Door access control method and system

Abstract

The present application discloses a door entry control system combined with three-dimensional recognition technology, which includes: a projection unit, an infrared camera, a control and computing unit, and an actuator. The projection unit is used to project a surface structured light field of infrared light to a predetermined area and a uniform illumination light field of infrared light. When the infrared camera projects infrared light, it collects facial images. The control and computing unit is connected to the projection unit and the infrared camera, and uses a machine learning method to perform facial recognition based on the facial image collected by the infrared camera to determine whether the face in the image is authorized; and based on the judgment result, a control instruction for unlocking or locking is issued. The actuator receives the control instruction from the control and computing unit, and performs unlocking or locking operations accordingly. The present application also discloses a door entry control method.

Description

Technical Field

[0001] The present invention generally relates to a door entry control method and system, and specifically, to a door entry control method and a door entry control system combined with three-dimensional recognition technology, such as a smart door lock, a smart access control system, etc. Background Art

[0002] Entrance control systems, such as door locks and access control systems, involve the security of controlled areas and use various identity authentication methods to control the access of personnel.

[0003] At present, door locks and access control products based on face recognition generally use two-dimensional face recognition technology, which lacks the ability to effectively obtain three-dimensional face information and relies only on shooting two-dimensional image information for recognition. They are easily attacked by counterfeit means such as photos, tablets, masks, and are affected by facial makeup, accessories, lighting changes, etc. Door locks and access control products using three-dimensional recognition technology are expected to overcome the above problems and improve product safety and ease of use.

[0004] However, traditional 3D imaging equipment uses laser line scanning or projection of multiple changing stripe structured lights to perform 3D imaging of objects. The objects must remain still during the acquisition process and maintain a stable and good external lighting environment. Body sensing 3D acquisition equipment uses a single structured light projection and shooting method to obtain 3D images. Since the above-mentioned equipment needs to achieve medium and long-distance 3D imaging, the equipment is large in size and has high power consumption. The above-mentioned 3D imaging / recognition technologies are difficult to apply to door locks and access control products.

[0005] In addition, since door locks are generally powered by batteries, they need to be able to work continuously for several months to a year without replacing the batteries, and the computing power of the computing chips that can be used in door locks is relatively low, so when applying 3D face recognition technology to door locks, there are great limitations in power consumption and cost. It can be seen that in addition to security and miniaturization, door locks have low power consumption and low computing requirements that are different from other products. Therefore, it is not suitable to directly transplant the 3D recognition technology solutions applied to other products to smart door locks. Summary of the invention

[0006] The purpose of the present invention is to provide a door entry control system combined with three-dimensional recognition technology, in particular, an intelligent door lock and access control system combined with three-dimensional recognition technology, and provide a corresponding door entry control method, which at least partially solves the above-mentioned deficiencies in the prior art.

[0007] According to one aspect of the present invention, a door entry control system is provided, which includes: a projection unit, an infrared camera, a control and computing unit, and an actuator. The projection unit is used to project a surface structured light field of infrared light and a uniform illumination light field of infrared light to a predetermined area. When the projection unit projects infrared light to the predetermined area, the infrared camera collects an image of a face located in the predetermined area. The control and computing unit is connected to the projection unit and the infrared camera, and includes a processor and a memory, wherein the memory stores processor-readable instructions, and the processor-readable instructions enable the processor to perform the following processing: perform face recognition based on the face image collected by the infrared camera using a machine learning method, and determine whether the face in the image is authorized; and issue a control instruction for unlocking or locking according to the judgment result. The actuator receives the control instruction from the control and computing unit, and performs unlocking or locking operations accordingly.

[0008] Preferably, the door entry control system may also include a proximity sensor for detecting whether someone enters the predetermined area, the proximity sensor includes a laser ranging sensor or a far-infrared sensing sensor, and the control and calculation unit may control the operation of the projection unit and the infrared camera according to the detection result of the proximity sensor.

[0009] Preferably, the surface structured light field projected by the projection unit is a structured light field to which spatially coded structured light technology is applied.

[0010] In some advantageous embodiments, the control and computing unit is configured to control the projection unit to simultaneously project the surface structured light field and the uniform illumination light field; and the processor-readable instructions in the control and computing unit enable the processor to use a machine learning method to perform face recognition based on a face image captured under the simultaneous illumination of the surface structured light field and the uniform illumination light field.

[0011] In some advantageous embodiments, the projection unit comprises a common light source and an optical element for simultaneously forming the surface structured light field and the uniform illumination light field based on the common light source. The optical element preferably comprises a diffractive optical element, namely, DOE.

[0012] In some advantageous embodiments, the brightness ratio of the simultaneously projected uniform illumination light field to the surface structured light field is in the range of 1:2 to 1:100, preferably in the range of 1:2 to 1:10.

[0013] Preferably, the working field angle of the projection unit and the infrared camera is in the range of 30° to 120°, and the working distance is in the range of 0.3 to 1 meter.

[0014] In some embodiments, the door entry control system may not include a visible light camera.

[0015] The door entry control system can be implemented as a battery-powered smart door lock or a continuously powered smart access control.

[0016] According to another aspect of the present invention, a door access control method is also provided, which includes: projecting a surface structured light field of infrared light and a uniform illumination light field of infrared light to a predetermined area; while projecting, collecting an image of a face located in the predetermined area; performing face recognition based on the face image using a machine learning method to determine whether the face in the image is authorized; and performing an unlocking or locking operation based on the determination result.

[0017] Preferably, the door entry control method may further include: using a proximity sensor to detect whether a person enters the predetermined area.

[0018] Preferably, the projected surface structured light field is a structured light field to which spatially coded structured light technology is applied.

[0019] In some advantageous embodiments, the projection of the surface structured light field and the uniform illumination light field is performed simultaneously; and the face recognition is achieved based on a face image captured under the simultaneous illumination of the surface structured light field and the uniform illumination light field.

[0020] In some advantageous embodiments, the brightness ratio of the simultaneously projected uniform illumination light field to the surface structured light field is in the range of 1:2 to 1:100, preferably in the range of 1:2 to 1:10. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:

[0022] Figure 1 is a schematic flow chart of an example of a door entry control method according to an embodiment of the present invention;

[0023] Figure 2 is a schematic block diagram of an example of a door entry control system according to an embodiment of the present invention;

[0024] Figure 3 is a schematic diagram of an example of a projection unit that can be applied to the door entry control method and system according to an embodiment of the present invention;

[0025] Figure 4 According to an embodiment of the present invention, Figure 3 Schematic diagram of a smart door lock with a projection unit shown;

[0026] Figure 5 is a schematic diagram of another example of a projection unit applicable to the door entry control method and system according to an embodiment of the present invention; and

[0027] Figure 6 According to an embodiment of the present invention, Figure 5 Schematic diagram of a smart door lock with a projection unit shown. DETAILED DESCRIPTION

[0028] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It should also be noted that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0029] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0030] First, refer to Figure 1 The following describes a door entry control method according to an embodiment of the present invention. Figure 1 In the example shown, the door entry control method 1 includes the following processing:

[0031] S10: Using a proximity sensor to detect whether a person approaches the area where door entry control is implemented (i.e., the predetermined area), and if so, a trigger signal is issued;

[0032] S20: Projecting a surface structured light field of infrared light and a uniform illumination light field of infrared light into the predetermined area;

[0033] S30: using an infrared camera to collect an image of a face in a predetermined area;

[0034] S40: Performing face recognition based on the face image using a machine learning method to determine whether the face in the image is authorized; and

[0035] S50: Perform an unlocking or locking operation based on the result of determining whether the face is authorized.

[0036] It should be noted that step S10 is an optional feature in the preferred example of the door entry control method according to the embodiment of the present invention. The door entry control method according to the embodiment of the present invention is not limited to including the subsequent projection and shooting / image acquisition work triggered by detecting the approach of a person. For example, in some implementations, the door entry control / recognition operation of the door entry control system can be triggered by a person manually pressing a button. Alternatively, in other implementations, especially when the corresponding door entry control system is continuously powered, the projection and image acquisition work can be performed continuously or automatically triggered intermittently.

[0037] In process S20, two different infrared light fields are projected onto the predetermined area, one is a surface structured light field, and the other is a uniform illumination light field.

[0038] The structured light field of infrared light is projected onto the face to form a three-dimensional information image. In the present invention, the projected surface structured light field is preferably a structured light field that applies spatially encoded structured light technology, and more preferably a random scattered spot array pattern that applies spatially encoded structured light technology. The structured light field that applies spatially encoded structured light technology makes it possible to obtain three-dimensional information in the entire structured light field area by projecting only a single structured light pattern (structured light field), without the need for scanning operations as in traditional three-dimensional imaging / detection equipment, thereby simplifying operations, saving time, and improving user experience. At the same time, this also means reducing the power consumption of the door entry control system, which is particularly beneficial for battery-powered door lock products, for example.

[0039] The infrared light illumination field is mainly used to assist in acquiring two-dimensional images of infrared light. It can serve as fill light under low-light conditions and help resist interference from ambient stray light.

[0040] According to some advantageous embodiments of the present invention, the projection of the surface structured light field and the uniform illumination light field is carried out simultaneously. This means that the projected infrared light uniform illumination light field will form the background light of the structured light light field, and when applying the structured light three-dimensional imaging / detection technology, it is usually required to minimize the brightness of the background light / stray light when projecting the structured light. However, the inventors of the present application have found that by controlling the brightness ratio of the structured light field and the illumination light field, and / or through the adaptive modification of the subsequent image processing algorithm, it is feasible to project the structured light field and the illumination light field at the same time. By projecting two infrared light fields at the same time and correspondingly performing facial image acquisition under the condition of simultaneous projection, the operations of projection and image acquisition are simplified, time is saved, the user experience is correspondingly improved, and it helps to reduce power consumption.

[0041] According to some advantageous embodiments, the brightness ratio between the uniform illumination light field and the surface structured light field that are projected simultaneously may be in the range of 1:2 to 1:100, preferably in the range of 1:2 to 1:10.

[0042] However, it should be understood that the door entry control method according to the embodiment of the present invention is not limited to simultaneously projecting the above-mentioned surface structured light field and the illumination light field of infrared light.

[0043] In process S30, an image of a face located in the predetermined area is collected. Although process S20 and process S30 are shown as being sequential, in practice, the projection in process S20 may be started first and continued to process S30, or the projection in process S20 and the image collection in process S30 may be controlled to be performed synchronously.

[0044] After collecting the face image, the door control method 1 enters the process S40, wherein, according to the embodiment of the present invention, the three-dimensional data and the two-dimensional data are processed in the computing unit, and the three-dimensional feature extraction and face recognition algorithm model operation are performed using machine learning (such as deep learning) methods to give a recognition result and determine whether the face in the image is authorized. Using the three-dimensional face information for recognition can greatly improve the anti-attack capability and reduce the false recognition rate.

[0045] In the embodiment discussed above in which the structured light field and the illumination light field are projected simultaneously in process S20, correspondingly, face recognition in process S40 is also achieved based on the face image collected under the simultaneous illumination of the surface structured light field and the illumination light field.

[0046] Next, in processing S50, according to the result of the judgment, the operation of unlocking (allowing passage) or locking (not allowing passage) in the door entry control is performed. Such operation can be performed by, for example, a certain actuator, which can be mechanically and / or electronically driven and / or controlled.

[0047] The above describes an example of a door entry control method according to an embodiment of the present invention. Figures 2 to 6 A door entry control system that can be used to implement the above-mentioned door entry control method is introduced in detail.

[0048] Figure 2 An example of a door entry control system according to an embodiment of the present invention is shown. In this example, the door entry control system 2 includes a projection unit 10, an infrared camera 20, a control and calculation unit 30, and an actuator 40.

[0049] The projection unit 10 is used to project a surface structured light field of infrared light and a uniform illumination light field of infrared light to a predetermined area (an area for door entry monitoring). When the projection unit 10 projects infrared light to the predetermined area, the infrared camera 20 collects an image of a face in the predetermined area. The control and calculation unit 30 is connected to the projection unit and the infrared camera 20, and receives the image collected by the infrared camera 20. According to an embodiment of the present invention, the control and calculation unit 30 uses a machine learning method to perform face recognition based on the face image collected by the infrared camera 20, determines whether the face in the image is authorized, and issues an unlocking or locking control instruction to the actuator 40 according to the judgment result. In order to realize the above functions, the control and calculation unit 30 may include a processor 31 and a memory 32, and the memory 32 stores processor-readable instructions, which, when executed by the processor 31, causes the processor 31 to perform the above-mentioned face recognition and other processing. The actuator 40 receives the control instruction from the control and calculation unit 30 and performs the unlocking or locking operation accordingly.

[0050] like Figure 2 As shown, the door control system 2 may also include a proximity sensor 50 for detecting whether a person enters the predetermined area. When the proximity sensor 50 detects that a person is approaching, a trigger signal may be sent to the control and computing unit 30. The control and computing unit 30 controls the operation of the projection unit 10 and the infrared camera 20 according to the detection result of the proximity sensor 50. The proximity sensor is used for automatic human body sensing and triggering, and the face acquisition device has a large coverage range, and there is no need to manually align the recognition device, so that non-sensing unlocking can be achieved, and the user experience is good. The use of the induction trigger method also helps to reduce the standby power consumption of the system.

[0051] The proximity sensor 50 can be in a continuous or intermittent working state, for example, it can sense the change of the position of people in a predetermined area in the environment at intervals (for example, 0.3 to 1 second). The predetermined area can be, for example, the monitoring range of the door lock or access control system for door entry control (for example, within about 1m in front of the door lock or access control). According to an embodiment of the present invention, the proximity sensor 50 can preferably adopt a laser ranging sensor or a far infrared sensing sensor. As proximity sensors, laser ranging sensors and infrared sensing sensors have the advantages of small size, low power consumption and low cost.

[0052] The control and calculation unit 30 includes a processor 31 and a memory 32, and may also include peripheral circuits (such as Figure 4 and Figure 6 The control and computing unit 30 may be integrated or distributed, for example, it may include more than one processor and / or more than one memory that are separately distributed but can work together. Figure 2In the example shown, after receiving the trigger signal from the proximity sensor 10 , the control and calculation unit 30 starts the projection unit 10 to project a certain structured light field and provide fill light illumination.

[0053] The infrared camera 20 in the present invention may include only one camera, which is conducive to miniaturization of the device and to reducing costs. In addition to the lens itself, the infrared camera 20 may include a filter (not shown), which is used to filter interfering light, preferably a near-infrared bandpass filter, with a bandpass range of, for example, 780 to 1100 nm. The lens working distance of the infrared camera 20 is preferably 0.3 to 1 meter, and the field of view angle is preferably 30° to 120°.

[0054] The door entry control method and system according to the embodiment of the present invention can be implemented using only one infrared camera, and there is no need to use a visible light camera to achieve three-dimensional imaging or perform manual alignment (the door lock and access control system in the prior art usually requires the use of a visible light camera in conjunction with a display screen for manual alignment). This is conducive to the miniaturization of the device and to reducing the power consumption of image acquisition. Of course, it should be understood that the method and system according to the embodiment of the present invention can work in conjunction with a visible light camera, and the present invention is not limited in this respect.

[0055] The control and calculation unit 30 controls the projection unit 10 and the infrared camera 20 to work together to collect the three-dimensional image light field of the face, and processes the three-dimensional data, and uses machine learning (such as deep learning) methods to extract three-dimensional features and calculate the face recognition algorithm model to give a recognition result. According to the recognition result, the actuator 40 is controlled to perform operations of allowing or not allowing passage.

[0056] The actuator 40 performs the operation of allowing or not allowing passage according to the control instruction of the control and computing unit 30 , which can be achieved, for example, by mechanical and / or electronic control of a corresponding door lock or access control system.

[0057] In order to project two infrared light fields, the projection unit 10 may have different implementation methods.

[0058] Figure 3An example of a projection unit applicable to the door access control method and system according to an embodiment of the present invention is schematically shown. In this example, the projection unit 10 includes two separate modules for forming and projecting different light fields, namely a structured light module 11 and a lighting module 12. It should be understood that the door access control system 2 according to an embodiment of the present invention is not limited to including only one structured light module 11, but a corresponding number of structured light modules 11 can be set according to the number of different structured light fields to be projected. The light source used by the structured light module 11 and the lighting module 12 is preferably in the near-infrared band, for example, with a wavelength in the range of 780 to 1100 nm. The use of the near-infrared band can avoid interference from visible light and reduce interference to the user.

[0059] The projection pattern of the structured light module 11 can be different types of structured light fields, preferably using a random scattered spot array pattern. The structured light module 11 preferably uses a light source based on an LD (Laser Diode) or VCSEL laser (Vertical Cavity Surface Emitting Laser), and is combined with a DOE (Diffractive Optical Elements) pattern generator structure. The projection angle α1 of the structured light module 11 is preferably between 30° and 120°, and the working distance (the distance between the projected near position F and the far position B and the structured light module 11) is preferably between 30° and 120°. Figure 3 and Figure 5 The range from L1 to L2 shown in FIG. 1 is preferably at least 0.3 to 1 meter. For example Figure 3 As shown in FIG. 1 , the distance L1 between the near position F to which the structured light module 11 can project and the structured light module 11 is 0.3 meters, and the distance L2 between the far position B and the structured light module 11 is 1 meter.

[0060] The lighting module 12 may be a light source using LED (Light Emitting Diode), LD or VCSEL, and may work in combination with a light homogenizing device. LED is preferably used as the light source, and DOE is preferably used as the light homogenizing device. The wavelength of light preferably used for lighting is also near-infrared, such as in the range of 780 to 1100 nm. The projection angle α2 and the working distance of the lighting module are the same or similar to those of the structured light module 11. In addition, in some advantageous embodiments, the structured light module 11 and the lighting module 12 are constructed so that the brightness ratio of the illumination light field and the structured light field is controlled between 1:2 and 1:100, preferably in the range of 1:2 to 1:10.

[0061] Figure 4 The embodiment of the present invention is shown in combination with Figure 3Schematic diagram of the smart door lock 100 with a projection unit shown. The structure and layout shown in this figure are only schematic and non-restrictive, and will not be described in detail here.

[0062] Figure 5 Another example of a projection unit applicable to the door entry control method and system according to an embodiment of the present invention is schematically shown. In this example, the projection unit 10 has an integrated structure for projecting both a structured light field and a uniform illumination light field. Specifically, Figure 5 The projection unit 10 shown includes a common light source 10a and an optical element 10b for simultaneously forming a surface structured light field and a uniform illumination light field based on the common light source. The optical element 10b preferably includes a diffractive optical element, i.e., DOE. It should be noted that the use of DOE to project a structured light field (structured light pattern) is itself accompanied by background illumination caused by stray light, and the projection unit of the present invention that can simultaneously project a structured light field and an illumination light field is different from this in that it realizes a background illumination light field with controlled contrast and uniformity while projecting the structured light field. In a preferred embodiment, Figure 5 The brightness ratio of the illumination light field and the surface structured light field simultaneously projected by the projection unit 10 in the example shown is controlled between 1:2 and 1:100, within the range of 1:2 to 1:10.

[0063] In addition, similar to Figure 3 The example shown, Figure 5 The working field angle α3 of the projection unit 10 in the illustrated example is in the range of 30° to 120°, and the working distance is in the range of 0.3 to 1 meter.

[0064] Figure 6 According to an embodiment of the present invention, Figure 5 Schematic diagram of a smart door lock with a projection unit shown. The structure and layout shown in this figure are only schematic and not restrictive, and will not be described in detail here.

[0065] The above describes the door entry control method and system according to the embodiment of the present invention in combination with the accompanying drawings and different examples. The door entry control system according to the embodiment of the present invention can be implemented as a battery-powered smart door lock or a sustainably powered smart access control, and the present invention is not limited thereto.

[0066] In addition, the door access control method and system of the present invention can also be used in combination with other authority control methods / systems such as fingerprints, door cards, passwords, keys, etc.

[0067] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the above features are replaced with (but not limited to) technical features with similar functions disclosed in the present application.

Claims

1. A door entry control system, comprising: A projection unit is used to simultaneously project a surface structured light field of near-infrared light and a uniform illumination light field of near-infrared light to a predetermined area, wherein: The wavelengths of the surface structured light field and the uniform illumination light field are both within the range of 780 to 1100 nm; The projection unit includes a common light source and an optical element for simultaneously forming the surface structured light field and the uniform illumination light field based on the common light source; and The brightness ratio of the uniform illumination light field and the surface structured light field projected simultaneously is in the range of 1:2 to 1:10; a single infrared camera, comprising a near-infrared bandpass filter having a bandpass range of 780 to 1100 nm, wherein when the projection unit projects near-infrared light to the predetermined area, the infrared camera is used to collect an image of a human face located in the predetermined area; A control and computing unit connected to the projection unit and the infrared camera, and comprising a processor and a memory, wherein the memory stores processor-readable instructions, the control and computing unit is configured to control the projection unit to simultaneously project the surface structured light field and the uniform illumination light field, and the processor-readable instructions enable the processor to perform the following processing: using a machine learning method to perform face recognition based on a face image captured by the infrared camera under the simultaneous illumination of the surface structured light field and the uniform illumination light field, to determine whether the face in the image is authorized; and issuing a control instruction to unlock or lock the door according to the determination result; and The actuator receives the control instruction from the control and calculation unit and executes the unlocking or locking operation accordingly.

2. The door entry control system as described in claim 1 further includes a proximity sensor for detecting whether someone enters the predetermined area, the proximity sensor includes a laser ranging sensor or a far-infrared sensing sensor, and the control and calculation unit controls the operation of the projection unit and the infrared camera according to the detection result of the proximity sensor.

3. The door entry control system according to claim 1, wherein: The surface structured light field projected by the projection unit is a structured light field to which the spatially coded structured light technology is applied.

4. The door entry control system according to claim 1, wherein: The optical element comprises a diffractive optical element, namely, DOE.

5. The door entry control system according to claim 1, wherein: The working field angle of the projection unit and the infrared camera is within the range of 30° to 120°, and the working distance is within the range of 0.3 to 1 meter.

6. The door entry control system according to claim 1, wherein: The access control system does not include a visible light camera.

7. The door entry control system according to claim 1, wherein: The door entry control system is a battery-powered smart door lock or a sustainably powered smart access control.

8. A door entry control method applied to the door entry control system according to claim 1, comprising: Using a common light source and an optical element for simultaneously forming the surface structured light field and the uniform illumination light field based on the common light source, a surface structured light field of near-infrared light and a uniform illumination light field of near-infrared light are simultaneously projected to a predetermined area, wherein a brightness ratio of the simultaneously projected uniform illumination light field to the surface structured light field is within a range of 1:2 to 1:10; While projecting, using a single infrared camera including a near infrared bandpass filter to collect an image of a human face located in the predetermined area, wherein the bandpass range of the near infrared bandpass filter is 780 to 1100 nm; Using a machine learning method, face recognition is performed based on the face image collected under the simultaneous illumination of the surface structured light field and the uniform illumination light field to determine whether the face in the image is authorized; as well as According to the judgment result, the unlocking or locking operation is performed.

9. The door entry control method according to claim 8, further comprising: A proximity sensor is used to detect whether a person enters the predetermined area.

10. The door entry control method according to claim 8 or 9, wherein: The projected surface structured light field is a structured light field to which spatially coded structured light technology is applied.

Images