing action, and the above closing operation is repeated;

[0018] 2.3) If both Nx and Ny reach the number of cycles N, the program ends, otherwise repeat steps 2.1) and 2.2);

[0019] 3) Recognition rate statistics

[0020] Successful recognition rate of range hood opening operation gesture recognition control function

[0021] Successful recognition rate of gesture recognition control function for range hood closing operation

[0022] The total recognition rate of the range hood gesture recognition control function

[0023] The present invention determines whether the operation is valid by comparing whether the pressure change after the gesture operation exceeds the tolerance range before the operation, thereby realizing accurate judgment of the working state of the range hood.

[0024] The bionic hand motion control module comprises a four-degree-of-freedom mechanical intelligent arm and a bionic hand mounted thereon. The bionic hand is driven by the four-degree-of-freedom mechanical intelligent arm to simulate human hand movements in front of the range hood.

[0025] The data acquisition module comprises a universal shaped tube and a pressure gauge mounted thereon, wherein the pressure gauge is supported by the universal shaped tube and arranged near the air inlet of the range hood.

[0026] Beneficial effects:

[0027] 1) The present invention provides a solution for automatically detecting and evaluating the gesture recognition control function of intelligent range hoods for third-party laboratories or internal laboratories of enterprises by automatically counting the recognition rate, one of the most important indicators of the gesture recognition control function of intelligent range hoods. This can reduce the detection workload of testers, improve detection efficiency, and has strong practicality.

[0028] 2) The present invention is an automatic detection and evaluation scheme that can conveniently detect the recognition rate of long-term continuous repetitive actions, thereby judging the performance stability and reliability of the gesture recognition control function of the smart range hood under continuous use.

[0029] 3) The present invention realizes user configuration of the waving direction, number of waving cycles, waving time interval, waving speed, and the distance between the bionic hand and the air inlet of the range hood, meets the diversity of the product's intelligent gesture control mode and working logic, ensures the comprehensiveness of the test, and can adapt to the control requirements of different range hoods.

[0030] 4) The present invention determines whether the operation is valid by comparing whether the pressure change after the gesture operation exceeds the tolerance range before the operation, thereby realizing accurate judgment of the working state of the range hood.

[0031] 5) The present invention realizes the automatic generation of the test report, which can reduce the workload of the test personnel and further improve the work efficiency of the test personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is a block diagram of the components of the detection and evaluation device used in the method of this embodiment;

[0033] Figure 2 It is an automated test processing logic diagram. DETAILED DESCRIPTION

[0034] The present invention will be further described in detail below in conjunction with specific examples and drawings. It should be noted that the following examples are only preferred embodiments of the present invention and should not be construed as limiting the embodiments and protections of the present invention.

[0035] Figure 1 is a block diagram of the components of the detection and evaluation device used in the method of this embodiment, such as Figure 1 As shown, it mainly includes a host computer, a bionic hand motion control module and an instrument data acquisition module. The bionic hand motion control module is mainly composed of a bionic hand driven by a four-degree-of-freedom mechanical intelligent arm. The instrument data acquisition module is mainly composed of a communication-type digital pressure gauge supported by a stainless steel universal shaping tube. The stainless steel universal shaping tube has great flexibility and can be rotated, bent and shaped at will. The above-mentioned host computer communicates with the intelligent arm through Uart and communicates with the pressure gauge through RS485.

[0036] The detection and evaluation method for the gesture recognition control function of the intelligent range hood of the present embodiment controls the bionic hand motion control module by the host computer to simulate the gesture movements of a human hand in front of the range hood, collects the pressure of the air inlet of the range hood by the instrument data acquisition module, and then the host computer automatically calculates the recognition rate of the gesture recognition control function of the range hood according to the change of the air inlet pressure of the range hood after each gesture movement, and forms a detection report according to the template.

[0037] The following is the specific configuration information of the software and hardware of this embodiment:

[0038] 1. Hardware Platform

[0039]

[0040] 2. Software Platform

[0041] operating system Windows 10 Development Environment Python development environment and tools

[0042] In this embodiment, multiple parameters can be flexibly configured through the user parameter configuration interface of the host computer, including the bionic hand waving direction (usually including the up and down direction and the left and right direction), the number of waving cycles (one round trip is one cycle, such as one up and down or left and right movement), the waving time interval, the waving speed, and the distance between the bionic hand and the air inlet of the range hood. After completing the parameter setting and saving, the host computer will reconfigure the control logic of the bionic hand motion control module according to the above user settings, so as to conveniently and conveniently meet the diversity of the product's gesture control mode and working logic, ensure the comprehensiveness of the test, and better adapt to the control requirements of different range hoods. By setting the number of waving cycles, this embodiment can conveniently realize the recognition rate detection of long-term continuous repetitive actions, so as to judge the performance stability and reliability of the gesture recognition control function of the intelligent range hood under continuous use.

[0043] This embodiment configures the distance parameter between the initial position of the bionic hand and the air inlet of the range hood, and then calculates the distance between the bionic hand and the air inlet of the range hood after the move based on the distance, thereby realizing program control of the distance.

[0044] This embodiment also generates a wind pressure curve diagram by automatically recording the wind pressure peak point during the sampling process, thereby monitoring the working status of the range hood in real time.

[0045] The host computer of this embodiment evaluates the recognition rate, stability and reliability of the on and off operation gesture recognition control function of the range hood in the following manner:

[0046] 1. Set the initial conditions, including setting:

[0047] a) Initial air inlet pressure: P0 (obtained by the user inputting the ambient atmospheric pressure in this embodiment)

[0048] b) Initial normal opening times: x = 0

[0049] c) Initial normal closing times: y = 0

[0050] d) Total number of initial start-up operations: Nx = 0

[0051] e) Total number of initial closing operations: Ny = 0

[0052] f) Test cycle number: N (assign the number of swing cycles set by the user to N)

[0053] 2. Test Controls

[0054] 2.1) Start operation

[0055] a) controlling the bionic hand to swing out according to the user's setting of the swing direction of the bionic hand;

[0056] b) After waiting for a certain period of time, the current pressure value P1 is read. If it exceeds the tolerance range of P0 (P0±10%), it is determined that the opening control operation is successful, the normal opening times x=x+1, and the total opening operation times Nx=Nx+1; if the range hood is not successfully opened, the bionic hand is controlled to reset to the initial position of the opening action, and the above opening operation is repeated;

[0057] 2.2) Close operation

[0058] a) Control the bionic hand to wave back;

[0059] b) After waiting for a certain period of time, the current pressure value P2 is read. If it exceeds the tolerance range of P1 (P1±10%), it is determined that the closing control operation is successful, the normal closing number y=y+1, and the total number of closing operations Ny=Ny+1; if the range hood is not successfully closed, the bionic hand is controlled to reset to the initial position of the closing action, and the above closing operation is repeated;

[0060] 2.3) If both Nx and Ny reach the number of cycles N, the program ends, otherwise repeat steps 2.1) and 2.2);

[0061] 3) Recognition rate statistics

[0062] Successful recognition rate of range hood opening operation gesture recognition control function

[0063] Successful recognition rate of gesture recognition control function for range hood closing operation

[0064] The total recognition rate of the range hood gesture recognition control function

[0065] The present invention determines whether the operation is valid by comparing whether the pressure change after the gesture operation exceeds the tolerance range before the operation, thereby realizing accurate judgment of the working state of the range hood.

[0066] The following takes a range hood product as an example to introduce its detection steps in detail. The flow chart is as follows Figure 2 As shown, for this range hood, the gesture is waved from left to right once, the range hood is turned on and starts to suck air, and the gesture is waved from right to left once, the range hood is turned off:

[0067] Step 1: Turn on the host computer power control switch and open the host computer test system software; Step 2: Enter user parameters

[0068] Including parameters such as the waving direction of the bionic hand, the number of waving cycles, the waving time interval, the waving speed, the distance between the bionic hand and the air inlet of the range hood, the ambient air pressure (103kpa), the report number, the sample number, etc.;

[0069] Step 3: Adjust the pressure point of the stainless steel universal shaped tube connected to the digital pressure gauge close to the air inlet of the range hood without hindering the movement of the bionic hand;

[0070] Step 4: Connect the power supply to the range hood to be tested;

[0071] Step 5: Click the "Start" button on the test system page of the host computer, and the host computer sends a command to start the test;

[0072] Step 6: The host computer initializes program variables, sets the normal opening times x and normal closing times y to 0, and sets the opening operation times Nx and closing operation times Ny to 0;

[0073] Step 7: The host computer issues an opening operation command, controls the bionic hand to wave from left to right once, and waits for Tx seconds so that the range hood can be fully opened to facilitate pressure detection. The completion of one opening operation is recorded, and Nx increases by 1;

[0074] Step 8: Dynamically monitor the air pressure change at the air inlet of the range hood. If the range hood fails to open successfully, control the bionic hand to reset to the initial position of the opening action and repeat step 7;

[0075] Step 9: If the action of turning on the range hood is a valid turning-on operation (Nx times that do not exceed the target action times N are considered valid turning-on operations), then a normal turning-on is recorded as completed, and x increases by 1.

[0076] Step 10: The host computer issues a closing operation command, controls the bionic hand to wave from right to left once, and waits for Ty seconds so that the range hood can be completely closed to facilitate pressure detection. It is recorded that one opening operation is completed, and Ny increases by 1;

[0077] Step 11: Dynamically monitor the air pressure change at the range hood air inlet. If the range hood is not successfully closed, control the bionic hand to reset to the initial position of the closing action and repeat step 10;

[0078] Step 12: If the action of shutting down the range hood is a valid shut-down operation (Ny that does not exceed the target number of actions N is considered a valid shut-down operation), then one normal shut-down is recorded, and y increases by 1;

[0079] Step 13: If both Nx and Ny reach the target number of operations, the program ends, otherwise repeat steps 7 to 12;

[0080] Step 14: After the test is completed, the program calculates and outputs the test results. The host computer clicks the "Generate Report" button to generate a test report. The test-related information entered into the system at the beginning of the test, such as report number, customer information, machine model, rated voltage, rated power, frequency, illuminance, ambient atmospheric pressure, temperature and humidity, etc., is automatically imported into the report, which not only improves the test efficiency, but also facilitates the traceability of relevant information.

[0081] The present invention provides a solution for automatically detecting and evaluating the gesture recognition control function of intelligent range hoods for third-party laboratories or internal laboratories of enterprises by automatically counting the recognition rate, one of the most important indicators of the gesture recognition control function of intelligent range hoods. This solution can reduce the detection workload of testers, improve detection efficiency, and has strong practicality.

[0082] The present invention adopts modular design, so that each component of the device can be quickly replaced or upgraded according to the actual use of products and functional needs, thereby improving the flexibility and future adaptability of the tooling.

Claims

1. A detection and evaluation method for the gesture recognition control function of an intelligent range hood, characterized in that: The host computer controls the bionic hand motion control module to simulate the gestures of a human hand in front of the range hood, and the pressure of the air inlet of the range hood is collected through the instrument data acquisition module. Then, the host computer automatically counts the recognition rate of the gesture recognition control function of the range hood according to the change of the pressure after each gesture, and forms a detection report according to the template.

2. The detection and evaluation method according to claim 1, characterized in that: The host computer is provided with a user parameter configuration interface, which supports user-configured parameters including the bionic hand waving direction such as up and down or left and right, the number of waving cycles, the waving time interval, the waving speed, and the distance between the bionic hand and the air inlet of the range hood. After the user completes the parameter setting and saves it, the host computer reconfigures the control logic of the bionic hand motion control module according to the above user settings.

3. The detection and evaluation method according to claim 2, characterized in that: The distance parameter between the initial position of the bionic hand and the air inlet of the range hood is configured, and then the distance between the bionic hand and the air inlet of the range hood after the movement is calculated based on the distance parameter, thereby realizing program control of the distance.

4. The detection and evaluation method according to claim 1, characterized in that: The specific steps of automatically counting the recognition rate of the range hood gesture recognition control function according to the change of the pressure after each gesture action by the host computer are as follows: 1) Set initial conditions, including: Initial air inlet pressure P0; initial normal opening times x=0; initial normal closing times y=0; total initial opening times Nx=0; total initial closing times Ny=0; number of test cycles N; 2) Test Control 2.1) Start operation a) controlling the bionic hand to swing out according to the user's setting of the swing direction of the bionic hand; b) After waiting for a certain period of time, the current pressure value P1 is read. If it exceeds the tolerance range of P0, it is determined that the opening control operation is successful, the normal opening times x=x+1, and the total opening operation times Nx=Nx+1; if the range hood is not successfully opened, the bionic hand is controlled to reset to the initial position of the opening action, and the above opening operation is repeated; 2.2) Close operation a) Control the bionic hand to wave back; b) After waiting for a certain period of time, the current pressure value P2 is read. If it exceeds the tolerance range of P1, it is determined that the closing control operation is successful, the normal closing number y=y+1, and the total number of closing operations Ny=Ny+1; if the range hood is not successfully closed, the bionic hand is controlled to reset to the initial position of the closing action, and the above closing operation is repeated; 2.3) If both Nx and Ny reach the number of cycles N, the program ends, otherwise repeat steps 2.1) and 2.2); 3) Recognition rate statistics Successful recognition rate of range hood opening operation gesture recognition control function Successful recognition rate of gesture recognition control function for range hood closing operation The total recognition rate of the range hood gesture recognition control function 5. The detection and evaluation method according to claim 1, characterized in that: The bionic hand motion control module comprises a four-degree-of-freedom mechanical intelligent arm and a bionic hand mounted thereon. The bionic hand is driven by the four-degree-of-freedom mechanical intelligent arm to simulate human hand movements in front of the range hood.

6. The detection and evaluation method according to claim 1, characterized in that: The data acquisition module comprises a universal shaped tube and a pressure gauge mounted thereon, wherein the pressure gauge is supported by the universal shaped tube and arranged near the air inlet of the range hood.

Images