Recognition device, door lock and recognition method

Abstract

The invention provides a recognition device, a door lock and a recognition method, and the recognition device comprises an infrared projector which is used for projecting infrared light to a target area; the multi-channel camera collects reflected light of a to-be-recognized target in the target area and obtains a first image of the to-be-recognized target; the infrared camera collects infrared light reflected by a to-be-recognized target in the target area and obtains a second image with the recognized target; the infrared projector, the multi-channel camera and the infrared camera are in communication connection with the processor, and the processor has a face brushing working mode, a palm brushing working mode and a cat eye working mode. The processor is used for acquiring a first image and a second image, switching among the face brushing working mode, the palm brushing working mode and the cat eye working mode on the basis of the distance of a to-be-recognized target, and recognizing the to-be-recognized target on the basis of the acquired first image and second image when the processor is in the face brushing working mode and the palm brushing working mode.

Description

Technical Field

[0001] The present invention relates to the field of optical technology, and in particular to an identification device, a door lock and an identification method. Background Art

[0002] With the gradual popularization of smart door locks, a smart door lock often integrates multiple biometric devices, such as fingerprint recognition devices and face recognition devices that have been widely used, as well as palm vein recognition devices that are gradually emerging; in addition to biometric devices, smart door locks will also integrate cat-eye cameras for remote monitoring.

[0003] In the prior art, fingerprint recognition devices, face recognition devices, vein recognition devices, and cat-eye cameras installed in door locks all need to be installed separately, and their structures and functions cannot interfere with each other. If these devices are integrated separately on the door lock, the overall device will be too large and the cost will be too high. Summary of the invention

[0004] A major advantage of the present invention is that it provides an identification device, a door lock and an identification method, wherein the identification device integrates face recognition function, palm vein and cat's eye function into one device, effectively improving the integration of the door lock end device, sharing of the three function realization devices, and greatly reducing the cost of the overall device.

[0005] Another advantage of the present invention is that it provides an identification device, a door lock and an identification method, wherein the identification device includes a multi-channel camera, the multi-channel camera can obtain color image information and infrared image information, and the ordinary infrared camera can obtain infrared image information. When entering the face scanning mode, the two images can be integrated and subjected to face detection and key point detection respectively, and key point matching can be achieved, thereby obtaining 3D information of the face area, which can better realize face liveness detection.

[0006] Another advantage of the present invention is that it provides an identification device, a door lock and an identification method, wherein the image provided by the multi-channel camera of the identification device contains color image information and infrared image information, and combined with the infrared image provided by an ordinary camera, in the face scanning and palm scanning modes, compared with a single infrared image, it can provide more prosthesis clues and achieve safer liveness detection.

[0007] Another advantage of the present invention is that it provides an identification device, a door lock and an identification method, wherein the area information of human body detection obtained by the identification device in the cat's eye mode can narrow the area range of face detection in the subsequent face scanning mode to achieve a faster face scanning effect.

[0008] Another advantage of the present invention is that it provides an identification device, a door lock, and an identification method, wherein the identification device has a wide range of applications, and for some usage scenarios requiring extreme security, dual verification of face scanning and palm scanning can be selected based on this set of equipment.

[0009] Another advantage of the present invention is that it provides an identification device, a door lock, and an identification method, wherein the identification device can choose to turn off face scanning for some people who are more sensitive to faces and only use palm scanning for verification.

[0010] According to one aspect of the present invention, an identification device of the present invention that can achieve the aforementioned objects and other objects and advantages includes:

[0011] An infrared projector for projecting infrared light toward a target area;

[0012] A multi-channel camera collects reflected light of the target to be identified in the target area and obtains a first image of the target to be identified;

[0013] An infrared camera collects reflected infrared light of an identified target in the target area and obtains a second image with the identified target; and

[0014] A processor, wherein the infrared projector, the multi-channel camera, and the infrared camera are communicatively connected to the processor, the processor has a face-scanning working mode, a palm-scanning working mode, and a cat's eye working mode, and switches between the face-scanning working mode, the palm-scanning working mode, and the cat's eye working mode based on the distance of the target to be identified, when the processor is in the face-scanning working mode and the palm-scanning working mode, the target to be identified is identified based on the acquired first image and the second image; when the processor is in the cat's eye working mode, the target to be identified is identified based on the first image.

[0015] According to one embodiment of the present invention, the processor controls the operating power of the infrared projector, and when the processor switches the operating mode, the processor adjusts the operating power of the infrared projector to adapt to the current operating mode.

[0016] According to one embodiment of the present invention, the processor sets a face-scanning working distance interval, a palm-scanning working distance interval, and a cat's eye working distance interval for the working mode of the recognition device, and adjusts its working mode according to the set working distance interval and the distance of the target to be recognized.

[0017] According to one embodiment of the present invention, the palm brush working distance interval is 5-15 cm, the face brush working distance interval is 30-120 cm, and a distance greater than 120 cm is a cat's eye working distance interval.

[0018] According to one embodiment of the present invention, the power of the infrared projector in the palm brushing working mode is greater than the power in the face brushing working mode and the power in the cat's eye working mode.

[0019] According to one embodiment of the present invention, it further includes a distance sensor, wherein the distance sensor is communicatively connected to the processor, and the distance sensor is used to detect the distance of the identification target within the target area and transmit the collected information to the processor.

[0020] According to an embodiment of the present invention, the distance sensor is an infrared distance sensor.

[0021] According to one embodiment of the present invention, the processor includes a control module, a detection module and a recognition module, wherein the control module switches the working mode based on the acquired distance information of the target to be recognized, wherein the detection module includes a face detection unit, a key point detection unit and a liveness detection unit, wherein the face detection unit performs face detection based on a first image to obtain a first face detection result; performs face detection based on a second image to obtain a second face detection result; the key point detection unit performs key point detection based on the first image to obtain the face key point position of the first image; performs key point detection based on the second image to obtain the face key point position of the second image; the liveness detection unit performs liveness detection based on the face key point position of the first image and the face key point position of the second image, and the recognition module performs face recognition based on the first face detection result and the face key point position.

[0022] According to one embodiment of the present invention, the detection module further includes a palm detection unit and an image processing unit, wherein the palm detection unit performs palm detection based on the first image and obtains a palm detection result of the first image, and performs palm detection based on the second image and obtains a palm detection result of the second image; the key point detection unit obtains a first key position of the hand based on the palm detection result of the first image, and obtains a second key position of the hand based on the palm detection result of the second image; the image processing unit rotates and crops the palm having a palm vein area based on the obtained palm detection result of the first image and the first key position of the hand to obtain a palm vein image of the first image, and rotates and crops the palm having a palm vein area based on the obtained palm detection result of the second image and the second key position of the hand to obtain a palm vein image of the second image; the liveness detection unit performs liveness detection based on the obtained palm vein image of the first image and the palm vein image of the second image to obtain a liveness detection result; the recognition module performs palm vein recognition based on the acquired palm vein image and obtains a palm vein recognition result.

[0023] According to one embodiment of the present invention, when the control module sets the current working mode to the cat-eye working mode, the control module controls the infrared camera to turn off, wherein the infrared projector projects infrared light outward, wherein the multi-channel camera captures a first image, and the detection module performs human body detection based on the first image, and outputs the current monitoring video when a human body is detected.

[0024] According to one embodiment of the present invention, the multi-channel camera is a wide-angle camera.

[0025] According to one embodiment of the present invention, it further includes a light intensity sensor and an infrared low-pass filter, wherein the light intensity sensor is connected to the processor and is used to collect the intensity of visible light in the target area, and the infrared low-pass filter is movably arranged at the front end of the multi-channel camera, and the position of the infrared low-pass filter is adjusted based on the light intensity of the target area collected by the light intensity sensor.

[0026] According to one embodiment of the present invention, when the recognition device is in the face scanning working mode or the palm scanning working mode, the infrared low-pass filter remains at the front end of the multi-channel camera; in the cat's eye mode, the light intensity sensor senses the intensity of visible light in the ambient light, and in the daytime or in an environment with good lighting conditions, the infrared low-pass filter is maintained at the front end of the multi-channel camera, and in low light conditions or at night, the infrared low-pass filter is moved to the outside of the multi-channel camera.

[0027] According to another aspect of the present application, the present application further provides a door lock, comprising:

[0028] Lock body; and

[0029] The identification device as described above is electrically connected to the lock body, wherein the identification device generates a corresponding control instruction based on the identification result and transmits the control instruction to the lock body, and the lock body is opened or closed based on the control instruction.

[0030] According to another aspect of the present application, the present application further provides an identification method, comprising the following steps:

[0031] (a) measuring the distance of the target to be identified in the target area;

[0032] (b) switching between a face scanning working mode, a palm scanning working mode and a cat's eye working mode based on the distance of the target to be identified; and

[0033] (c) In the face scanning working mode, a first image and a second image are collected, wherein the first image contains infrared and color information, and the second image contains infrared information, and face recognition is performed based on the first image and the second image; in the palm scanning working mode, a first image and a second image are collected, wherein the first image contains infrared and color information, and the second image contains infrared information, and palm vein recognition is performed based on the first image and the second image; in the cat's eye working mode, the first image is collected, and human body detection is performed based on the first image.

[0034] According to one embodiment of the present invention, in the step (b) of the recognition method, a face-scanning working distance interval, a palm-scanning working distance interval and a cat's eye working distance interval are set, and the working mode is switched according to the set working distance interval and the distance of the target to be recognized.

[0035] According to one embodiment of the present invention, the palm brush working distance interval is 5-15 cm, the face brush working distance interval is 30-120 cm, and a distance greater than 120 cm is a cat's eye working distance interval.

[0036] According to one embodiment of the present invention, in the step (b) of the identification method, infrared light is projected onto the target area, and the projection power of the infrared light is adjusted when the working mode is switched, wherein the projection power of the infrared light in the palm brushing working mode is greater than the projection power in the face brushing working mode and the cat's eye working mode.

[0037] According to one embodiment of the present invention, in step (c) of the recognition method, when in the face scanning working mode, the following steps are further included:

[0038] (c1.1) performing face detection based on the first image to obtain a first face detection result; performing face detection based on the second image to obtain a second face detection result;

[0039] (c1.2) performing key point detection based on the first image to obtain the positions of the key points of the face of the first image; performing key point detection based on the second image to obtain the positions of the key points of the face of the second image; and

[0040] (c1.3) Performing liveness detection based on the obtained facial key point positions of the first image and the obtained facial key point positions of the second image to obtain a liveness detection result.

[0041] According to one embodiment of the present invention, in step (c) of the recognition method, when in the face scanning working mode, the following steps are further included:

[0042] (c1.4) If the liveness detection result passes, face recognition is performed based on the first face detection result and the positions of the face key points to obtain the face recognition result.

[0043] According to an embodiment of the present invention, in step (c) of the identification method, when in the palm brushing working mode, the following steps are further included:

[0044] (c2.1) performing palm detection based on the first image and obtaining the palm detection result of the first image; performing palm detection based on the second image and obtaining the palm detection result of the second image

[0045] (c2.2) acquiring a first key position of the hand based on the palm detection result of the first image, and acquiring a second key position of the hand based on the palm detection result of the second image;

[0046] (c2.3) rotating and cropping the palm having the palm vein region based on the palm detection result of the first image and the first key position of the hand to obtain a palm vein image of the first image; rotating and cropping the palm having the palm vein region based on the palm detection result of the second image and the second key position of the hand to obtain a palm vein image of the second image; and

[0047] (c2.4) Performing liveness detection based on the obtained palm vein image of the first image and the obtained palm vein image of the second image to obtain a liveness detection result.

[0048] According to one embodiment of the present invention, the step (b) of the identification method further comprises the following steps:

[0049] (b.1) collecting the intensity of visible light in the target area; and

[0050] (b.2) In the face scanning working mode and the palm scanning working mode, the infrared low-pass filter is kept in front of the camera; in the cat's eye mode, the position of the infrared low-pass filter in front of the camera is adjusted according to the light intensity of the collected target area.

[0051] Further objects and advantages of the present invention will be fully apparent from an understanding of the following description and the accompanying drawings.

[0052] These and other objects, features and advantages of the present invention will be more fully understood from the following detailed description and accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] The technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. In the accompanying drawings, unless otherwise specified, the same reference numerals are used to represent the same components. Among them:

[0054] Figure 1 It is a simplified structural diagram of an identification device according to a first preferred embodiment of the present invention.

[0055] Figure 2 is a simplified diagram of the identification device according to the first preferred embodiment of the present invention.

[0056] Figure 3A and Figure 3B It is a side view of the identification device according to the above preferred embodiment of the present application.

[0057] Figure 4 It is a schematic diagram of the structural framework of the processor of the identification device according to the above preferred embodiment of the present application.

[0058] Figure 5 It is a flow chart of the recognition device in the face scanning working mode according to the above preferred embodiment of the present application.

[0059] Figure 6 It is a flow chart of the identification device in the palm-swiping working mode according to the above preferred embodiment of the present application.

[0060] Figure 7 It is a flow chart of the identification device in cat's eye working mode according to the above preferred embodiment of the present application.

[0061] Figure 8 It is a schematic structural diagram of a door lock according to another preferred embodiment of the present invention.

[0062] Fig. 9 It is a flowchart of a recognition method according to another preferred embodiment of the present invention. DETAILED DESCRIPTION

[0063] It should be pointed out that the embodiments shown in the drawings are only used as examples to specifically and vividly explain and illustrate the concept of the present invention. The size and structure thereof are not necessarily drawn to scale, nor do they constitute a limitation to the concept of the present invention.

[0064] The directional terms such as up, down, left, right, front, back, front, back, top, bottom, etc. mentioned or may be mentioned in this specification are defined relative to the structures shown in the various drawings. They are relative concepts and may change accordingly according to different positions and different usage conditions. Therefore, these or other directional terms should not be interpreted as restrictive terms.

[0065] Referring to the drawings of this application specification Figures 1 to 7As shown, a recognition device according to the first preferred embodiment of the application is explained in the following description. The recognition device is hereinafter referred to as the recognition device. The recognition device can be applied to the device of face recognition and palm vein recognition at the same time, and the device can realize the remote monitoring function similar to cat's eye. The recognition device includes an infrared projector 10, a multi-channel camera 20, an infrared camera 30, and a processor 40, wherein the infrared projector 10, the multi-channel camera 20, and the infrared camera 30 are connected to the processor 40 in a communicative manner, and the processor 40 controls the working state of the infrared projector 10, the multi-channel camera 20, and the infrared camera 30. The infrared projector 10 is used to project infrared light into a target area. When a target to be identified appears in the target area, the multi-channel camera 20 collects the reflected light of the target to be identified and obtains the first image corresponding to the target to be identified; the infrared camera 30 collects the reflected light of the infrared light of the target to be identified and obtains the second image of the target to be identified.

[0066] In this preferred embodiment of the present application, the recognition device has a face-scanning working mode, a palm-scanning working mode, and a cat's eye working mode, and the processor 40 switches between the face-scanning working mode, the palm-scanning working mode, and the cat's eye working mode based on the distance of the target to be recognized. The first image acquired by the multi-channel camera 20 and the second image acquired by the infrared camera 30 are transmitted to the processor 40, and the processor 40 recognizes the target to be recognized based on the first image and the second image.

[0067] When the recognition device is in the face-swiping working mode, the first image obtained by the multi-channel camera 20 and the second image obtained by the infrared camera are facial images corresponding to the target to be identified, and the processor 40 performs liveness detection and facial recognition based on the first image and the second image. When the recognition device is in the palm-swiping working mode, the first image obtained by the multi-channel camera 20 and the second image obtained by the infrared camera are palm or fingerprint images corresponding to the target to be identified, and the processor 40 performs liveness detection and palm vein recognition based on the first image and the second image to obtain a palm vein recognition result. When the recognition device is in the cat's eye working mode, the processor 40 controls the infrared camera 30 to turn off, and performs human body detection based on the first image.

[0068] It can be understood that in this preferred embodiment of the present application, the recognition device integrates the face recognition function, palm vein and cat's eye functions into one device, which effectively improves the integration of the door lock end device, shares the three functions to achieve the device, and greatly reduces the cost of the overall equipment.

[0069] The processor 40 sets the face-swiping working distance interval, the palm-swiping working distance interval and the cat's eye working distance interval for the working mode of the recognition device. When the target to be identified is in the palm-swiping working distance interval, the processor 40 switches to the palm-swiping working mode and adjusts the working power of the infrared projector 10 to meet the needs of the multi-channel camera 20 and the infrared camera 30; when the identification target is in the face-swiping working distance interval, the processor switches to the face-swiping working mode and adjusts the working efficiency of the infrared projector 10 to meet the needs of the multi-channel camera 20 and the infrared camera 30; when the target to be identified is in the cat's eye working distance interval, or face-swiping and palm-swiping fail, the processor 40 switches to the cat's eye working mode and adjusts the working efficiency of the infrared projector 10.

[0070] As an example, in this preferred embodiment of the present application, the palm working distance interval is (5-15cm), the face working distance interval is (30-120cm), and the distance greater than 120cm is the cat's eye working distance interval. It can be understood that in the present application, the above working distance intervals are only used as examples, not limitations, so the above working distance intervals can be adjusted according to actual needs.

[0071] It is worth mentioning that the processor 40 adjusts the working power of the infrared projector 10 when switching the working mode, wherein the power of the infrared projector 10 in the palm brushing working mode is greater than the power in the face brushing working mode and the cat's eye working mode.

[0072] like Figure 1 As shown, the identification device further includes a distance sensor 50, wherein the distance sensor 50 is communicatively connected to the processor 40, and the distance sensor 50 is used to detect the distance of the identification target within the target area and transmit the collected information to the processor 40. The processor 40 adjusts the working mode and the working power of the infrared projector 10 based on the distance information collected by the distance sensor 50.

[0073] Preferably, in this preferred embodiment of the present application, the distance sensor 50 is implemented as an infrared distance sensor. Optionally, in another optional implementation of the present application, the processor 40 obtains the distance information of the target to be identified based on the image information captured by the multi-channel camera 20 and the infrared camera 30. That is, the processor 40 obtains the distance of the target to be identified based on the pre-calibrated binocular camera composed of the multi-channel camera 20 and the infrared camera 30 through parallax calculation between the binocular cameras.

[0074] The processor 40 controls the working power of the infrared projector 10 according to different working modes, and identifies the target to be identified based on the images captured by the multi-channel camera 20 and the infrared camera 30.

[0075] like Figure 4 As shown, the processor 40 includes a control module 41, a detection module 42 and an identification module 43, wherein the control module 41 switches between the palm-swiping working mode, the face-swiping working mode and the cat's eye working mode based on the acquired distance information of the target to be identified, and adjusts the working power of the infrared projector 10 in a specific working mode so that the working power of the infrared projector 10 adapts to the current working mode.

[0076] That is to say, the control module 41 adjusts the working mode of the recognition device according to the set face-scanning working distance interval, the palm-scanning working distance interval, the cat's eye working distance interval and the acquired distance information of the target to be identified, and when a specific working mode is selected, the control module 41 adjusts the working power of the infrared projector 10 under the current working mode.

[0077] like Figure 5 As shown, when in the face scanning working mode, the infrared projector 10 projects infrared light into the target area, the multi-channel camera 20 captures the first image of the target to be identified, and the infrared camera 30 captures the second image of the target to be identified, wherein when there is ambient light, the first image contains infrared and color information, and the second image only contains infrared information.

[0078] When the control module 41 sets the current working mode to the face scanning working mode, the detection module 42 performs face liveness detection on the target to be identified based on the first image and the second image to identify whether the target to be identified is a live face. The detection module 42 includes a face detection unit 421, a key point detection unit 422, and a liveness detection unit 423, wherein the face detection unit 421 performs face detection based on the first image to obtain a first face detection result; performs face detection based on the second image to obtain a second face detection result; the key point detection unit 422 performs key point detection based on the first image to obtain the face key point position of the first image; performs key point detection based on the second image to obtain the face key point position of the second image; the liveness detection unit 423 performs liveness detection based on the obtained face key point position of the first image and the face key point position of the second image to obtain a liveness detection result.

[0079] The detection module 42 of the processor 40 determines that the live face detection is passed based on the first image and the second image, and the recognition module 43 performs face recognition according to the first face detection result and the positions of the key points of the face.

[0080] It is understandable that the multi-channel camera 20 can obtain color image information and infrared image information, and the infrared camera 30 can obtain infrared image information. Therefore, the multi-channel camera 20 and the infrared camera 30 can obtain local 3D information of the target to be identified by binocular matching. That is to say, the multi-channel camera 20 and the infrared camera 30 are arranged on the same plane, and the multi-channel camera 20 and the infrared camera 30 are arranged adjacent to each other, so that the field of view of the multi-channel camera 20 at least partially overlaps or coincides with the field of view of the infrared camera 30, so that when the multi-channel camera 20 and the infrared camera 30 shoot the target to be identified in the target area, the distance information of the target to be identified can be obtained according to the shot image.

[0081] When the recognition device is in the face scanning working mode, the images taken by the multi-channel camera 20 and the infrared camera 30 are respectively subjected to face detection and key point detection by the processor 40, and key point matching can be achieved, thereby obtaining 3D information of the face area, which can better realize face liveness detection.

[0082] like Figure 6 As shown, when in the palm-brushing working mode, the infrared projector 10 projects infrared light into the target area, the multi-channel camera 20 captures the first image of the target to be identified, and the infrared camera 30 captures the second image of the target to be identified, wherein when there is ambient light, the first image contains infrared and color information, and the second image only contains infrared information.

[0083] When the control module 41 sets the current working mode to the palm brushing working mode, the detection module 42 performs palm liveness detection on the target to be identified based on the first image and the second image to identify whether the target to be identified is a live palm.

[0084] like Figure 4As shown, the detection module 42 further includes a palm detection unit 424 and an image processing unit 425, wherein the palm detection unit 424 performs palm detection based on the first image and obtains the palm detection result of the first image; the palm detection unit 424 performs palm detection based on the second image and obtains the palm detection result of the second image. The key point detection unit 422 of the detection module 42 obtains the first key position of the hand based on the palm detection result of the first image, and obtains the second key position of the hand based on the palm detection result of the second image. The image processing unit 425 rotates and crops the palm with the palm vein area based on the palm detection result of the first image and the first key position of the hand to obtain the palm vein image of the first image; the image processing unit 425 rotates and crops the palm with the palm vein area based on the palm detection result of the second image and the second key position of the hand to obtain the palm vein image of the second image. The liveness detection unit 423 of the detection module 42 performs liveness detection based on the obtained palm vein image of the first image and the palm vein image of the second image to obtain the liveness detection result. The recognition module 43 performs palm vein recognition according to the acquired palm vein image and obtains a palm vein recognition result.

[0085] like Figure 7 As shown, when the control module 41 sets the current working mode to the cat-eye working mode, the control module controls the infrared camera 30 to turn off, wherein the infrared projector 10 projects infrared light outward, wherein the multi-channel camera captures a first image, and the detection module 42 performs human body detection based on the first image, and can output the current monitoring video when a human body is detected.

[0086] It is worth mentioning that in the cat's eye mode, the area information of human body detection obtained can be used to narrow the area range of face detection in the subsequent face scanning mode to achieve a faster face scanning effect.

[0087] Preferably, in this preferred embodiment of the present application, the multi-channel camera 20 is a wide-angle camera, and the multi-channel camera 20 is a camera that can accept all bands, and the received bands may include: narrow-band single band (infrared band, such as 850nm / 940nm, etc.), wide-area band (visible light band).

[0088] It is worth mentioning that in this preferred embodiment of the present application, in the face scanning and palm scanning modes, the image provided by the multi-channel camera 20 contains color image information and infrared image information combined with the infrared image provided by the infrared camera 30. Compared with the single infrared image, it can provide more prosthesis clues and achieve safer liveness detection.

[0089] like Figure 3A and Figure 3B As shown, the recognition device further includes a light intensity sensor 60 and an infrared low-pass filter 70, wherein the light intensity sensor 60 is connected to the processor 40, and is used to collect the intensity of visible light in the target area, and the infrared low-pass filter 70 is movably arranged at the front end of the multi-channel camera 20, and the position of the infrared low-pass filter 70 is adjusted based on the light intensity of the target area collected by the light intensity sensor 60.

[0090] When the recognition device is in the face-scanning working mode or the palm-scanning working mode, the infrared low-pass filter 70 remains at the front end of the multi-channel camera 20, wherein the image formed by the multi-channel camera 20 contains infrared information and color information. In the cat's eye mode, the light intensity sensor 60 senses the intensity of visible light in the ambient light. During the day or in an environment with good lighting conditions, the infrared low-pass filter 70 will remain at the front end of the multi-channel camera 20 to ensure the accuracy of the color image. In low-light conditions or at night, based on the information obtained by the light intensity sensor 60, the infrared low-pass filter 70 is moved to the outside of the multi-channel camera 20 to enhance the perception of infrared light, thereby providing better low-light images.

[0091] Referring to the drawings of this application specification Figure 8 As shown, a door lock according to another preferred embodiment of the present application is explained in the following description. The door lock includes a lock body 100 and an identification device 200, wherein the identification device 200 is arranged on the lock body 100, and the identification device 200 and the lock body 100 are electrically connected, the identification device 200 is used to identify the target to be identified in the target area, and convert the identification information into an instruction to transmit to the lock body 100, and the lock core 200 performs an opening or closing operation according to the instruction information sent by the identification device 200.

[0092] It is worth mentioning that the recognition device 200 can provide command information to the lock body 100 according to different command requirements. For example, for some use scenarios that require extreme security, dual verification of face and palm swiping can be selected based on this set of equipment, that is, the recognition device 200 needs to obtain the correct face information and palm information of the target to be identified before sending the unlocking command to the lock body 100. For some people who are more sensitive to human faces, they can choose to turn off face swiping and only use palm swiping verification, that is, the recognition device 200 only needs to obtain the correct palm information of the target to be identified before sending the unlocking command to the lock body 100.

[0093] The identification device 200 is the same as the identification device in the first preferred embodiment, and will not be described in detail. When the identification device 200 is in the cat's eye mode, the infrared camera 30 of the identification device 200 is turned off, and the multi-channel camera 20 captures the surveillance video in real time.

[0094] Referring to the drawings of this application specification Fig. 9 As shown, the identification method according to another preferred embodiment of the present application is explained in the following description. The identification method comprises the following steps:

[0095] (a) measuring the distance of the target to be identified in the target area;

[0096] (b) switching between a face scanning working mode, a palm scanning working mode and a cat's eye working mode based on the distance of the target to be identified; and

[0097] (c) In the face scanning working mode, a first image and a second image are collected, wherein the first image contains infrared and color information, and the second image contains infrared information, and face recognition is performed based on the first image and the second image; in the palm scanning working mode, a first image and a second image are collected, wherein the first image contains infrared and color information, and the second image contains infrared information, and palm vein recognition is performed based on the first image and the second image; in the cat's eye working mode, the first image is collected, and human body detection is performed based on the first image.

[0098] In step (b) of the identification method of the above preferred embodiment of the present application, a face-scanning working distance interval, a palm-scanning working distance interval and a cat's eye working distance interval are set, and the working mode is switched according to the set working distance interval and the distance of the target to be identified.

[0099] As an example, in this preferred embodiment of the present application, the palm working distance interval is (5-15cm), the face working distance interval is (30-120cm), and the distance greater than 120cm is the cat's eye working distance interval. It can be understood that in the present application, the above working distance intervals are only used as examples, not limitations, so the above working distance intervals can be adjusted according to actual needs.

[0100] In step (b) of the identification method of the above-mentioned preferred embodiment of the present application, infrared light is projected onto the target area, and the projection power of the infrared light is adjusted when the working mode is switched, wherein the projection power of the infrared light in the palm brushing working mode is greater than the projection power in the face brushing working mode and the cat's eye working mode.

[0101] When the target to be identified is in the palm-brushing working distance interval, the processor 40 switches to the palm-brushing working mode and adjusts the working power of the infrared projector 10 to adapt to the needs of the multi-channel camera 20 and the infrared camera 30; when the target to be identified is in the face-brushing working distance interval, the processor switches to the face-brushing working mode and adjusts the working efficiency of the infrared projector 10 to adapt to the needs of the multi-channel camera 20 and the infrared camera 30; when the target to be identified is in the cat's eye working distance interval, or face or palm brushing fails, the processor 40 switches to the cat's eye working mode and adjusts the working efficiency of the infrared projector 10.

[0102] In the step (c) of the identification method of the above preferred embodiment of the present application, when in the face scanning working mode, the following steps are further included:

[0103] (c1.1) performing face detection based on the first image to obtain a first face detection result; performing face detection based on the second image to obtain a second face detection result;

[0104] (c1.2) performing key point detection based on the first image to obtain the positions of the key points of the face of the first image; performing key point detection based on the second image to obtain the positions of the key points of the face of the second image; and

[0105] (c1.3) Performing liveness detection based on the obtained facial key point positions of the first image and the obtained facial key point positions of the second image to obtain a liveness detection result.

[0106] In the step (c) of the identification method of the above preferred embodiment of the present application, when in the face scanning working mode, the following steps are further included:

[0107] (c1.4) If the liveness detection result passes, face recognition is performed based on the first face detection result and the positions of the face key points to obtain the face recognition result.

[0108] In the step (c) of the identification method of the preferred embodiment of the present application, when in the palm brushing working mode, the following steps are further included:

[0109] (c2.1) performing palm detection based on the first image and obtaining the palm detection result of the first image; performing palm detection based on the second image and obtaining the palm detection result of the second image

[0110] (c2.2) acquiring a first key position of the hand based on the palm detection result of the first image, and acquiring a second key position of the hand based on the palm detection result of the second image;

[0111] (c2.3) rotating and cropping the palm having the palm vein region based on the palm detection result of the first image and the first key position of the hand to obtain a palm vein image of the first image; rotating and cropping the palm having the palm vein region based on the palm detection result of the second image and the second key position of the hand to obtain a palm vein image of the second image; and

[0112] (c2.4) Performing liveness detection based on the obtained palm vein image of the first image and the obtained palm vein image of the second image to obtain a liveness detection result.

[0113] In the step (c) of the identification method of the above preferred embodiment of the present application, when in the cat's eye working mode, the following step is further included: outputting the surveillance video when a person is detected.

[0114] Furthermore, the step (b) of the identification method in the above preferred embodiment of the present application further includes the following steps:

[0115] (b.1) collecting the intensity of visible light in the target area; and

[0116] (b.2) In the face scanning working mode and the palm scanning working mode, the infrared low-pass filter is kept in front of the camera; in the cat's eye mode, the position of the infrared low-pass filter in front of the camera is adjusted according to the light intensity of the collected target area.

[0117] It is worth mentioning that in cat-eye mode, the light intensity sensor senses the intensity of visible light in the ambient light. During the day or in an environment with good lighting conditions, the infrared low-pass filter will remain in front of the camera to ensure the accuracy of the color image. In low-light conditions or at night, based on the information obtained by the light intensity sensor, the infrared low-pass filter will be automatically moved away to enhance the perception of infrared light, thereby providing better low-light images.

[0118] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The purpose of the present invention has been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

[0119] The technical scope of the present invention is not limited to the contents in the above description. Those skilled in the art can make various deformations and modifications to the above embodiments without departing from the technical idea of ​​the present invention, and these deformations and modifications all belong to the protection scope of the present invention.

Claims

1. An identification device, characterized in that include: An infrared projector for projecting infrared light toward a target area; A multi-channel camera collects reflected light of the target to be identified in the target area and obtains a first image of the target to be identified; The infrared camera collects the reflected infrared light of the identified target in the target area and obtains a second image with the identified target; as well as A processor, wherein the infrared projector, the multi-channel camera, and the infrared camera are communicatively connected to the processor, the processor has a face-scanning working mode, a palm-scanning working mode, and a cat's eye working mode, and switches between the face-scanning working mode, the palm-scanning working mode, and the cat's eye working mode based on the distance of the target to be identified, when the processor is in the face-scanning working mode and the palm-scanning working mode, the target to be identified is identified based on the acquired first image and the second image; when the processor is in the cat's eye working mode, the target to be identified is identified based on the first image.

2. The identification device according to claim 1, wherein the processor controls the working power of the infrared projector, and adjusts the working power of the infrared projector to adapt to the current working mode when the processor switches the working mode.

3. The recognition device according to claim 2, wherein the processor sets a face-scanning working distance interval, a palm-scanning working distance interval, and a cat's eye working distance interval for the working mode of the recognition device, and adjusts its working mode according to the set working distance interval and the distance of the target to be identified.

4. The identification device according to claim 3, wherein the palm-scanning working distance interval is 5-15 cm, the face-scanning working distance interval is 30-120 cm, and a distance greater than 120 cm is a cat's eye working distance interval.

5. The identification device according to claim 2, wherein the power of the infrared projector in the palm-swiping working mode is greater than the power in the face-swiping working mode and the power in the cat's eye working mode.

6. The identification device according to claim 1, further comprising a distance sensor, wherein the distance sensor is communicatively connected to the processor, and the distance sensor is used to detect the distance of the target to be identified in the target area and transmit the collected information to the processor. The identification device according to claim 6 , wherein the distance sensor is an infrared distance sensor.

8. The recognition device according to claim 6, wherein the processor comprises a control module, a detection module and a recognition module, wherein the control module switches the working mode based on the distance information of the target to be recognized, wherein the detection module comprises a face detection unit, a key point detection unit and a liveness detection unit, wherein the face detection unit performs face detection based on the first image to obtain a first face detection result; performs face detection based on the second image to obtain a second face detection result; and the key point detection unit performs key point detection based on the first image to obtain the position of the key points of the face of the first image; Performing key point detection based on the second image to obtain the positions of facial key points of the second image; The liveness detection unit performs liveness detection based on the obtained facial key point positions of the first image and the facial key point positions of the second image, and the recognition module performs face recognition based on the first face detection result and the facial key point positions.

9. The recognition device according to claim 8, wherein the detection module further comprises a palm detection unit and an image processing unit, wherein the palm detection unit performs palm detection based on the first image and obtains a palm detection result of the first image, and performs palm detection based on the second image and obtains a palm detection result of the second image; the key point detection unit obtains a first key position of the hand based on the palm detection result of the first image, and obtains a second key position of the hand based on the palm detection result of the second image; the image processing unit rotates and crops the palm having a palm vein area based on the obtained palm detection result of the first image and the first key position of the hand to obtain a palm vein image of the first image, and rotates and crops the palm having a palm vein area based on the obtained palm detection result of the second image and the second key position of the hand to obtain a palm vein image of the second image; the liveness detection unit performs liveness detection based on the obtained palm vein image of the first image and the palm vein image of the second image to obtain a liveness detection result; the recognition module performs palm vein recognition based on the obtained palm vein image and obtains a palm vein recognition result.

10. The identification device according to claim 9, wherein when the control module sets the current working mode to the cat's eye working mode, the control module controls the infrared camera to turn off, wherein the infrared projector projects infrared light outward, wherein the multi-channel camera captures a first image, and the detection module performs human body detection based on the first image, and outputs the current monitoring video when a human body is detected. The recognition device according to claim 1 , wherein the multi-channel camera is a wide-angle camera.

12. The identification device according to claim 6 further comprises a light intensity sensor and an infrared low-pass filter, wherein the light intensity sensor is connected to the processor and is used to collect the intensity of visible light in the target area, and the infrared low-pass filter is movably arranged at the front end of the multi-channel camera, and the position of the infrared low-pass filter is adjusted based on the light intensity of the target area collected by the light intensity sensor.

13. The recognition device according to claim 12, wherein when the recognition device is in the face scanning working mode or the palm scanning working mode, the infrared low-pass filter remains at the front end of the multi-channel camera; in the cat's eye mode, the light intensity sensor senses the intensity of visible light in the ambient light, and in the daytime or in an environment with good lighting conditions, the infrared low-pass filter is maintained at the front end of the multi-channel camera, and in low light conditions or at night, the infrared low-pass filter is moved to the front end of the multi-channel camera.

14. Door lock, characterized in that, include: Lock body; and The identification device as described in any one of claims 1 to 13 is electrically connected to the lock body, wherein the identification device generates a corresponding control instruction based on the identification result, and transmits the control instruction to the lock body, and the lock body is opened or closed based on the control instruction.

15. The identification method, characterized in that The steps include: (a) measuring the distance of the target to be identified in the target area; (b) switching between a face scanning working mode, a palm scanning working mode and a cat's eye working mode based on the distance of the target to be identified; as well as (c) In the face scanning working mode, a first image and a second image are collected, wherein the first image contains infrared and color information, and the second image contains infrared information, and face recognition is performed based on the first image and the second image; in the palm scanning working mode, a first image and a second image are collected, wherein the first image contains infrared and color information, and the second image contains infrared information, and palm vein recognition is performed based on the first image and the second image; in the cat's eye working mode, the first image is collected, and human body detection is performed based on the first image.

16. The identification method according to claim 15, wherein in the step (b) of the identification method, a face-scanning working distance interval, a palm-scanning working distance interval and a cat's eye working distance interval are set, and the working mode is switched according to the set working distance interval and the distance of the target to be identified.

17. The identification method according to claim 16, wherein the palm-scanning working distance interval is 5-15 cm, the face-scanning working distance interval is 30-120 cm, and a distance greater than 120 cm is a cat's eye working distance interval.

18. The recognition method according to claim 16, wherein in the step (b) of the recognition method, infrared light is projected onto the target area, and the projection power of the infrared light is adjusted when the working mode is switched, wherein the projection power of the infrared light in the palm brushing working mode is greater than the projection power in the face brushing working mode and the cat's eye working mode.

19. The identification method according to claim 16, wherein in step (c) of the identification method, when in the face scanning working mode, further comprising the following steps: (c1.1) performing face detection based on the first image to obtain a first face detection result; performing face detection based on the second image to obtain a second face detection result; (c1.2) performing key point detection based on the first image to obtain the positions of facial key points of the first image; Performing key point detection based on the second image to obtain the positions of facial key points of the second image; as well as (c1.3) Performing liveness detection based on the obtained facial key point positions of the first image and the obtained facial key point positions of the second image to obtain a liveness detection result.

20. The identification method according to claim 19, wherein in step (c) of the identification method, when in the face scanning working mode, further comprising the following steps: (c1.4) If the liveness detection result passes, face recognition is performed based on the first face detection result and the positions of the face key points to obtain the face recognition result.

21. The identification method according to claim 20, wherein in step (c) of the identification method, when in the palm brushing working mode, further comprising the following steps: (c2.1) performing palm detection based on the first image and obtaining the palm detection result of the first image; performing palm detection based on the second image and obtaining the palm detection result of the second image (c2.2) acquiring a first key position of the hand based on the palm detection result of the first image, and acquiring a second key position of the hand based on the palm detection result of the second image; (c2.3) rotating and cropping the palm having the palm vein region based on the palm detection result of the first image and the first key position of the hand to obtain a palm vein image of the first image; rotating and cropping the palm having the palm vein region based on the palm detection result of the second image and the second key position of the hand to obtain a palm vein image of the second image; as well as (c2.4) Performing liveness detection based on the obtained palm vein image of the first image and the obtained palm vein image of the second image to obtain a liveness detection result.

22. The identification method according to claim 20, wherein the step (b) of the identification method further comprises the following steps: (b.1) collecting the intensity of visible light in the target area; and (b.2) In the face scanning working mode and the palm scanning working mode, the infrared low-pass filter is kept in front of the camera; in the cat's eye mode, the position of the infrared low-pass filter in front of the camera is adjusted according to the light intensity of the collected target area.

Images