Elevator calling system and method based on gesture recognition

Abstract

The invention relates to an elevator calling system and method based on gesture recognition. The landing calling system based on gesture recognition can comprise an image acquisition module, a gesture recognition module and a landing calling module, wherein the image acquisition module is configured to acquire a to-be-recognized gesture image; the image preprocessing module is electrically connected with the image acquisition module and is configured to preprocess the to-be-recognized gesture image acquired by the image acquisition module; the image recognition module is electrically connected with the image preprocessing module and is configured to recognize digits represented by gestures in the preprocessed to-be-recognized gesture image so as to output signals corresponding to the digits; and a controller configured to communicate with the image recognition module to receive a signal corresponding to the number output by the image recognition module, and to make an elevator call according to the signal.

Description

Technical Field

[0001] The present invention relates to an elevator calling system and method based on gesture recognition, which can determine the floor to which a user will go by recognizing the user's gesture, thereby making an elevator call. Background Art

[0002] Elevators are an indispensable tool for people to commute, travel, and go home every day. In places with dense flow of people, elevator buttons are used frequently. Viruses can survive on the surface of metal or plastic products for hours or even days depending on the type, ambient temperature, humidity and other factors. This means that the virus may survive on elevator buttons for a long time. When an infected person with the virus takes the elevator, the droplets carrying the virus sprayed by coughing or talking will adhere to the surface of the elevator buttons. Other people riding the elevator may be infected by touching the elevator buttons.

[0003] Gesture recognition technology refers to the technology that realizes human-computer interaction by recognizing various hand postures, movements and finger movements. Gesture recognition technology can be based on image recognition technology or sensor-based motion recognition technology.

[0004] Gesture recognition technology based on image recognition technology uses computer vision algorithms to analyze images, extract gesture features, and recognize various gesture actions. The advantage is that you can control or interact with the device by using simple gestures without touching the device.

[0005] Static gesture recognition, also known as static 2D gesture recognition, can recognize the meaning of gestures based on gesture images without using motion sensors.

[0006] The accuracy of static gesture recognition depends mainly on factors such as the diversity and complexity of gestures, lighting conditions, and the technology and algorithms used. Generally speaking, for simple gestures such as numbers and letters, the accuracy of static gesture recognition can reach more than 90%, or even as high as 98% to 99%. Static gesture recognition technology has achieved good results in practical applications.

[0007] The information disclosed in this Background of the Invention section is only intended to enhance understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that the information constitutes the prior art already known to a person skilled in the art. Summary of the invention

[0008] Various aspects of the present invention are directed to solving the above-mentioned problems in the prior art, and through the following detailed description of the specification, those skilled in the art will clearly understand other technical problems not mentioned herein.

[0009] In order to solve the above problems, on the one hand, the present invention provides an elevator calling system based on gesture recognition, which may include: an image acquisition module, an image preprocessing module, an image recognition module and a controller, wherein the image acquisition module may be configured to acquire a gesture image to be recognized; the image preprocessing module may be electrically connected to the image acquisition module and configured to preprocess the gesture image to be recognized acquired by the image acquisition module; the image recognition module may be electrically connected to the image preprocessing module and configured to recognize the number represented by the gesture in the preprocessed gesture image to be recognized, so as to output a signal corresponding to the number; the controller may be configured to communicate with the image recognition module, so as to receive the signal corresponding to the number output by the image recognition module, and make an elevator call according to the signal.

[0010] In an exemplary embodiment of the present invention, the image recognition module can be configured to: extract a hand area based on a gesture segmentation method in a preprocessed gesture image to be recognized; extract one or more gesture features from the extracted hand area; match the extracted one or more gesture features with the gesture features of a pre-stored gesture template to determine the number represented by the gesture.

[0011] In an exemplary embodiment of the present invention, the image recognition module can be configured as follows: after extracting the hand area, when only one hand area is extracted, the number represented by the gesture is determined to be a single-digit number; when two hand areas are extracted, the number represented by the gesture is determined to be a two-digit number.

[0012] In an exemplary embodiment of the present invention, the image recognition module can be configured as follows: when it is determined that the number represented by the gesture is a two-digit number, the number represented by the right hand gesture in the gesture image is determined as the high digit of the two-digit number, and the number represented by the left hand gesture in the gesture image is determined as the low digit of the two-digit number.

[0013] The elevator calling system based on gesture recognition according to an exemplary embodiment of the present invention may further include: a display module configured to display the number recognized by the image recognition module for the user to confirm whether the recognized number is correct.

[0014] In an exemplary embodiment of the present invention, when the user confirms that the number recognized by the image recognition module is correct, the image recognition module sends a signal corresponding to the recognized number to the controller. When the user confirms that the number recognized by the image recognition module is incorrect, the display module can serve as an input device, allowing the user to input the correct number through the display module, and sending a signal corresponding to the correct number input by the user to the controller.

[0015] In an exemplary embodiment of the present invention, the image preprocessing module may be configured to: perform denoising and information enhancement on the gesture image to be recognized acquired by the image acquisition module.

[0016] In an exemplary embodiment of the present invention, the image acquisition module may be configured to: detect a human body, and when a human body is detected, start acquiring gesture images to be recognized.

[0017] The elevator calling system based on gesture recognition according to an exemplary embodiment of the present invention may further include: a voice announcement module electrically connected to the image recognition module and configured to announce the number recognized by the image recognition module.

[0018] On the other hand, the present invention provides an elevator calling method based on gesture recognition, which may include: an image acquisition module, collecting gesture images to be recognized; an image preprocessing module, preprocessing the gesture images to be recognized collected by the image acquisition module; an image recognition module, recognizing the number represented by the gesture in the preprocessed gesture image to be recognized, to output a signal corresponding to the number; a controller, receiving the signal corresponding to the number output by the image recognition module, and calling the elevator according to the signal.

[0019] In an exemplary embodiment of the present invention, an image recognition module extracts a hand area from a preprocessed gesture image to be recognized based on a gesture segmentation method; extracts one or more gesture features from the extracted hand area; and matches the extracted one or more gesture features with gesture features of a pre-stored gesture template to determine the number represented by the gesture.

[0020] In an exemplary embodiment of the present invention, after extracting the hand region, when only one hand region is extracted, the image recognition module determines that the number represented by the gesture is a single-digit number; when two hand regions are extracted, the image recognition module determines that the number represented by the gesture is a two-digit number.

[0021] In an exemplary embodiment of the present invention, when it is determined that the number represented by the gesture is a two-digit number, the image recognition module determines the number represented by the right hand gesture in the gesture image as the high digit of the two-digit number, and determines the number represented by the left hand gesture in the gesture image as the low digit of the two-digit number.

[0022] The elevator calling method based on gesture recognition according to an exemplary embodiment of the present invention may further include: displaying, by a display module, the number recognized by the image recognition module, so that the user can confirm whether the recognized number is correct.

[0023] In an exemplary embodiment of the present invention, when the user confirms that the number recognized by the image recognition module is correct, the image recognition module sends a signal corresponding to the recognized number to the controller. When the user confirms that the number recognized by the image recognition module is incorrect, the user inputs the correct number through the display module as an input device, and a signal corresponding to the correct number input by the user is sent to the controller.

[0024] In an exemplary embodiment of the present invention, the image preprocessing module performs denoising and information enhancement on the gesture image to be recognized acquired by the image acquisition module.

[0025] In an exemplary embodiment of the present invention, the image acquisition module detects a human body, and when a human body is detected, it starts to acquire gesture images to be recognized.

[0026] The elevator calling method based on gesture recognition according to an exemplary embodiment of the present invention may further include: a voice announcement module announcing the number recognized by the image recognition module. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] The above and other objects, features and other advantages of the present invention will be more clearly understood through the following detailed description presented in conjunction with the accompanying drawings, in which:

[0028] Figure 1 A control block diagram of an elevator call system based on gesture recognition according to an exemplary embodiment of the present invention is shown;

[0029] Figure 2 A schematic diagram showing gestures corresponding to numbers 0 to 9 according to an exemplary embodiment of the present invention;

[0030] Figure 3A and Figure 3B A schematic diagram showing an example of a gesture indicating a two-digit floor according to an exemplary embodiment of the present invention;

[0031] Figure 4 The flowchart is a diagram showing an elevator calling method based on gesture recognition according to an exemplary embodiment of the present invention.

[0032] It should be understood that the drawings are not drawn to scale, but show various features that illustrate the basic principles of the invention in a slightly simplified manner. In the drawings of the present invention, the same reference numerals represent the same or equivalent parts of the present invention. DETAILED DESCRIPTION

[0033] Reference will now be made in detail to various embodiments of the present invention, examples of which are shown in the accompanying drawings and described below. Although the present invention will be described in conjunction with exemplary embodiments of the present invention, it should be understood that this specification is not intended to limit the present invention to the exemplary embodiments of the present invention described. On the other hand, the present invention is intended to cover not only the exemplary embodiments of the present invention, but also various alternative embodiments, modified embodiments, equivalent embodiments and other embodiments that may be included in the spirit and scope of the present invention defined by the appended claims.

[0034] Hereinafter, an elevator calling system and method based on gesture recognition according to an exemplary embodiment of the present invention will be described in detail with reference to the accompanying drawings.

[0035] Figure 1 FIG. 4 is a control block diagram of an elevator calling system based on gesture recognition according to an exemplary embodiment of the present invention.

[0036] like Figure 1 As shown, the elevator calling system based on gesture recognition may include: an image acquisition module 10 , an image preprocessing module 20 , an image recognition module 30 , a display module 40 , a voice broadcasting module 50 and a controller 60 .

[0037] The image acquisition module 10 may be a camera and is disposed near the elevator call operation panel in the elevator car for acquiring images. The user of the elevator may stand in front of the image acquisition module 10. When it is determined that someone is in front of the image acquisition module 10, the image acquisition module 10 may start to acquire the user's gesture image to be recognized and send the acquired user's gesture image to be recognized to the image preprocessing module 20. Conventional human body recognition technology may be used to determine whether someone is in front of the image acquisition module 10.

[0038] The image preprocessing module 20 may be electrically connected to the image acquisition module 10, and preprocess the gesture image to be recognized acquired by the image acquisition module 10, and send the preprocessed gesture image to be recognized to the image recognition module 30. The preprocessing may include denoising and information enhancement processing of the gesture image.

[0039] The image recognition module 30 can be electrically connected to the image preprocessing module 20, and use a gesture recognition algorithm to recognize the number represented by the gesture in the preprocessed gesture image to be recognized, so as to output a signal corresponding to the number. The image recognition module 30 can send the signal corresponding to the number to the controller 60 to determine the floor selected by the user.

[0040] Currently, the gesture recognition algorithms that are widely used include: gesture recognition algorithms based on template matching, gesture recognition algorithms based on machine learning, and gesture recognition algorithms based on deep learning.

[0041] The gesture recognition algorithm based on template matching collects a series of gesture images in advance and stores the gesture images. The gesture images to be recognized are matched with the pre-stored gesture templates to achieve gesture recognition. The template matching method is relatively simple and easy to implement, and is suitable for recognizing simple gestures with a small number, such as gestures representing numbers and letters.

[0042] Gesture recognition algorithms based on machine learning achieve gesture recognition by learning gesture features and classification models in the training set. Common machine learning methods include support vector machines, random forests, neural networks, etc. Compared with template matching methods, gesture recognition methods based on machine learning can handle more complex gestures, but they require a large number of gesture training samples and strong feature extraction and classification capabilities.

[0043] The gesture recognition algorithm based on deep learning is a type of machine learning. It uses a multi-layer neural network model and can process high-dimensional data and complex features. Commonly used deep learning models include convolutional neural network (CNN), recurrent neural network (RNN), etc.

[0044] In an exemplary embodiment of the present invention, the image recognition module 30 may use a gesture recognition algorithm based on template matching to recognize the pre-processed gesture image to be recognized. The process of gesture recognition generally includes: gesture segmentation, gesture feature extraction and gesture recognition.

[0045] First, the image recognition module 30 can perform gesture segmentation on the preprocessed gesture image to be recognized. Gesture segmentation refers to extracting the hand area in the gesture image. Specifically, the hand area is located in the gesture image including the hand area and the background area, and then the located hand area is separated from the background area to extract the hand area. Static gesture recognition only needs to extract the hand area in a single frame image. Image segmentation methods include region-based image segmentation methods and edge-based image segmentation methods. The region-based image segmentation method is an algorithm that divides an image into different parts with similar pixel features. The pixels in the hand area and the background area have different pixel features, and the gesture image can be segmented into the hand area and the background area based on these two different pixel features. The edge-based image segmentation method performs edge detection based on the discontinuity of the pixel grayscale at the edge position, thereby segmenting the image based on the edge. The hand area has a relatively obvious edge, and the gesture image can be segmented into the hand area and the background area based on the edge of the hand. The present invention can adopt any of the above two image segmentation methods.

[0046] Next, the image recognition module 30 can further perform gesture feature extraction in the extracted hand area for gesture recognition. Gesture features refer to features that best reflect the meaning of a gesture. Gesture features may include pixel value features, edge features, shape features, etc. Specifically, the image recognition module 30 can extract one or more gesture features in the hand area. The more features extracted, the more accurate the result of gesture recognition. However, extracting too many features will affect the speed of gesture recognition. The image recognition module 30 can perform gesture recognition based on the extracted one or more gesture features.

[0047] Next, the image recognition module 30 may compare the gesture features of the gesture image to be recognized with the gesture features of the pre-stored gesture template. The pre-stored gesture template includes each gesture corresponding to the numbers 0 to 9. Specifically, the image recognition module 30 may calculate the similarity between the gesture image to be recognized and the pre-stored gesture image based on the comparison result of the gesture features of the gesture image to be recognized with the gesture features of the pre-stored gesture template. The image recognition module 30 may further determine the gesture template with the highest similarity among the pre-stored gesture templates whose similarity with the gesture image to be recognized is higher than a threshold value (e.g., 90%). The image recognition module 30 may recognize the number represented by the gesture in the gesture image to be recognized as the number represented by the gesture template with the highest similarity. The image recognition module 30 may send a signal corresponding to the number represented by the gesture in the recognized gesture image to the controller 60, so that the controller 60 may call the elevator according to the received signal.

[0048] Figure 2 Schematic diagram showing gestures corresponding to numbers 0 to 9 according to an exemplary embodiment of the present invention. Figure 2 As can be understood, for each number, multiple gesture templates corresponding to the number can be pre-stored in the image recognition module 30. Hand images taken from different angles can be pre-stored as gesture templates to improve the accuracy of gesture recognition. Figure 2 The gesture shown may represent the number 3, and there is another gesture that may also represent the number 3, and both gestures may be stored in the image recognition module 30.

[0049] The display module 40 may be a touch display screen and may be used to replace a traditional elevator call operation panel with physical buttons. When the image acquisition module 10 acquires a gesture image to be recognized by the user, the display module 40 may display the shooting range of the image acquisition module 10 in real time to ensure that the user makes a gesture within the shooting range of the image acquisition module 10. When the image recognition module 30 recognizes the number represented by the gesture in the gesture image, the display module 40 may display the number so that the user can confirm whether the number recognized by the image recognition module 30 is correct.

[0050] In the process of identifying the gesture image, the user needs to continue to maintain the gesture. If the recognized number is correct, the user needs to continue to maintain the gesture for more than a predetermined time. If the recognized number is incorrect, the user can stop the hand gesture. Therefore, after the recognized number is displayed by the display module 40, it can be determined whether the user continues to maintain the gesture for more than a predetermined time. When it is determined that the user continues to maintain the gesture for more than a predetermined time, an elevator call can be made according to the recognized number. Conversely, when it is determined that the user does not continue to maintain the gesture for more than a predetermined time, the elevator call may not be made. Preferably, the predetermined time can be 2s.

[0051] The voice announcement module 50 may be a speaker disposed in the elevator car. When the image recognition module 30 recognizes the number represented by the gesture in the gesture image, the voice announcement module 50 may announce the number for the user to confirm whether the number recognized by the image recognition module 30 is correct.

[0052] When the user confirms that the number recognized by the image recognition module 30 is correct, the image recognition module 30 may send a signal corresponding to the recognized number to the controller 60. When the user confirms that the number recognized by the image recognition module 30 is incorrect, the user may make a gesture again, or the display module 40 may serve as an input device so that the user may directly input the correct floor through the display module 40.

[0053] The controller 60 may be a server of an elevator system for controlling the operation of the elevator, and may wirelessly communicate with the image recognition module 30. The controller 60 may control the operation of the elevator according to the number received and recognized by the image recognition module 30, and transport the user to the floor corresponding to the recognized number.

[0054] After extracting the hand area, when only one hand area is extracted, the image recognition module 30 can determine that the number represented by the gesture in the gesture image to be recognized is a single-digit number (i.e., one of the numbers 1 to 9); when two hand areas are extracted, the image recognition module 30 can determine that the number represented by the gesture in the gesture image to be recognized is a two-digit number (i.e., the number 10 and numbers above 10).

[0055] Since the direction of the image captured by the camera and the direction of the user have a mirror relationship, the left side of the image captured by the camera is actually the right side of the user, and the right side of the image captured by the camera is actually the left side of the user. When the user makes two gestures with both hands at the same time, it is assumed that the number represented by the user's left hand represents the high digit (i.e., the tens digit) in the two-digit number, and the number represented by the user's right hand represents the low digit (i.e., the ones digit) in the two-digit number. For example, when the user makes a gesture representing the number 1 with his left hand and makes a gesture representing the number 2 with his right hand, it can be considered that the floor the user wants to go to is the 12th floor. Accordingly, when performing gesture recognition, the image recognition module 30 can determine the number represented by the right-hand gesture (corresponding to the user's left hand) in the collected gesture image to be recognized as the high digit (i.e., the tens digit) in the two-digit number, and determine the number represented by the left-hand gesture (corresponding to the user's right hand) in the collected gesture image to be recognized as the low digit (i.e., the ones digit) in the two-digit number.

[0056] Figure 3A and Figure 3B FIG. 4 is a schematic diagram showing an example of a gesture indicating a two-digit floor according to an exemplary embodiment of the present invention. Figure 3A and Figure 3B Two gesture images captured by the image capture module 10 are respectively shown, which have a mirror relationship with the direction of the user.

[0057] like Figure 3A As shown, in the collected gesture image, the number represented by the right gesture (corresponding to the user's left hand) is 1, and in the collected gesture image, the number represented by the left gesture (corresponding to the user's right hand) is 2. At this time, the image recognition module 30 can recognize that the gesture made by the user is 12.

[0058] like Figure 3B As shown, in the collected gesture image, the number represented by the right hand gesture (corresponding to the user's left hand) is 0, and the number represented by the left hand gesture (corresponding to the user's right hand) is 1. At this time, the image recognition module 30 can recognize that the gesture made by the user is 01, that is, 1. In other words, when the high digit (i.e., the tens digit) of the two-digit number is recognized as 0, the floor selected by the user is a single-digit floor.

[0059] In another exemplary embodiment, the image recognition module 30 of the elevator call system based on gesture recognition of the present invention may have a face recognition function. The user of the elevator may be registered and the registration information may be stored. After completing the registration, when the user enters the elevator car, the elevator call system based on gesture recognition may perform face recognition on the user through the image recognition module 30, and authenticate the user through face recognition to determine whether the user is a registered user. The registration information may include the floor information where the user lives. If it is determined that the user is a registered user, the controller 60 of the elevator call system based on gesture recognition may call the elevator according to the registration information to send the user to the floor where he lives. In the above process, the user does not need to touch the elevator button, thereby reducing the health risks that may be caused by calling the elevator. The above process can be achieved by adding a face recognition algorithm to the image recognition module 30 to realize the face recognition function and storing the user's registration information in the controller 60 without adding additional hardware equipment.

[0060] Figure 4 FIG. 1 is a flowchart showing a method for calling an elevator based on gesture recognition according to an exemplary embodiment of the present invention. Figure 1 The elevator calling system based on gesture recognition is shown to execute the elevator calling method based on gesture recognition.

[0061] like Figure 4 As shown, in step S101, the image acquisition module may determine whether there is a person in front of the image acquisition module. When it is determined that there is a person in front of the image acquisition module, the image acquisition module may start to acquire the user's gesture image to be recognized.

[0062] In step S102, the image preprocessing module may preprocess the gesture image of the user to be recognized collected by the image collection module. The image preprocessing module may perform denoising and information enhancement processing on the gesture image of the user.

[0063] In step S103, the image recognition module may perform gesture segmentation on the preprocessed gesture image to be recognized. The image recognition module may separate the hand region from the background region using a region-based image segmentation method and / or an edge-based image segmentation method to extract the hand region. When only one hand region is extracted, the image recognition module may determine that the number represented by the gesture is a single-digit number. When two hand regions are extracted, the image recognition module may determine that the number represented by the gesture is a two-digit number. When it is determined that the number represented by the gesture is a two-digit number, the image recognition module may determine the number represented by the right-hand gesture in the gesture image to be recognized as the high digit of the two-digit number, and determine the number represented by the left-hand gesture in the gesture image to be recognized as the low digit of the two-digit number. In step S104, the image recognition module may extract one or more gesture features in the extracted hand region for gesture recognition.

[0064] In step S105, the image recognition module may compare the gesture features of the gesture image to be recognized extracted with the gesture features of the pre-stored gesture templates to calculate the similarity between the gesture image to be recognized and the pre-stored gesture images. The image recognition module may determine the gesture template with the highest similarity among the gesture templates, and recognize the number represented by the gesture in the gesture image to be recognized as the number represented by the gesture template with the highest similarity.

[0065] In step S106, the image recognition module may display the result of the gesture recognition, that is, the number represented by the gesture in the recognized gesture image, through the display module and / or announce it through the voice announcement module.

[0066] In step S107, the user can confirm whether the recognized result is correct by viewing the numbers displayed on the display module and / or the numbers broadcast by the voice broadcast module.

[0067] When the user confirms that the recognition result is correct (step S107 is yes), in step S108, the image recognition module can send a signal corresponding to the recognized number to the controller, so that the controller can determine the floor selected by the user based on the received signal and call the elevator according to the floor selected by the user.

[0068] When the user confirms that the recognized result is incorrect (step S107 is no), the user can make a gesture again in step S109. In this case, after the user makes a gesture again, it is possible to return to step S101 and re-execute steps S101 to S107.

[0069] Alternatively, when the user confirms that the recognized result is incorrect (step S107 is no), in step S110, the user can directly input the correct floor through the display module as an input device. In this case, the correct floor input by the user can be sent to the controller. When the controller receives the correct floor input by the user, step S108 is executed, and the controller can call the elevator according to the floor input by the user.

[0070] The elevator calling system and method based on gesture recognition according to the exemplary embodiment of the present invention can enable the user to make a contactless elevator call by making gestures representing numbers without touching the elevator buttons, thereby reducing the health risks that may be caused by elevator calling.

[0071] The above descriptions of the specific exemplary embodiments of the present invention are presented for the purpose of illustration and description. They are not intended to be exhaustive or to limit the present invention to the precise form disclosed, and it is obvious that many modifications and changes can be made according to the above teachings. The selection and description of the exemplary embodiments are to explain certain principles of the present invention and their practical applications, so that those skilled in the art can make and use various exemplary embodiments of the present invention and their different alternative forms and modified forms. The scope of the present invention is intended to be limited by the appended claims and their equivalents.

Claims

1. An elevator calling system based on gesture recognition, comprising: An image acquisition module configured to acquire a gesture image to be recognized; An image preprocessing module, which is electrically connected to the image acquisition module and is configured to preprocess the gesture image to be recognized acquired by the image acquisition module; an image recognition module, which is electrically connected to the image preprocessing module and is configured to recognize the number represented by the gesture in the preprocessed gesture image to be recognized, so as to output a signal corresponding to the number; as well as The controller is configured to communicate with the image recognition module to receive a signal corresponding to the number output by the image recognition module and to make an elevator call according to the signal.

2. The elevator calling system based on gesture recognition according to claim 1, wherein: The image recognition module is configured to: extract a hand area based on a gesture segmentation method in a preprocessed gesture image to be recognized; extract one or more gesture features from the extracted hand area; and match the extracted one or more gesture features with gesture features of a pre-stored gesture template to determine the number represented by the gesture.

3. The elevator calling system based on gesture recognition according to claim 2, wherein: The image recognition module is configured to: after extracting the hand area, when only one hand area is extracted, determine that the number represented by the gesture is a single-digit number; when two hand areas are extracted, determine that the number represented by the gesture is a two-digit number.

4. The elevator calling system based on gesture recognition according to claim 3, wherein: The image recognition module is configured to: when it is determined that the number represented by the gesture is a two-digit number, determine the number represented by the right hand gesture in the gesture image as the high digit of the two-digit number, and determine the number represented by the left hand gesture in the gesture image as the low digit of the two-digit number.

5. The elevator calling system based on gesture recognition according to claim 1, further comprising: The display module is configured to display the number recognized by the image recognition module so that the user can confirm whether the recognized number is correct.

6. The elevator calling system based on gesture recognition according to claim 5, wherein: When the user confirms that the number recognized by the image recognition module is correct, the image recognition module sends a signal corresponding to the recognized number to the controller; When the user confirms that the number recognized by the image recognition module is incorrect, the display module serves as an input device, allowing the user to input the correct number through the display module, and sending a signal corresponding to the correct number input by the user to the controller.

7. The elevator calling system based on gesture recognition according to claim 1, wherein: The image preprocessing module is configured to perform denoising and information enhancement on the gesture image to be recognized acquired by the image acquisition module.

8. The elevator calling system based on gesture recognition according to claim 1, wherein: The image acquisition module is configured to detect a human body, and when a human body is detected, start acquiring gesture images to be recognized.

9. The elevator calling system based on gesture recognition according to claim 1, further comprising: The voice broadcast module is electrically connected to the image recognition module and is configured to broadcast the numbers recognized by the image recognition module.

10. A method for calling an elevator based on gesture recognition, comprising: The image acquisition module acquires the gesture image to be recognized; The image preprocessing module preprocesses the gesture image to be recognized acquired by the image acquisition module; The image recognition module recognizes the number represented by the gesture in the preprocessed gesture image to be recognized, so as to output a signal corresponding to the number; The controller receives a signal corresponding to the number output by the image recognition module, and makes an elevator call according to the signal.

11. The elevator calling method according to claim 10, wherein: The image recognition module extracts the hand area from the preprocessed gesture image to be recognized based on the gesture segmentation method; extracts one or more gesture features from the extracted hand area; and matches the extracted one or more gesture features with the gesture features of the pre-stored gesture template to determine the number represented by the gesture.

12. The elevator calling method according to claim 11, wherein: After extracting the hand area, when only one hand area is extracted, the image recognition module determines that the number represented by the gesture is a single-digit number. When two hand areas are extracted, the image recognition module determines that the number represented by the gesture is a two-digit number.

13. The elevator calling method according to claim 12, wherein: When it is determined that the number represented by the gesture is a two-digit number, the image recognition module determines the number represented by the right hand gesture in the gesture image as the high digit of the two-digit number, and determines the number represented by the left hand gesture in the gesture image as the low digit of the two-digit number.

14. The elevator calling method according to claim 10, further comprising: The display module displays the number recognized by the image recognition module so that the user can confirm whether the recognized number is correct.

15. The elevator calling method according to claim 14, wherein: When the user confirms that the number recognized by the image recognition module is correct, the image recognition module sends a signal corresponding to the recognized number to the controller; When the user confirms that the number recognized by the image recognition module is incorrect, the user inputs a correct number through the display module as an input device, and a signal corresponding to the correct number input by the user is sent to the controller.

16. The elevator calling method according to claim 10, wherein: The image preprocessing module performs denoising and information enhancement on the gesture image to be recognized acquired by the image acquisition module.

17. The elevator calling method according to claim 10, wherein: The image acquisition module detects the human body, and when the human body is detected, it starts to acquire the gesture image to be recognized.

18. The elevator calling method according to claim 10, further comprising: The voice broadcast module broadcasts the numbers recognized by the image recognition module.

Images