An intelligent projector based on biometric recognition function

Abstract

This invention belongs to the field of control and discloses a smart projector based on biometric recognition, including a camera module, a face recognition module, and a control module. The camera module acquires a user's facial image and transmits it to the face recognition module. The face recognition module identifies the facial image acquired by the camera module and determines whether the similarity between the acquired facial image and pre-stored facial images in the face recognition module exceeds a set similarity threshold, obtaining a judgment result. The control module controls the power-on process of the smart projector based on the judgment result. Unlike existing methods that rely on power-on passwords for protection, this invention controls the power-on process of the smart projector through face recognition, effectively improving the projector's data security.

Description

Technical Field

[0001] This invention relates to the field of control, and more particularly to an intelligent projector based on biometric recognition. Background Technology

[0002] A smart projector is a projector with a built-in operating system. Users can operate the smart projector via a remote control. Currently, most smart projectors protect their data by setting a power-on password. However, these passwords can be compromised, making this method insufficiently secure. Summary of the Invention

[0003] The purpose of this invention is to disclose a smart projector based on biometric recognition function, and to solve the problem of how to improve the data security of smart projectors.

[0004] To achieve the above objectives, the present invention provides the following technical solution:

[0005] This invention provides an intelligent projector based on biometric recognition, including a shooting module, a face recognition module, and a control module;

[0006] The camera module is used to capture the user's facial image and transmit the facial image to the facial recognition module;

[0007] The face recognition module is used to recognize the face images acquired by the shooting module, determine whether the similarity between the face image acquired by the shooting module and the face images pre-stored in the face recognition module is greater than a set similarity threshold, and obtain the judgment result.

[0008] The control module is used to control the power-on process of the smart projector based on the judgment result.

[0009] Preferably, the shooting module includes a first camera unit that is wiredly connected to the data input port of the smart projector. The first camera unit is used to capture the user's face and obtain the user's facial image.

[0010] Preferably, the camera module includes a wireless communication device disposed inside the projector and a second camera unit capable of wirelessly communicating with the wireless communication device. The second camera unit is used to capture the user's face and obtain an image of the user's face.

[0011] Preferably, the wireless communication device is either Bluetooth or WiFi.

[0012] Preferably, the second camera unit includes any one of a camera on a smartphone, a camera on a tablet, or a camera on a laptop.

[0013] Preferably, the face recognition module includes a receiving unit, a storage unit, a similarity calculation unit, and a judgment unit;

[0014] The receiving unit is used to receive facial images transmitted from the shooting module;

[0015] The storage unit is used to save pre-stored face images and preset similarity thresholds;

[0016] The similarity calculation unit is used to calculate the similarity between the face image acquired by the shooting module and the face image pre-stored in the face recognition module;

[0017] The judgment unit is used to determine whether the similarity is greater than the set similarity threshold and obtain the judgment result.

[0018] Preferably, the judgment result is that the similarity is greater than the set similarity threshold or the similarity is less than or equal to the set similarity threshold.

[0019] Preferably, the power-on process of the smart projector is controlled based on the judgment result, including:

[0020] If the judgment result is that the similarity is greater than the set similarity threshold, then control the smart projector to project the startup desktop;

[0021] If the similarity result is less than or equal to the set similarity threshold, the smart projector will project a preset prompt text to inform the user that the face recognition failed.

[0022] Preferably, it also includes a remote control module;

[0023] The remote control module is used to send remote control signals to the smart projector.

[0024] Preferably, the remote control signal is a Bluetooth signal or an infrared signal.

[0025] Unlike existing methods that rely on power-on passwords for protection, this invention uses facial recognition to control the power-on process of a smart projector, which can effectively improve the data security of the projector. Attached Figure Description

[0026] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0027] Figure 1 This is a schematic diagram of an intelligent projector based on biometric recognition function according to the present invention.

[0028] Figure 2 This is a schematic diagram of the face recognition module of the present invention. Detailed Implementation

[0029] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this application or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0030] like Figure 1 As shown in one embodiment, the present invention provides an intelligent projector based on biometric recognition function, including a shooting module, a face recognition module and a control module;

[0031] The camera module is used to capture the user's facial image and transmit the facial image to the facial recognition module;

[0032] The face recognition module is used to recognize the face images acquired by the shooting module, determine whether the similarity between the face image acquired by the shooting module and the face images pre-stored in the face recognition module is greater than a set similarity threshold, and obtain the judgment result.

[0033] The control module is used to control the power-on process of the smart projector based on the judgment result.

[0034] Unlike existing methods that rely on power-on passwords for protection, this invention uses facial recognition to control the power-on process of a smart projector, which can effectively improve the data security of the projector.

[0035] Traditional methods of setting a power-on password are not very practical in situations with many people around, because others may be able to see the entered password at close range, leading to the leakage of the power-on password and affecting the security of the data in the smart projector.

[0036] Preferably, the shooting module includes a first camera unit that is wiredly connected to the data input port of the smart projector. The first camera unit is used to capture the user's face and obtain the user's facial image.

[0037] Preferably, the user's face is photographed to obtain a facial image, including:

[0038] The camera continuously captures images of the area in front of the first camera unit, obtaining multiple real-time images.

[0039] Multiple real-time images are filtered to obtain a face image.

[0040] Specifically, the method of filtering multiple images can effectively increase the probability of obtaining high-quality facial images, because if only one image is captured, a blurry image may be obtained.

[0041] Preferably, multiple real-time images are filtered to obtain a face image, including:

[0042] S1, acquire the earliest captured real-time image erpho and the latest captured real-time image latpho from multiple real-time images;

[0043] S2, perform face region detection on erpho to obtain the face region blkfco in erpho;

[0044] S3, expand the face region in Erpho to obtain the expanded region;

[0045] S4, obtain the set of pixels in latpho that have the same coordinates as the pixels in the extended region, faccoef;

[0046] S5, perform face region detection on the pixels in the set faccoef to obtain the face region blkfct in latpho;

[0047] S6, obtain the minimum x-coordinate minxo, the maximum x-coordinate maxxo, the minimum y-coordinate minyo, and the maximum y-coordinate maxyo of the pixels in blkfco.

[0048] S7, obtain the minimum x-coordinate minxt, the maximum x-coordinate maxxt, the minimum y-coordinate minyt, and the maximum y-coordinate maxyt of the pixels in blkfct;

[0049] S8, obtain the detection area, the coordinate range of the detection area is:

[0050]

[0051] min means to get the smaller value in the parentheses, and max means to get the larger value in the parentheses;

[0052] S9, calculate the detection coefficient of the detection region in each real-time image;

[0053] S10 uses the real-time image with the highest monitoring coefficient as the face image.

[0054] Because the user's head may move during continuous shooting, the face region obtained from Erpho cannot be directly used as the face region in other real-time images.

[0055] In this invention, when filtering multiple inspection images, face detection is not performed on each real-time image separately. Instead, face regions in the two real-time images are obtained by calculating erpho and latpho. Then, the detection regions are calculated from these face regions. Finally, the detection coefficients for each real-time image are calculated using these detection regions, and the final selected face image is obtained using these detection coefficients. Compared to directly performing face detection on each real-time image, this invention effectively shortens the time required to obtain the detection regions, thereby reducing the time required to obtain face images and consequently shortening the time required for face recognition of the user, resulting in a better user experience.

[0056] Compared to calculating detection coefficients for the entire real-time image, the detection region includes both face regions and a small portion of non-face regions in the real-time image. This makes the obtained detection coefficients more representative of the face regions. Therefore, the real-time image with the best face region quality can be used as the final face image.

[0057] In addition, when obtaining the face region in latpho, this invention does not directly obtain the face region of the entire image, but expands the face region of erpho to obtain an expanded region, and then detects the face region in the expanded region, which effectively shortens the time required to obtain blkfct.

[0058] Preferably, face region detection can be performed using algorithms such as Cascade CNN or DenseBox.

[0059] Preferably, the face region in Erpho is expanded to obtain an expanded region, including:

[0060] S31, calculate the difference in shooting time between erpho and latpho, timdf:

[0061] timdf=shtm latpho -shtm erpho

[0062] Among them, shtm latpho and shtm erpho These represent the shooting times for the erpho and latpho images, respectively.

[0063] S32, the extended region is obtained using the following formula:

[0064]

[0065] Where timst represents the preset time length and chs represents the preset constant coefficient.

[0066] Specifically, the degree of expansion is mainly related to the value of timdf. The larger the timdf value, the greater the expansion degree and the larger the expansion area. This is because a larger timdf increases the probability of the user's head moving. Therefore, the size of the expansion area in this invention can adaptively change with the timdf value, effectively avoiding an expansion area that is too large or too small. If the expansion area is too small, the proportion of the face region in the LATPHO may be too low, making it impossible to obtain a complete face region. Conversely, if the expansion area is too large, it will result in an excessively long face region acquisition time.

[0067] Preferably, the detection coefficient is calculated as follows:

[0068]

[0069] Among them, chekidx g The detection coefficients of the detection region g are represented by w1 and w2, which are preset to represent the window weight and edge weight, respectively. w1 + w2 = 1. Ni is the number of window pixels contained in phost, and phost is the set of window pixels contained in the detection region g. i Let be the grayscale value of pixel i in the window, stgray represent the preset grayscale variance, phobl represent the set of edge pixels in the detection region g, and gray j nei represents the grayscale value of pixel j. j The top(nei) represents the set of pixels in the 8-neighborhood of pixel j. j ) indicates obtaining nei j Mi represents the maximum grayscale value of the pixels in phobl; stbls represents the preset grayscale constant.

[0070] The detection coefficient is calculated from two aspects: the variance of gray values ​​of window pixels and the relationship between gray values ​​of edge pixels and neighboring pixels. The larger the variance of gray values ​​of window pixels and the smaller the minimum difference of gray values ​​between edge pixels and neighboring pixels, the larger the detection coefficient of the detection area, thus selecting the real-time image with the best face region quality.

[0071] A larger grayscale variance indicates a higher content of detail information in the detection area. Conversely, a smaller minimum difference in grayscale values ​​between edge pixels and their neighboring pixels indicates a lower probability that the edge pixels are noise, resulting in better quality of the face region. Existing technologies generally calculate image quality by calculating all pixels, but this method is less effective in this invention. Firstly, it has low computational efficiency, and secondly, the calculated quality reflects the overall quality and cannot capture the quality of the actual area being used. Consequently, the real-time image used as a face image is not of the highest quality.

[0072] Preferably, the method for obtaining window pixels is as follows:

[0073] The detection area is divided into multiple sub-areas.

[0074] Obtain the center of each sub-region and use the center pixel of the sub-region as the window pixel.

[0075] Specifically, multiple sub-regions are of the same size. By acquiring the window pixels, the number of pixels involved in the detection coefficients can be reduced, thereby improving the calculation efficiency of the detection coefficients.

[0076] Preferably, the camera module includes a wireless communication device disposed inside the projector and a second camera unit capable of wirelessly communicating with the wireless communication device. The second camera unit is used to capture the user's face and obtain an image of the user's face.

[0077] Specifically, by setting up a second camera unit capable of wireless communication, the range of facial recognition for the smart projector is effectively expanded, allowing users to perform facial recognition even without being near the smart projector.

[0078] Preferably, the wireless communication device is either Bluetooth or WiFi.

[0079] Preferably, the second camera unit includes any one of a camera on a smartphone, a camera on a tablet, or a camera on a laptop.

[0080] Specifically, by utilizing the cameras on existing computing devices, facial recognition can be fully leveraged, further reducing the cost of smart projectors.

[0081] Preferably, such as Figure 2 As shown, the face recognition module includes a receiving unit, a storage unit, a similarity calculation unit, and a judgment unit;

[0082] The receiving unit is used to receive facial images transmitted from the shooting module;

[0083] The storage unit is used to save pre-stored face images and preset similarity thresholds;

[0084] The similarity calculation unit is used to calculate the similarity between the face image acquired by the shooting module and the face image pre-stored in the face recognition module;

[0085] The judgment unit is used to determine whether the similarity is greater than the set similarity threshold and obtain the judgment result.

[0086] Specifically, similarity can be calculated based on the similarity of image features.

[0087] Preferably, the judgment result is that the similarity is greater than the set similarity threshold or the similarity is less than or equal to the set similarity threshold.

[0088] Preferably, the power-on process of the smart projector is controlled based on the judgment result, including:

[0089] If the judgment result is that the similarity is greater than the set similarity threshold, then control the smart projector to project the startup desktop;

[0090] If the similarity result is less than or equal to the set similarity threshold, the smart projector will project a preset prompt text to inform the user that the face recognition failed.

[0091] Specifically, in this invention, the smart projector powers on upon receiving a power-on command and then projects text on the screen prompting the user to perform facial recognition. If facial recognition is successful, the user is taken to the desktop; otherwise, the user is not taken to the desktop, thus ensuring the security of the data in the smart projector.

[0092] Preferably, it also includes a remote control module;

[0093] The remote control module is used to send remote control signals to the smart projector.

[0094] Preferably, the remote control signal is a Bluetooth signal or an infrared signal.

[0095] Specifically, transmitting remote control signals via Bluetooth eliminates the need to point the signal directly at the smart projector, making remote control more convenient.

[0096] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent projector based on biometric recognition function, characterized in that, It includes a camera module, a face recognition module, and a control module; The camera module is used to capture the user's facial image and transmit the facial image to the facial recognition module; The face recognition module is used to recognize the face images acquired by the shooting module, determine whether the similarity between the face image acquired by the shooting module and the face images pre-stored in the face recognition module is greater than a set similarity threshold, and obtain the judgment result. The control module is used to control the power-on process of the smart projector based on the judgment result; The shooting module includes a first camera unit that is wired to the data input port of the smart projector. The first camera unit is used to capture the user's face and obtain the user's facial image. The user's face is photographed to obtain an image of the user's face, including: The camera continuously captures images of the area in front of the first camera unit, obtaining multiple real-time images. Multiple real-time images are filtered to obtain face images; Multiple real-time images are filtered to obtain face images, including: S1, acquire the earliest captured real-time image erpho and the latest captured real-time image latpho from multiple real-time images; S2, perform face region detection on erpho to obtain the face region blkfco in erpho; S3 expands the face region in Erpho to obtain the expanded region, which includes: S31, calculate the difference in shooting time between erpho and latpho : in, and These represent the shooting times for the erpho and latpho images, respectively. S32, the extended region is obtained using the following formula: Where timst represents the preset time length and chs represents the preset constant coefficient; S4, obtain the set of pixels in latpho that have the same coordinates as the pixels in the extended region, faccoef; S5, perform face region detection on the pixels in the set faccoef to obtain the face region blkfct in latpho; S6, obtain the minimum x-coordinate minxo, the maximum x-coordinate maxxo, the minimum y-coordinate minyo, and the maximum y-coordinate maxyo of the pixels in blkfco. S7, obtain the minimum x-coordinate minxt, the maximum x-coordinate maxxt, the minimum y-coordinate minyt, and the maximum y-coordinate maxyt of the pixels in blkfct; S8, obtain the detection area, the coordinate range of the detection area is: min means to get the smaller value in the parentheses, and max means to get the larger value in the parentheses; S9, calculate the detection coefficient of the detection region in each real-time image. The calculation method for the detection coefficient is as follows: in, Indicates the detection area The detection coefficient, and The preset values ​​represent window weights and edge weights, respectively. , The number of pixels in the window contained in the phost, where phost is the detection region. The collection of window pixels contained therein Let be the grayscale value of pixel i in the window, stgray represent the preset grayscale variance, and phobl represent the detection area. The set of edge pixels in the image. This represents the grayscale value of pixel j. Let j represent the set of pixels in the 8-neighborhood of pixel j. Indicates obtaining The maximum grayscale value of the pixels in the image. This indicates the number of pixels contained in the phobl; This represents the preset grayscale value constant; S10 uses the real-time image with the highest monitoring coefficient as the face image.

2. The intelligent projector based on biometric recognition function according to claim 1, characterized in that, The shooting module includes a wireless communication device installed inside the projector and a second camera unit capable of wirelessly communicating with the wireless communication device. The second camera unit is used to capture the user's face and obtain the user's facial image.

3. A smart projector based on biometric recognition function according to claim 2, characterized in that, The wireless communication device is either Bluetooth or WiFi.

4. A smart projector based on biometric recognition function according to claim 2, characterized in that, The second camera unit includes any one of a camera on a smartphone, a camera on a tablet, or a camera on a laptop.

5. A smart projector based on biometric recognition function according to claim 1, characterized in that, The face recognition module includes a receiving unit, a storage unit, a similarity calculation unit, and a judgment unit; The receiving unit is used to receive facial images transmitted from the shooting module; The storage unit is used to save pre-stored face images and preset similarity thresholds; The similarity calculation unit is used to calculate the similarity between the face image acquired by the shooting module and the face image pre-stored in the face recognition module; The judgment unit is used to determine whether the similarity is greater than the set similarity threshold and obtain the judgment result.

6. A smart projector based on biometric recognition function according to claim 5, characterized in that, The judgment result is either the similarity is greater than the set similarity threshold or the similarity is less than or equal to the set similarity threshold.

7. A smart projector based on biometric recognition function according to claim 6, characterized in that, Based on the judgment result, the power-on process of the smart projector is controlled, including: If the judgment result is that the similarity is greater than the set similarity threshold, then control the smart projector to project the startup desktop; If the similarity result is less than or equal to the set similarity threshold, the smart projector will project a preset prompt text to inform the user that the face recognition failed.

8. A smart projector based on biometric recognition function according to claim 1, characterized in that, It also includes a remote control module; The remote control module is used to send remote control signals to the smart projector.

9. A smart projector based on biometric recognition function according to claim 8, characterized in that, The remote control signal is either a Bluetooth signal or an infrared signal.

Images