CN112949505A - MCU-based offline face recognition intelligent door lock and control method - Google Patents
MCU-based offline face recognition intelligent door lock and control method Download PDFInfo
- Publication number
- CN112949505A CN112949505A CN202110245678.6A CN202110245678A CN112949505A CN 112949505 A CN112949505 A CN 112949505A CN 202110245678 A CN202110245678 A CN 202110245678A CN 112949505 A CN112949505 A CN 112949505A
- Authority
- CN
- China
- Prior art keywords
- face
- module
- door lock
- intelligent door
- mcu
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims description 28
- 239000004973 liquid crystal related substance Substances 0.000 claims abstract description 21
- 229910052751 metal Inorganic materials 0.000 claims abstract description 17
- 239000002184 metal Substances 0.000 claims abstract description 17
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 claims abstract description 11
- 229910052744 lithium Inorganic materials 0.000 claims abstract description 11
- HEZMWWAKWCSUCB-PHDIDXHHSA-N (3R,4R)-3,4-dihydroxycyclohexa-1,5-diene-1-carboxylic acid Chemical compound O[C@@H]1C=CC(C(O)=O)=C[C@H]1O HEZMWWAKWCSUCB-PHDIDXHHSA-N 0.000 claims abstract description 6
- 238000000605 extraction Methods 0.000 claims description 11
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 238000013527 convolutional neural network Methods 0.000 claims description 6
- 230000000694 effects Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 238000003708 edge detection Methods 0.000 claims description 3
- 230000006698 induction Effects 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000013499 data model Methods 0.000 abstract description 3
- 238000013139 quantization Methods 0.000 abstract description 3
- 230000006870 function Effects 0.000 description 20
- 238000010586 diagram Methods 0.000 description 3
- 230000015654 memory Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 210000003813 thumb Anatomy 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/172—Classification, e.g. identification
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Biophysics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Lock And Its Accessories (AREA)
- Collating Specific Patterns (AREA)
Abstract
The utility model provides an off-line face identification intelligence lock based on MCU, intelligence lock include two cameras and camera collection module, take touch function's liquid crystal and display drive module, take fingerprint identification function's metal handle, central processor module and supporting control circuit, lithium cell and supporting power module. The central processing unit module and the matched control circuit comprise an RT1060 low-power-consumption processor of an ARM cortex M7F based on NXP, a matched power supply DCDC circuit, a relay control circuit and a switching value input and output circuit. Compared with the prior art, the invention integrates the face recognition function and the function of the traditional door lock, abandons the open source face recognition SDK with large traditional resource demand and high cost requirement based on GPU hardmac and linux operating system, is suitable for large-scale popularization in common families or the field of intelligent door locks with strict cost requirement, greatly improves the performance of algorithm operation, simultaneously keeps the precision of the quantization algorithm basically consistent with the original floating point data model, and improves the precision of inference.
Description
Technical Field
The invention relates to the technical field of intelligent door locks, in particular to an off-line type face recognition intelligent door lock based on an MCU (microprogrammed control unit) and a control method thereof.
Background
In recent years, with the wide application of artificial intelligence in different industries and different fields, more and more people enjoy the benefits brought by artificial intelligence, including face recognition technology. At present, most open-source face recognition SDKs are based on a GPU (graphics processing unit) hardmac and a linux operating system, have large resource requirements and high cost requirements, and are not beneficial to large-scale popularization and deployment in the field of intelligent door locks facing common families or having very strict cost requirements. Meanwhile, the general intelligent door lock with the face recognition, which is popular in the current market, almost adopts an online face characteristic value extraction and recognition method, and the door lock cannot be opened due to the fact that the face cannot be recognized in places where a network cannot be used under network faults or other special conditions. In addition, almost all present face identification's intelligent lock all adopts annex outer hanging, also is that traditional lock and face identification device part, adopts this kind of design to have trouble and security problem of installation, because the people face belongs to people's absolute privacy, in case reveal will cause irrecoverable loss, consequently unite two into one face identification function and traditional lock's function, the hardware attribute that possesses off-line face identification simultaneously and draw function and low cost will play very important effect to the popularization based on face identification's intelligent lock.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides an off-line face recognition intelligent door lock based on an MCU (microprogrammed control unit) as follows:
the technical scheme of the invention is realized as follows:
an off-line type face recognition intelligent door lock based on an MCU (microprogrammed control Unit), which comprises a double-camera and camera acquisition module, a liquid crystal and display driving module with a touch function, a metal handle with a fingerprint recognition function, a central processor module and a matched control circuit, a lithium battery and a matched power supply module,
the central processing unit module and the matched control circuit comprise:
an RT1060 low-power processor of ARM Cortex M7F based on NXP, wherein the RT1060 low-power processor is used for inputting, detecting, identifying and controlling outputting of human faces and fingerprints;
the matched power supply DCDC circuit is used for linearly converting the voltage output by the lithium battery into various paths of voltage required by the RT1060 low-power-consumption processor and voltage required by the matched control circuit;
the relay control circuit is mainly used for controlling the magnetic lock plunger to be matched with the switching value input and output circuit to complete the in and out of the magnetic lock plunger,
and the switching value input and output circuit is used for being matched with the magnetic lock plunger and completing the in-out operation of the magnetic lock plunger.
The dual-camera and camera acquisition module, the liquid crystal and display driving module with the touch function, the lithium battery and the matched power supply module are respectively and electrically connected with the central processor module and the matched control circuit.
Preferably, the double-camera and camera acquisition module is arranged on the front upper side of the intelligent door lock, forms an angle of 45 degrees with the plane of the intelligent door lock, comprises two CMOS full-high-definition cameras and is used for acquiring a human face image in real time and sending the image to the central processing unit module; the two CMOS full-high-definition cameras are fisheye cameras, and the human face is shot in the direction of 20-160 degrees of the shooting face.
Preferably, take touch function's liquid crystal and show drive module includes 2.6 inches, resolution ratio 240x 320's high bright OLED LCD screen, a supporting capacitive touch screen and a SPI serial ports drive circuit, the facial image that the camera was gathered passes through behind the central processing unit module, appear in real time to the OLED LCD screen on through the SPI interface for user perception face position, fingerprint and identification process, the discernment failure can show red, and the discernment successfully shows green, capacitive touch screen provides the user to the operation of interface prompt nature for local face eigenvalue and fingerprint eigenvalue etc. of drawing, the OLED LCD screen is installed the front side of intelligence lock, and intelligent lock main part are a plane.
Preferably, the metal handle with the fingerprint identification function comprises a fingerprint identification module and a metal handle, and the fingerprint identification module is embedded in the middle of the metal handle and used for collecting and identifying the fingerprint of a user.
Preferably, the smart door lock further comprises a conventional lock hole for unlocking with a key in case of emergency.
The invention also provides an intelligent door lock control method based on the MCU face recognition algorithm, which comprises the following steps:
(1) the intelligent door lock is characterized in that the intelligent door lock is integrated with two cameras to automatically track and shoot a human body entering a shooting area, one camera is used for shooting and tracking a human face, the other camera is used for shooting the depth of field of the human face, video images jointly shot by the two cameras are transmitted to the inside of the central processing unit module in real time, a three-dimensional human face 3D image is constructed through an image preprocessing program integrated in the central processing unit module, image chord degree transformation and human face image edge detection are carried out at the same time, and the position of the human face is automatically locked;
(2) the camera acquisition module is internally provided with an infrared induction sensor, can automatically sense the temperature of an area which enters a shooting area and is locked with the position of a human face, and ensures that the currently shot human face is a non-photo image by combining with a human face depth image;
(3) the central processor module automatically takes a picture of the image of the locked face position area and transmits the result to a face recognition inference algorithm integrated inside;
(4) when the inference algorithm is successfully identified, the central processor module sends a switching signal to the relay control module, the magnetic lock plunger is sucked in, and the door is automatically opened;
preferably, the inference algorithm mainly comprises: the method comprises 6 steps of face initial identification with a CNN neural network convolution algorithm, a face activity detection process, face quality classification, automatic face alignment, face depth identification with a CNN neural network regression algorithm, face characteristic value extraction and comparison, the whole process of face identification is completed, if face identification is successful, an upper dotted line outer frame of an OLED liquid crystal display is automatically updated, green represents that identification is successful, and red represents that identification is unsuccessful.
Preferably, the control method further comprises a face feature value storage step, and the face feature value extraction and face feature value storage step comprises the following steps:
(1) aligning the face of the user to the double cameras, and observing the face position on the liquid crystal screen;
(2) clicking any position of the capacitive touch screen, and selecting 'inputting a face' on a popped dialog box;
(3) after the selection is finished, the user aims at the camera station for 1-2 seconds, and at the moment, the face recognition algorithm can automatically detect the face and extract a characteristic value;
(4) and after the extraction is successful, a dialog box is popped up, the user inputs the user name to be stored in the face, the pinyin or the number is used for replacing the input, after the input is finished, the 'save' button is clicked, and at the moment, the face characteristic value of the user is stored in the local database.
Compared with the prior art, the invention has the following beneficial effects:
the off-line face recognition intelligent door lock based on the MCU and the control method combine the face recognition function and the function of the traditional door lock into a whole, abandon the open source face recognition SDK with large resource requirement and high cost requirement based on GPU hardmac and linux operating system, and are suitable for large-scale popularization and deployment in common families or the field of intelligent door locks with strict cost requirement; the MCU-oriented hardware scheme provided by the invention can build the minimum face recognition hardware scheme only by using the MCU (with certain FLASH and SRAM memories in the chip) and a DCDC power supply, and can be completely used in the field of door locks with strictly limited areas.
Drawings
FIG. 1 is a schematic circuit diagram of an off-line face recognition intelligent door lock based on an MCU (microprogrammed control unit);
FIG. 2 is an appearance schematic diagram of the off-line face recognition intelligent door lock based on the MCU;
FIG. 3 is a flow chart of the intelligent door lock control method based on the MCU face recognition algorithm of the present invention;
FIG. 4 is a flow chart of an off-line face recognition inference algorithm flow chart of the present invention;
FIG. 5 is an architecture diagram of an offline face recognition algorithm of the present invention.
In the figure: the device comprises two cameras 1, a camera acquisition module 2, a liquid crystal and display driving module 3 with a touch function, a metal handle 4 with a fingerprint identification function, a central processing unit module and a matched control circuit 5, and a lithium battery and a matched power supply module 6.
Detailed Description
The present invention will now be described more fully hereinafter with reference to the accompanying drawings, in which embodiments of the invention are shown.
As shown in fig. 1-2, an off-line face identification intelligence lock based on MCU, intelligence lock includes two cameras 1 and camera collection module 2, takes touch function's liquid crystal and display drive module 3, takes fingerprint identification function's metal handle 4, central processing unit module and supporting control circuit 5, lithium cell and supporting power module 6, central processing unit module and supporting control circuit 5 include:
an RT1060 low-power processor of ARM Cortex M7F based on NXP, wherein the RT1060 low-power processor is used for inputting, detecting, identifying and controlling outputting of human faces and fingerprints;
the matched power supply DCDC circuit is used for linearly converting the voltage output by the lithium battery into various paths of voltage required by the RT1060 low-power-consumption processor and voltage required by the matched control circuit;
the relay control circuit is mainly used for controlling the magnetic lock plunger to be matched with the switching value input and output circuit to complete the in and out of the magnetic lock plunger,
and the switching value input and output circuit is used for being matched with the magnetic lock plunger and completing the in-out operation of the magnetic lock plunger.
The dual-camera and camera acquisition module, the liquid crystal and display driving module with the touch function, the lithium battery and the matched power supply module are respectively and electrically connected with the central processor module and the matched control circuit.
The double cameras 1 and the camera acquisition module 2 are arranged on the front upper side of the intelligent door lock, form an angle of 45 degrees with the plane of the intelligent door lock, and comprise two CMOS full-high-definition cameras, wherein the CMOS full-high-definition cameras are positioned on the upper part of the liquid crystal and display driving module 3 with the touch function and are used for acquiring face images in real time and sending the images to the central processing unit module; the two CMOS full-high-definition cameras are fisheye cameras, and the human face is shot in the direction of 20-160 degrees of the shooting face.
Take touch function's liquid crystal and show drive module 3 includes 2.6 inches, resolution ratio 240x 320's high bright OLED LCD screen, a supporting capacitive touch screen and a SPI serial ports drive circuit, the face image that the camera was gathered passes through behind the central processing unit module, appear in real time on the OLED LCD screen through the SPI interface for user's perception face position, fingerprint and identification process, the discernment failure can show red, and the discernment successfully shows green, in addition, indicate the colour can not be restricted to red and green, still can include other, capacitive touch screen provides the user to interface suggestive's operation, be used for local extraction face eigenvalue and fingerprint eigenvalue etc, the OLED LCD screen is installed the front side of intelligence lock, and intelligent lock main part is a plane.
The metal handle with the fingerprint identification function comprises a fingerprint identification module and a metal handle, wherein the fingerprint identification module is embedded in the middle of the metal handle and used for collecting and identifying fingerprints of a user, and the fingerprint identification module is arranged on the front side of the metal handle. The distance between the fingerprint identification module and the tail end of the metal handle is 13cm, so that a user can hold the metal handle and simultaneously ensure that a thumb is directly pressed on a fingerprint identification area of the fingerprint identification module, and the user can conveniently open the door.
The intelligent door lock further comprises a traditional lock hole, wherein the traditional lock hole is used for unlocking by a key under an emergency condition, or the intelligent door lock can be unlocked by the key under the condition that the battery is not electrified.
As shown in fig. 3 and 5, the invention also provides an intelligent door lock control method based on the MCU face recognition algorithm, which comprises the following steps:
(1) the intelligent door lock is characterized in that the intelligent door lock is integrated with two cameras to automatically track and shoot a human body entering a shooting area, one camera is used for shooting and tracking a human face, the other camera is used for shooting the depth of field of the human face, video images jointly shot by the two cameras are transmitted to the inside of the central processing unit module in real time, a three-dimensional human face 3D image is constructed through an image preprocessing program integrated in the central processing unit module, image chord degree transformation and human face image edge detection are carried out at the same time, and the position of the human face is automatically locked;
(2) the camera acquisition module is internally provided with an infrared induction sensor, can automatically sense the temperature of an area which enters a shooting area and is locked with the position of a human face, and ensures that the currently shot human face is a non-photo image by combining with a human face depth image;
(3) the central processor module automatically takes a picture of the image of the locked face position area and transmits the result to a face recognition inference algorithm integrated inside;
(4) when the inference algorithm is successfully identified, the central processor module sends a switching signal to the relay control module, the magnetic lock plunger is sucked in, and the door is automatically opened;
the reasoning algorithm mainly comprises the following steps: the method comprises 6 steps of face initial identification with a CNN neural network convolution algorithm, a face activity detection process, face quality classification, automatic face alignment, face depth identification with a CNN neural network regression algorithm, face characteristic value extraction and comparison, the whole process of face identification is completed, if face identification is successful, an upper dotted line outer frame of an OLED liquid crystal display is automatically updated, green represents that identification is successful, and red represents that identification is unsuccessful.
The control method also comprises face characteristic value storage, and the face characteristic value extraction and face characteristic value storage comprise the following steps:
(1) aligning the face of the user to the double cameras, and observing the face position on the liquid crystal screen;
(2) clicking any position of the capacitive touch screen, and selecting 'inputting a face' on a popped dialog box;
(3) after the selection is finished, the user aims at the camera station for 1-2 seconds, and at the moment, the face recognition algorithm can automatically detect the face and extract a characteristic value;
(4) and after the extraction is successful, a dialog box is popped up, the user inputs the user name to be stored in the face, the pinyin or the number is used for replacing the input, after the input is finished, the 'save' button is clicked, and at the moment, the face characteristic value of the user is stored in the local database.
The invention provides an offline type face recognition intelligent door lock based on an MCU (microprogrammed control Unit), which adopts an RT1060 low-power-consumption processor of ARM Cortex M7F based on NXP, simultaneously, hardware resources of the door lock comprise a memory, a processor instruction speed and a DSP floating point instruction, and various parameters of a face model of the MCU oriented to strict limitation of resources are obtained through deep optimization and cutting of an AI training model for face recognition. As shown in fig. 4, on the inference side, the floating point instruction with DSP and the MAC unit are fully combined to accelerate the operation of the model algorithm, thereby greatly improving the performance of the algorithm operation. Meanwhile, the precision of the quantization algorithm is basically consistent with that of the original floating point data model, and the precision of reasoning is improved.
Because the door lock is required to be integrated, the MCU-oriented hardware scheme provided by the invention can build a minimum face recognition hardware scheme only by using the MCU (with a certain FLASH and SRAM memories in the chip) and a DCDC power supply, and can be completely used in the field of door locks with strictly limited areas.
Aiming at the face recognition scheme of the MCU, the invention adopts the MCU with low cost, low power consumption and low size. The simultaneous fast start is also a feature of the MCU. The RTOS employed by the MCU enables system startup control to be within a few hundred milliseconds, whereas Linux kernel plus GUI with MPU typically requires more than 10 seconds to start. In this regard, the IoT face recognition module will bring a good user experience even if the entire system is powered off and restarted.
The circuit structure and the control method are integrated, so that the off-line face recognition intelligent door lock based on the MCU and the control method combine the face recognition function and the function of the traditional door lock, abandon the open source face recognition SDK based on a GPU hardmac and a linux operating system with high traditional resource requirements and high cost requirements, are suitable for large-scale popularization and deployment in common families or the field of intelligent door locks with strict cost requirements, greatly improve the running performance of the algorithm by adopting the algorithm, basically keep the precision of the quantization algorithm consistent with the precision of the original floating point data model, and improve the precision of reasoning.
Claims (8)
1. An off-line face recognition intelligent door lock based on an MCU (microprogrammed control Unit), which is characterized by comprising a double-camera and camera acquisition module, a liquid crystal and display driving module with a touch function, a metal handle with a fingerprint recognition function, a central processor module and a matched control circuit, a lithium battery and a matched power supply module,
the central processing unit module and the matched control circuit comprise:
an RT1060 low-power processor of ARM Cortex M7F based on NXP, wherein the RT1060 low-power processor is used for inputting, detecting, identifying and controlling outputting of human faces and fingerprints;
the matched power supply DCDC circuit is used for linearly converting the voltage output by the lithium battery into various paths of voltage required by the RT1060 low-power-consumption processor and voltage required by the matched control circuit;
the relay control circuit is mainly used for controlling the magnetic lock plunger to be matched with the switching value input and output circuit to complete the in and out of the magnetic lock plunger,
and the switching value input and output circuit is used for being matched with the magnetic lock plunger and completing the in-out operation of the magnetic lock plunger.
The dual-camera and camera acquisition module, the liquid crystal and display driving module with the touch function, the lithium battery and the matched power supply module are respectively and electrically connected with the central processor module and the matched control circuit.
2. The MCU-based offline type face recognition intelligent door lock is characterized in that the double-camera and camera acquisition module is arranged on the front upper side of the intelligent door lock and forms a 45-degree angle with the plane of the intelligent door lock, and comprises two CMOS full-high-definition cameras for acquiring face images in real time and sending the images to the central processing unit module; the two CMOS full-high-definition cameras are fisheye cameras, and the human face is shot in the direction of 20-160 degrees of the shooting face.
3. The MCU-based offline face recognition intelligent door lock is characterized in that the touch-function liquid crystal and display driving module comprises a 2.6-inch high-brightness OLED liquid crystal screen with the resolution of 240x320, a matched capacitive touch screen and an SPI serial port driving circuit, a face image collected by the camera is displayed on the OLED liquid crystal screen through an SPI interface in real time after passing through the central processing unit module, the face image is used for a user to sense the face position, the fingerprint and the recognition process, the red color can be displayed when the recognition fails, the green color can be displayed when the recognition succeeds, the interface prompt operation is provided for the user by the capacitive touch screen and used for locally extracting the face characteristic value, the fingerprint characteristic value and the like, the OLED liquid crystal screen is installed on the front side of the intelligent door lock, and the face image and the fingerprint characteristic value and the main body of the intelligent door lock form a plane.
4. The MCU-based offline type face recognition intelligent door lock is characterized in that the metal handle with the fingerprint recognition function comprises a fingerprint recognition module and a metal handle, wherein the fingerprint recognition module is embedded in the middle of the metal handle and is used for collecting and recognizing fingerprints of users.
5. The MCU-based offline face-recognition smart door lock according to claim 3, wherein said smart door lock further comprises a conventional lock hole, said conventional lock hole is used for key unlocking in case of emergency.
6. An intelligent door lock control method based on an MCU face recognition algorithm is characterized by comprising the following steps:
(1) the intelligent door lock is characterized in that the intelligent door lock is integrated with two cameras to automatically track and shoot a human body entering a shooting area, one camera is used for shooting and tracking a human face, the other camera is used for shooting the depth of field of the human face, video images jointly shot by the two cameras are transmitted to the inside of the central processing unit module in real time, a three-dimensional human face 3D image is constructed through an image preprocessing program integrated in the central processing unit module, image chord degree transformation and human face image edge detection are carried out at the same time, and the position of the human face is automatically locked;
(2) the camera acquisition module is internally provided with an infrared induction sensor, can automatically sense the temperature of an area which enters a shooting area and is locked with the position of a human face, and ensures that the currently shot human face is a non-photo image by combining with a human face depth image;
(3) the central processor module automatically takes a picture of the image of the locked face position area and transmits the result to a face recognition inference algorithm integrated inside;
(4) and when the inference algorithm is successfully identified, the central processor module sends a switching signal to the relay control module, the magnetic lock bolt is sucked in, and the door is automatically opened.
7. The intelligent door lock control method based on the MCU face recognition algorithm as claimed in claim 6, wherein the inference algorithm mainly comprises: the method comprises 6 steps of face initial identification with a CNN neural network convolution algorithm, a face activity detection process, face quality classification, automatic face alignment, face depth identification with a CNN neural network regression algorithm, face characteristic value extraction and comparison, the whole process of face identification is completed, if face identification is successful, an upper dotted line outer frame of an OLED liquid crystal display is automatically updated, green represents that identification is successful, and red represents that identification is unsuccessful.
8. The intelligent door lock control method based on the MCU face recognition algorithm as claimed in claim 7, wherein the control method further comprises face feature value storage, and the face feature value extraction and face feature value storage comprise the following steps:
(1) aligning the face of the user to the double cameras, and observing the face position on the liquid crystal screen;
(2) clicking any position of the capacitive touch screen, and selecting 'inputting a face' on a popped dialog box;
(3) after the selection is finished, the user aims at the camera station for 1-2 seconds, and at the moment, the face recognition algorithm can automatically detect the face and extract a characteristic value;
(4) and after the extraction is successful, a dialog box is popped up, the user inputs the user name to be stored in the face, the pinyin or the number is used for replacing the input, after the input is finished, the 'save' button is clicked, and at the moment, the face characteristic value of the user is stored in the local database.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110245678.6A CN112949505A (en) | 2021-03-05 | 2021-03-05 | MCU-based offline face recognition intelligent door lock and control method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110245678.6A CN112949505A (en) | 2021-03-05 | 2021-03-05 | MCU-based offline face recognition intelligent door lock and control method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112949505A true CN112949505A (en) | 2021-06-11 |
Family
ID=76247873
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110245678.6A Pending CN112949505A (en) | 2021-03-05 | 2021-03-05 | MCU-based offline face recognition intelligent door lock and control method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112949505A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1568489A (en) * | 2001-10-17 | 2005-01-19 | 拜尔丹泰梯系统公司 | Face imaging system for recordal and automated identity confirmation |
| US20060204050A1 (en) * | 2005-02-28 | 2006-09-14 | Kabushiki Kaisha Toshiba | Face authenticating apparatus and entrance and exit management apparatus |
| CN105096420A (en) * | 2015-07-31 | 2015-11-25 | 北京旷视科技有限公司 | Access control system and data processing method for same |
| CN208781311U (en) * | 2018-10-16 | 2019-04-23 | 麦域智能科技(上海)有限公司 | A kind of lock having human face scanning open function |
| CN209328094U (en) * | 2018-09-26 | 2019-08-30 | 深圳阜时科技有限公司 | Living body detection device and its access control system |
| CN209496743U (en) * | 2019-05-06 | 2019-10-15 | 南京机电职业技术学院 | A kind of electric-controlled lock device for door |
| CN110458041A (en) * | 2019-07-19 | 2019-11-15 | 国网安徽省电力有限公司建设分公司 | A kind of face identification method and system based on RGB-D camera |
| CN210402486U (en) * | 2019-11-20 | 2020-04-24 | 辽宁丽景科技有限公司 | Induction self-powered face recognition intelligent door lock based on TOF technology |
-
2021
- 2021-03-05 CN CN202110245678.6A patent/CN112949505A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1568489A (en) * | 2001-10-17 | 2005-01-19 | 拜尔丹泰梯系统公司 | Face imaging system for recordal and automated identity confirmation |
| US20060204050A1 (en) * | 2005-02-28 | 2006-09-14 | Kabushiki Kaisha Toshiba | Face authenticating apparatus and entrance and exit management apparatus |
| CN105096420A (en) * | 2015-07-31 | 2015-11-25 | 北京旷视科技有限公司 | Access control system and data processing method for same |
| CN209328094U (en) * | 2018-09-26 | 2019-08-30 | 深圳阜时科技有限公司 | Living body detection device and its access control system |
| CN208781311U (en) * | 2018-10-16 | 2019-04-23 | 麦域智能科技(上海)有限公司 | A kind of lock having human face scanning open function |
| CN209496743U (en) * | 2019-05-06 | 2019-10-15 | 南京机电职业技术学院 | A kind of electric-controlled lock device for door |
| CN110458041A (en) * | 2019-07-19 | 2019-11-15 | 国网安徽省电力有限公司建设分公司 | A kind of face identification method and system based on RGB-D camera |
| CN210402486U (en) * | 2019-11-20 | 2020-04-24 | 辽宁丽景科技有限公司 | Induction self-powered face recognition intelligent door lock based on TOF technology |
Non-Patent Citations (2)
| Title |
|---|
| ZLG致远电子: "NXP退出基于i.MX RT 106F本地人脸识别解决方案", pages 1 - 4, Retrieved from the Internet <URL:https://zhuanlan.zhihu.com/p/307982417> * |
| 北京海鹰科技情报研究所: "《海鹰智库丛书 人工智能技术篇》", 哈尔滨：哈尔滨工程大学出版社, pages: 169 - 45 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Hand gesture recognition using depth data | |
| CN108846924A (en) | Vehicle and car door solution lock control method, device and car door system for unlocking | |
| CN109243030B (en) | Control method and system of night non-contact access control system | |
| CN104933344B (en) | Mobile terminal user identity authentication device and method based on multi-biological characteristic mode | |
| CN105700363A (en) | Method and system for waking up smart home equipment voice control device | |
| CN101587485B (en) | Face information automatic login method based on face recognition technology | |
| CN101889928B (en) | Head gesture recognition technology-based wheelchair control method | |
| CN205427946U (en) | Hotel moves in management system | |
| CN109377616A (en) | A kind of access control system based on two-dimension human face identification | |
| CN106372576A (en) | Deep learning-based intelligent indoor intrusion detection method and system | |
| CN103905733A (en) | Method and system for conducting real-time tracking on faces by monocular camera | |
| Liu et al. | Action recognition based on 3d skeleton and rgb frame fusion | |
| CN110414381A (en) | Tracing type face identification system | |
| Zhang et al. | A survey on face anti-spoofing algorithms | |
| WO2021227585A1 (en) | Video image-based multi-modal emotion recognition system and method | |
| CN106887059A (en) | A kind of intelligent electronic lock system based on face recognition | |
| CN109636960A (en) | 3D Intelligent door lock capable of recognizing face and 3D face unlocking method | |
| CN112818722A (en) | Modular dynamically configurable living body face recognition system | |
| Huang et al. | Attendance system based on dynamic face recognition | |
| CN110188715A (en) | A kind of video human face biopsy method of multi frame detection ballot | |
| CN107680229A (en) | Gate control system and its control method based on phonetic feature and recognition of face | |
| Selvaraj et al. | Raspberry Pi based automatic door control system | |
| CN109841034A (en) | A kind of intelligent monitor system based on Internet of Things | |
| CN111401116B (en) | Bimodal emotion recognition method based on enhanced convolution and space-time LSTM network | |
| CN103019381B (en) | Control the method for display screen automatic backlight |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |