CN112412242A - Automatic door control and anti-pinch system based on binocular stereoscopic vision and method thereof - Google Patents
Automatic door control and anti-pinch system based on binocular stereoscopic vision and method thereof Download PDFInfo
- Publication number
- CN112412242A CN112412242A CN202011311588.4A CN202011311588A CN112412242A CN 112412242 A CN112412242 A CN 112412242A CN 202011311588 A CN202011311588 A CN 202011311588A CN 112412242 A CN112412242 A CN 112412242A
- Authority
- CN
- China
- Prior art keywords
- door
- control
- binocular
- automatic door
- detection module
- 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 abstract description 33
- 238000001514 detection method Methods 0.000 claims abstract description 57
- 239000011521 glass Substances 0.000 claims abstract description 44
- 238000005516 engineering process Methods 0.000 claims abstract description 22
- 230000008569 process Effects 0.000 claims abstract description 14
- 238000000605 extraction Methods 0.000 claims abstract description 10
- 238000005259 measurement Methods 0.000 claims abstract description 8
- 238000013528 artificial neural network Methods 0.000 claims abstract description 7
- 230000004888 barrier function Effects 0.000 claims abstract description 7
- 239000003638 chemical reducing agent Substances 0.000 claims description 8
- 230000033001 locomotion Effects 0.000 claims description 8
- 238000005286 illumination Methods 0.000 claims description 5
- 238000001914 filtration Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000013527 convolutional neural network Methods 0.000 claims description 3
- 238000003708 edge detection Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000001133 acceleration Effects 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000006073 displacement reaction Methods 0.000 description 2
- 239000000428 dust Substances 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 230000001360 synchronised effect Effects 0.000 description 2
- 241001465754 Metazoa Species 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000004080 punching Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E05—LOCKS; KEYS; WINDOW OR DOOR FITTINGS; SAFES
- E05F—DEVICES FOR MOVING WINGS INTO OPEN OR CLOSED POSITION; CHECKS FOR WINGS; WING FITTINGS NOT OTHERWISE PROVIDED FOR, CONCERNED WITH THE FUNCTIONING OF THE WING
- E05F15/00—Power-operated mechanisms for wings
- E05F15/70—Power-operated mechanisms for wings with automatic actuation
- E05F15/73—Power-operated mechanisms for wings with automatic actuation responsive to movement or presence of persons or objects
-
- E—FIXED CONSTRUCTIONS
- E05—LOCKS; KEYS; WINDOW OR DOOR FITTINGS; SAFES
- E05F—DEVICES FOR MOVING WINGS INTO OPEN OR CLOSED POSITION; CHECKS FOR WINGS; WING FITTINGS NOT OTHERWISE PROVIDED FOR, CONCERNED WITH THE FUNCTIONING OF THE WING
- E05F15/00—Power-operated mechanisms for wings
- E05F15/40—Safety devices, e.g. detection of obstructions or end positions
- E05F15/42—Detection using safety edges
-
- E—FIXED CONSTRUCTIONS
- E05—LOCKS; KEYS; WINDOW OR DOOR FITTINGS; SAFES
- E05F—DEVICES FOR MOVING WINGS INTO OPEN OR CLOSED POSITION; CHECKS FOR WINGS; WING FITTINGS NOT OTHERWISE PROVIDED FOR, CONCERNED WITH THE FUNCTIONING OF THE WING
- E05F15/00—Power-operated mechanisms for wings
- E05F15/60—Power-operated mechanisms for wings using electrical actuators
- E05F15/603—Power-operated mechanisms for wings using electrical actuators using rotary electromotors
- E05F15/632—Power-operated mechanisms for wings using electrical actuators using rotary electromotors for horizontally-sliding wings
- E05F15/643—Power-operated mechanisms for wings using electrical actuators using rotary electromotors for horizontally-sliding wings operated by flexible elongated pulling elements, e.g. belts, chains or cables
-
- E—FIXED CONSTRUCTIONS
- E05—LOCKS; KEYS; WINDOW OR DOOR FITTINGS; SAFES
- E05F—DEVICES FOR MOVING WINGS INTO OPEN OR CLOSED POSITION; CHECKS FOR WINGS; WING FITTINGS NOT OTHERWISE PROVIDED FOR, CONCERNED WITH THE FUNCTIONING OF THE WING
- E05F17/00—Special devices for shifting a plurality of wings operated simultaneously
-
- E—FIXED CONSTRUCTIONS
- E05—LOCKS; KEYS; WINDOW OR DOOR FITTINGS; SAFES
- E05F—DEVICES FOR MOVING WINGS INTO OPEN OR CLOSED POSITION; CHECKS FOR WINGS; WING FITTINGS NOT OTHERWISE PROVIDED FOR, CONCERNED WITH THE FUNCTIONING OF THE WING
- E05F15/00—Power-operated mechanisms for wings
- E05F15/70—Power-operated mechanisms for wings with automatic actuation
- E05F15/73—Power-operated mechanisms for wings with automatic actuation responsive to movement or presence of persons or objects
- E05F2015/767—Power-operated mechanisms for wings with automatic actuation responsive to movement or presence of persons or objects using cameras
-
- E—FIXED CONSTRUCTIONS
- E05—LOCKS; KEYS; WINDOW OR DOOR FITTINGS; SAFES
- E05F—DEVICES FOR MOVING WINGS INTO OPEN OR CLOSED POSITION; CHECKS FOR WINGS; WING FITTINGS NOT OTHERWISE PROVIDED FOR, CONCERNED WITH THE FUNCTIONING OF THE WING
- E05F17/00—Special devices for shifting a plurality of wings operated simultaneously
- E05F2017/005—Special devices for shifting a plurality of wings operated simultaneously for sliding wings
-
- E—FIXED CONSTRUCTIONS
- E05—LOCKS; KEYS; WINDOW OR DOOR FITTINGS; SAFES
- E05Y—INDEXING SCHEME ASSOCIATED WITH SUBCLASSES E05D AND E05F, RELATING TO CONSTRUCTION ELEMENTS, ELECTRIC CONTROL, POWER SUPPLY, POWER SIGNAL OR TRANSMISSION, USER INTERFACES, MOUNTING OR COUPLING, DETAILS, ACCESSORIES, AUXILIARY OPERATIONS NOT OTHERWISE PROVIDED FOR, APPLICATION THEREOF
- E05Y2900/00—Application of doors, windows, wings or fittings thereof
- E05Y2900/10—Application of doors, windows, wings or fittings thereof for buildings or parts thereof
- E05Y2900/13—Type of wing
- E05Y2900/132—Doors
Landscapes
- Power-Operated Mechanisms For Wings (AREA)
Abstract
The invention discloses an automatic door control and anti-pinch system based on binocular stereo vision and a method thereof, wherein the system comprises a control device, and a power assembly, a human body detection module, a door state detection module and an image acquisition device which are connected with the control device, wherein the image acquisition device acquires image video data of a glass door area, and the physical detection module performs semi-global stereo matching and characteristic extraction on left and right images acquired by a camera by utilizing a neural network so as to detect a human body; the door state processes the collected video and images to obtain barrier detection and automatic door state identification results; the control device is connected with the motor device through a control line; the control device sends out correct instructions according to the detection information of the human body detection module and the door state detection module to control the power assembly to drive the glass door to act so as to realize anti-pinch. The invention adopts the binocular stereo vision technology and has the characteristics of non-contact, passivity, high accuracy and good real-time performance, thereby being very suitable for the speed and distance measurement application in the processes of human body detection and automatic door opening and closing.
Description
Technical Field
The invention relates to an automatic door control system, in particular to an automatic door control and anti-pinch system based on binocular stereoscopic vision and a method thereof.
Background
In modern days, urbanization construction is faster and faster, various high-rise buildings are numerous, along with the development of the building and transportation industry, the automatic door industry also develops rapidly, and the automatic door is widely applied to a plurality of fields, so that the life of people is greatly improved. Automatic doors are not only popular in cities today, but almost every household in rural areas is also beginning to be equipped with automatic doors. The existing rural house gate is generally provided with two doors at the same time, the security door is arranged in the gate, and then the automatic door is arranged on the outer layer of the security door. The living habits in rural areas are different from those in cities, and the doors of rural houses are generally open in the daytime. However, in some areas, due to the climate problem, the door is opened all the time, so that the moisture and dust in the room are much, and therefore, the automatic door is installed to prevent the moisture and dust in the room, and the living environment is greatly improved.
When the existing automatic door is in a non-access-restricted working state, the automatic door adopts a microwave sensor to receive an object moving signal, and when the automatic door moves to about 500mm-1800mm away from the automatic door (the distance can be adjusted), the automatic door starts to be opened. The automatic door working method has some disadvantages. 1) If an animal is moving in the range of the sensor all the time, the sensing door can work, so that the sensing door is always in the working state of the switch, and the sensing system is easy to damage and break down. 2) The detection of the automatic door mainly adopts two technologies of a contact touch panel and an infrared light curtain, but both have certain defects. The automatic door can detect the barrier only when the barrier contacts the safety touch panel and generates certain pressure, the mode has poor precision, and the detection function can be realized only when the door contacts the barrier, thereby belonging to post detection and failure of pre-judgment, so that the accident of people is frequent; the light curtain improves the precision of obstacle detection to a great extent, and makes the door can detect without touching the obstacle, but still can't detect to the translucent object like pure water bucket, simultaneously because the infrared ray covers the regional gapped, causes the part to be in the low narrow degree obstacle in clearance also can't detect. There is therefore a need for further improvements in auto-induction door systems. 3) The automatic system that opens door of current house adopts the object in the sensor response scope mostly, open the door automatically through the controller, still be furnished with the smart card a bit, the card of punching the card gets into during the use, the smart card is lost easily, this kind of automatic system that opens door can only satisfy general needs, nevertheless brought the inconvenience for special crowd, for example old man, child and disabled person, it is inconvenient to pass in and out the door, this kind of system response scope is narrow, there is not identification function, can not satisfy more people's needs, and can bring certain potential safety hazard.
Disclosure of Invention
The invention aims to provide an automatic door control and anti-pinch system based on binocular stereo vision and a method thereof.
The technical scheme adopted by the invention is as follows:
the automatic door control and anti-pinch system based on the binocular stereo vision technology comprises a control device, and a power assembly, a human body detection module, a door state detection module and an image acquisition device which are connected with the control device, wherein the image acquisition device acquires image video data of a glass door area, and the human body detection module performs semi-global stereo matching and feature extraction on left and right images acquired by a camera by utilizing a neural network to detect a human body; the door state processes the collected video and images to obtain barrier detection and automatic door state identification results;
the power assembly comprises a hanger device, a belt device, a brake device and a motor device, wherein the belt device comprises a driving wheel and a driven wheel which are fixed on the upper part of the door frame and are driven by a belt, the driving wheel is driven by the motor device, the upper end of the glass door is fixedly connected with the hanger device, and the hanger device is connected with the belt through a belt fixing device and reciprocates left and right under the driving of the belt; the brake device is arranged at the upper part of the door frame and is correspondingly arranged at a plurality of limit points on the travel path of the hoisting device,
the control device is connected with the motor device through a control line; the control device sends out correct instructions according to the detection information of the human body detection module and the door state detection module to control the power assembly to drive the glass door to act so as to realize anti-pinch. .
Furthermore, as a preferred embodiment, the image acquisition device comprises an upper computer and two cameras connected with the upper computer, the two cameras are symmetrically arranged above the left side and the right side of the glass door, the two cameras are connected with the input port of the upper computer through a data line, and the output port of the upper computer is connected with a control device arranged above the glass door.
Further, as a preferred embodiment, the two cameras are binocular cameras, the cameras adopt standard PAL CVBS signals, the frame frequency is 25 frames/second, and infrared LEDs are arranged inside; the effective detection range of the camera is within the height range of 1.8 +/-0.05 m from the bottom of the glass door.
Further, as a preferred embodiment, the binocular camera comprises a lens, an image sensor, a digital signal processing chip and a PCB board wire, wherein the image sensor senses and records the information of external light and converts the information into current and digital information, and a computer operates and processes the current and digital information to obtain a data file of a digital image; the DSP chip is a core processor and is used for building a video processing system integrating video acquisition, video analysis and signal output; the PCB is a carrier of all components of the camera. The camera passes through the data line and is connected with the input port of host computer, and the output port of host computer position is connected with the controlling means who sets up in glass door top, and controlling means passes through the control line and is connected with motor device.
Specifically, the human body detection module performs semi-global stereo matching, feature extraction and other steps on left and right images acquired by the camera by using a twin false rcnn neural network, so as to realize human body detection.
Furthermore, as a preferred embodiment, the glass door includes two glass doors on the left and right sides, and every glass door is matched with two sets of hanger devices, and the hanger devices of the same set are fixed at the upper end of the corresponding glass door at intervals.
Further, as a preferred embodiment, an image processing chip is arranged in the door state detection module to automatically compensate the illumination of the original video of the image acquisition device, so as to determine a lens with a proper focal length and improve the compatibility of the video in definition and coverage.
Further, as a preferred embodiment, the control device adopts a programmable control element, the control device sends a working instruction to the motor device or the brake device by receiving signals from various external devices, and the motor device or the brake device completes the opening, closing, acceleration and deceleration actions of the automatic door according to the instruction.
The automatic door control and anti-pinch method based on the binocular stereoscopic vision technology adopts the automatic door control and anti-pinch system based on the binocular stereoscopic vision technology, and the method comprises the following steps:
step 1, when a living thing moves to a camera covering area, an upper computer immediately collects images and videos of the living thing in real time through the camera,
step 2, a human body detection module detects whether the human body is a human body; if yes, executing step 3; otherwise, the step is terminated;
step 3, the door state detection module acquires image information of the position where the image acquisition device acquires the person, and analyzes and processes the image information to obtain the accurate position and the motion track of the person;
step 4, judging whether the person is in the coverage area of the camera; if yes, sending a control instruction to the control device; otherwise, executing step 3;
and 5, amplifying the control command signal by the control device, and sending the amplified control command signal to the motor device and the brake device, wherein the motor device starts to work and drives the glass door to open through the speed reducer.
Further, as a preferred embodiment, the specific steps of step 3 are:
step 3-1, acquiring image video acquisition of a binocular camera, performing a self-adaptive median filtering algorithm on the image pair, and performing contrast enhancement on the image pair by using a histogram equalization method;
step 3-2, correcting the binocular image according to the camera parameters, and analyzing a binocular stereoscopic vision measurement model and a geometric transformation relation thereof;
3-3, designing a twin convolutional neural network to perform stereo matching and feature extraction on the binocular images, and finally generating a disparity map of the accurate binocular images;
and 3-4, performing automatic door edge detection by an interframe difference method to obtain edge information of the automatic door.
The image processing system is used for further improving the acquisition link, increasing functions of automatic illumination compensation and the like of an original video, determining a lens with a proper focal length, and improving the compatibility of the video on definition and a coverage range.
By adopting the technical scheme, when a living object moves to the camera covering area, the upper computer immediately collects images and videos of the living object in real time through the camera, detects whether the living object is a human body, if not, the step is terminated, if the living object is the human body, the upper computer collects image information of the position of the human body, analyzes and processes the image information to obtain the accurate position and the motion track of the human body, once the human body is positioned in the camera covering area, the upper computer sends a control instruction to the controller, the controller amplifies a signal of the control instruction and sends the signal to the motor, and the motor starts to work and drives the glass door to be opened through the speed reducer. The video collected by the camera is processed and analyzed to complete the measurement work of the displacement and the speed of the door and the detection of the obstacle in the automatic door opening and closing detection, and finally the anti-pinch function is realized. The control device is a command center of the automatic door, sends corresponding commands through a large-scale integrated block with a command program programmed inside, commands a motor or an electric lock system to work, and simultaneously, people adjust parameters such as door leaf opening speed, opening amplitude and the like through a main controller. The motor device of the invention provides the main power for opening and closing the door and controls the door leaf to operate in an accelerating and decelerating way. The driven wheel device is used for hanging the movable glass door leaf and simultaneously drives the glass door leaf to operate under the power traction. The brake device provides the driving force for opening and closing the door according to the instruction transmitted by the control device, and controls the door leaf to accelerate and decelerate. The belt is used for transmitting the power generated by the motor and dragging the hoisting device. The invention adopts the binocular stereo vision technology and has the characteristics of non-contact, passivity, high accuracy and good real-time performance, thereby being very suitable for the speed and distance measurement application in the processes of face detection and automatic door opening and closing.
Drawings
The invention is described in further detail below with reference to the accompanying drawings and the detailed description;
FIG. 1 is a block diagram of an automatic door control and anti-pinch system based on binocular stereo vision technology according to the present invention;
FIG. 2 is a schematic view of the connection structure of an automatic door control and anti-pinch system based on binocular stereo vision technology;
fig. 3 is a binocular distance measuring schematic diagram of the automatic door control and anti-pinch system based on the binocular stereo vision technology.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions of the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application.
As shown in one of fig. 1 to 3, the invention discloses an automatic door control and anti-pinch system based on a binocular stereo vision technology, which comprises a control device, and a power assembly, a human body detection module, a door state detection module and an image acquisition device which are connected with the control device, wherein the image acquisition device acquires image video data of a glass door area, and the human body detection module performs semi-global stereo matching and feature extraction on left and right images acquired by a camera by using a neural network to detect a human body; the door state processes the collected video and images to obtain barrier detection and automatic door state identification results;
the power assembly comprises a hanger device, a belt device, a brake device and a motor device, wherein the belt device comprises a driving wheel and a driven wheel which are fixed on the upper part of the door frame and are driven by a belt, the driving wheel is driven by the motor device, the upper end of the glass door is fixedly connected with the hanger device, and the hanger device is connected with the belt through a belt fixing device and reciprocates left and right under the driving of the belt; the brake device is arranged at the upper part of the door frame and is correspondingly arranged at a plurality of limit points on the travel path of the hoisting device,
the control device is connected with the motor device through a control line; the control device sends out correct instructions to control the glass door to act according to the detection information of the human body detection module and the door state detection module so as to realize anti-pinch.
The human body detection module acquires left and right images of a target to be detected through a binocular vision system, then performs image preprocessing on the acquired images, including filtering an original image pair, enhancing the images to improve the image quality, highlighting image characteristic details, then performs stereo matching on the binocular images by using a deep neural network to obtain a disparity map, then performs characteristic extraction, and finally obtains a detection result.
Furthermore, as a preferred embodiment, the image acquisition device comprises an upper computer and two cameras connected with the upper computer, the two cameras are symmetrically arranged above the left side and the right side of the glass door, the two cameras are connected with the input port of the upper computer through a data line, and the output port of the upper computer is connected with a control device arranged above the glass door.
Further, as a preferred embodiment, the two cameras are binocular cameras, the cameras adopt standard PAL CVBS signals, the frame frequency is 25 frames/second, and infrared LEDs are arranged inside; the effective detection range of the camera is within the height range of 1.8 +/-0.05 m from the bottom of the glass door.
Further, as a preferred embodiment, the binocular camera comprises a lens, an image sensor, a digital signal processing chip and a PCB board wire, wherein the image sensor senses and records the information of external light and converts the information into current and digital information, and a computer operates and processes the current and digital information to obtain a data file of a digital image; the DSP chip is a core processor and is used for building a video processing system integrating video acquisition, video analysis and signal output; the PCB is a carrier of all components of the camera. The camera passes through the data line and is connected with the input port of host computer, and the output port of host computer position is connected with the controlling means who sets up in glass door top, and controlling means passes through the control line and is connected with motor device.
As shown in fig. 2, the most important is the installation position of the camera, which is installed according to the actual situation, but it is ensured that the automatic glass door and the proper area inside and outside the glass door can be collected by the camera.
Specifically, the human body detection module performs semi-global stereo matching, feature extraction and other steps on left and right images acquired by the camera by using a twin false rcnn neural network, so as to realize human body detection.
Furthermore, as a preferred embodiment, the glass door includes two glass doors on the left and right sides, and every glass door is matched with two sets of hanger devices, and the hanger devices of the same set are fixed at the upper end of the corresponding glass door at intervals.
Further, as a preferred embodiment, an image processing chip is arranged in the door state detection module to automatically compensate the illumination of the original video of the image acquisition device, so as to determine a lens with a proper focal length and improve the compatibility of the video in definition and coverage.
Further, as a preferred embodiment, the control device adopts a programmable control element, the control device sends a working instruction to the motor device or the brake device by receiving signals from various external devices, and the motor device or the brake device completes the opening, closing, acceleration and deceleration actions of the automatic door according to the instruction.
Furthermore, as a preferred embodiment, the control device is a command center of the automatic door, sends out corresponding commands through a large-scale integrated block with a command program programmed inside, commands a motor or an electric lock system to work, and simultaneously, people adjust parameters such as door leaf opening speed and opening amplitude through the main controller.
The braking device consists of a motor and a speed reducer, and works according to a program set by an instruction driving device of the control device to drive the door body to open and close. The motion from the braking device to the door body needs a transmission mechanism in the middle, and the speed reducer is a part of the transmission mechanism. The speed reducer and the motor are connected into a whole to form the speed reducing motor.
The automatic door control and anti-pinch method based on the binocular stereoscopic vision technology adopts the automatic door control and anti-pinch system based on the binocular stereoscopic vision technology, and the method comprises the following steps:
step 1, when a living thing moves to a camera covering area, an upper computer immediately collects images and videos of the living thing in real time through the camera,
step 2, a human body detection module detects whether the human body is a human body; if yes, executing step 3; otherwise, the step is terminated;
step 3, the door state detection module acquires image information of the position where the image acquisition device acquires the person, and analyzes and processes the image information to obtain the accurate position and the motion track of the person;
step 4, judging whether the person is in the coverage area of the camera; if yes, sending a control instruction to the control device; otherwise, executing step 3;
and 5, amplifying the control command signal by the control device, and sending the amplified control command signal to the motor device and the brake device, wherein the motor device starts to work and drives the glass door to open through the speed reducer.
Further, as a preferred embodiment, the specific steps of step 3 are:
step 3-1, acquiring image video acquisition of a binocular camera, performing a self-adaptive median filtering algorithm on the image pair, and performing contrast enhancement on the image pair by using a histogram equalization method;
step 3-2, correcting the binocular image according to the camera parameters, and analyzing a binocular stereoscopic vision measurement model and a geometric transformation relation thereof;
3-3, designing a twin convolutional neural network to perform stereo matching and feature extraction on the binocular images, and finally generating a disparity map of the accurate binocular images;
and 3-4, performing automatic door edge detection by an interframe difference method to obtain edge information of the automatic door.
The image processing system is used for further improving the acquisition link, increasing functions of automatic illumination compensation and the like of an original video, determining a lens with a proper focal length, and improving the compatibility of the video on definition and a coverage range.
The invention adopts the technical scheme that a binocular camera is utilized to collect images and videos, an upper computer immediately collects images and videos of living objects in real time through the camera to detect whether the living objects are human bodies, if not, the steps are terminated, if the living objects are human bodies, image information of positions of the people is collected, the image information is analyzed and processed to obtain accurate positions and movement tracks of the people, once the people are located in a coverage area of the camera, the upper computer sends control instructions to a controller, the controller amplifies signals of the control instructions and sends the signals to a motor, the motor starts to work, the motor drives a glass door to open through a speed reducer, and meanwhile, the running direction, the speed and the rotating speed of the motor are monitored so as to inform the motor to apply force or reduce force at a certain time, enter the speed and the like. The motor does forward motion after obtaining a certain running current, transmits power to the synchronous belt, and transmits the power to the lifting appliance system by the synchronous belt to open the door; after the door body is opened, the controller makes a judgment, if the door body needs to be opened, the motor is informed to move in the positive direction, and the door body is opened. If the door needs to be closed, the motor is informed to do reverse motion, and the door body is closed. Therefore, various operation actions of the automatic door are realized, and meanwhile, videos collected by the camera are processed and analyzed to complete the measurement work of the displacement and the speed of the door and the detection of obstacles in the automatic door switch detection, and finally the anti-pinch function is realized.
The invention adopts the binocular stereo vision technology and has the characteristics of non-contact, passivity, high accuracy and good real-time performance, thereby being very suitable for the speed and distance measurement application in the processes of face detection and automatic door opening and closing.
It is to be understood that the embodiments described are only a few embodiments of the present application and not all embodiments. The embodiments and features of the embodiments in the present application may be combined with each other without conflict. The components of the embodiments of the present application, generally described and illustrated in the figures herein, can be arranged and designed in a wide variety of different configurations. Thus, the detailed description of the embodiments of the present application is not intended to limit the scope of the claimed application, but is merely representative of selected embodiments of the application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Claims (9)
1. Automatic door control and prevent pressing from both sides system based on binocular stereoscopic vision technique, its characterized in that: the system comprises a control device, and a power assembly, a human body detection module, a door state detection module and an image acquisition device which are connected with the control device, wherein the image acquisition device acquires image video data of a glass door area, and the human body detection module performs semi-global stereo matching and feature extraction on left and right images acquired by a camera by using a neural network to detect a human body; the door state processes the collected video and images to obtain barrier detection and automatic door state identification results;
the power assembly comprises a hanger device, a belt device, a brake device and a motor device, wherein the belt device comprises a driving wheel and a driven wheel which are fixed on the upper part of the door frame and are driven by a belt, the driving wheel is driven by the motor device, the upper end of the glass door is fixedly connected with the hanger device, and the hanger device is connected with the belt through a belt fixing device and reciprocates left and right under the driving of the belt; the brake device is arranged at the upper part of the door frame and is correspondingly arranged at a plurality of limit points on the travel path of the hoisting device,
the control device is connected with the motor device through a control line; the control device sends out correct instructions according to the detection information of the human body detection module and the door state detection module to control the power assembly to drive the glass door to act so as to realize anti-pinch.
2. The binocular stereo vision technology-based automatic door control and anti-pinch system of claim 1, wherein: the image acquisition device comprises an upper computer and two cameras connected with the upper computer, the two cameras are symmetrically arranged above the left side and the right side of the glass door, the two cameras are connected with an input port of the upper computer through a data line, and an output port of the upper computer is connected with a control device arranged above the glass door.
3. The binocular stereo vision technology-based automatic door control and anti-pinch system of claim 2, wherein: the two cameras are binocular cameras, adopt standard PAL CVBS signals, have a frame frequency of 25 frames/second and are internally provided with infrared LEDs; the effective detection range of the camera is within the height range of 1.8 +/-0.05 m from the bottom of the glass door.
4. The binocular stereo vision technology-based automatic door control and anti-pinch system of claim 3, wherein: the binocular camera comprises a lens, an image sensor, a digital signal processing chip and a PCB (printed circuit board) wire rod, wherein the image sensor senses and records information of external light and converts the information into current and digital information, and a computer operates and processes the current and digital information to obtain a data file of a digital image; the DSP chip is a core processor and is used for building a video processing system integrating video acquisition, video analysis and signal output; the PCB is a carrier of all components of the camera.
5. The binocular stereo vision technology-based automatic door control and anti-pinch system of claim 1, wherein: the glass door comprises a left glass door and a right glass door, each glass door is matched with two groups of hanger devices, and the hanger devices in the same group are fixed at the upper ends of the corresponding glass doors at intervals.
6. The binocular stereo vision technology-based automatic door control and anti-pinch system of claim 1, wherein: an image processing chip is arranged in the door state detection module to automatically compensate the illumination of the original video of the image acquisition device so as to determine a lens with a proper focal length.
7. The binocular stereo vision technology-based automatic door control and anti-pinch system of claim 1, wherein: the control device adopts a programmable control element, the control device sends a working instruction to the motor device or the brake device by receiving signals from various external equipment, and the motor device or the brake device completes the actions of opening, closing, accelerating and decelerating of the automatic door according to the instruction.
8. The binocular stereoscopic vision technology-based automatic door control and anti-pinch method adopts the binocular stereoscopic vision technology-based automatic door control and anti-pinch system of claims 1 to 7, and is characterized in that: the method comprises the following steps:
step 1, when a living thing moves to a camera covering area, an upper computer immediately collects images and videos of the living thing in real time through the camera,
step 2, a human body detection module detects whether the human body is a human body; if yes, executing step 3; otherwise, the step is terminated;
step 3, the door state detection module acquires image information of the position where the image acquisition device acquires the person, and analyzes and processes the image information to obtain the accurate position and the motion track of the person;
step 4, judging whether the person is in the coverage area of the camera; if yes, sending a control instruction to the control device; otherwise, executing step 3;
and 5, amplifying the control command signal by the control device, and sending the amplified control command signal to the motor device and the brake device, wherein the motor device starts to work and drives the glass door to open through the speed reducer.
9. The binocular stereo vision technology-based automatic door control and anti-pinch system of claim 8, wherein: the specific steps of the step 3 are as follows:
step 3-1, acquiring image video acquisition of a binocular camera, performing a self-adaptive median filtering algorithm on the image pair, and performing contrast enhancement on the image pair by using a histogram equalization method;
step 3-2, correcting the binocular image according to the camera parameters, and analyzing a binocular stereoscopic vision measurement model and a geometric transformation relation thereof;
3-3, designing a twin convolutional neural network to perform stereo matching and feature extraction on the binocular images, and finally generating a disparity map of the accurate binocular images;
and 3-4, performing automatic door edge detection by an interframe difference method to obtain edge information of the automatic door.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011311588.4A CN112412242A (en) | 2020-11-20 | 2020-11-20 | Automatic door control and anti-pinch system based on binocular stereoscopic vision and method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011311588.4A CN112412242A (en) | 2020-11-20 | 2020-11-20 | Automatic door control and anti-pinch system based on binocular stereoscopic vision and method thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112412242A true CN112412242A (en) | 2021-02-26 |
Family
ID=74778290
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011311588.4A Pending CN112412242A (en) | 2020-11-20 | 2020-11-20 | Automatic door control and anti-pinch system based on binocular stereoscopic vision and method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112412242A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113585913A (en) * | 2021-07-28 | 2021-11-02 | 广东蓝水花智能电子有限公司 | Automatic door and method for setting triggering area of automatic door |
| CN113585912A (en) * | 2021-07-28 | 2021-11-02 | 广东蓝水花智能电子有限公司 | Automatic door control method based on image processing technology and automatic door |
| CN117555258A (en) * | 2023-11-14 | 2024-02-13 | 济南巨光风田门业有限公司 | Electric upper door folding intelligent control management system based on artificial intelligence |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE19522760A1 (en) * | 1995-06-27 | 1997-04-10 | Dorma Gmbh & Co Kg | Automatic door operating system |
| CN101975009A (en) * | 2010-10-29 | 2011-02-16 | 无锡中星微电子有限公司 | Automatic door control device and method |
| CN104234575A (en) * | 2014-09-21 | 2014-12-24 | 王俭俭 | Control system and control method for automatic door |
| CN106401367A (en) * | 2016-12-09 | 2017-02-15 | 贵州大学 | Automatic induction door based on image recognition and control method of automatic induction door |
| JP2017061830A (en) * | 2015-09-25 | 2017-03-30 | 寺岡オート・ドアシステム株式会社 | Automatic door system |
| CN106910222A (en) * | 2017-02-15 | 2017-06-30 | 中国科学院半导体研究所 | Face three-dimensional rebuilding method based on binocular stereo vision |
| CN209924740U (en) * | 2019-01-31 | 2020-01-10 | 东莞市桥特电子科技有限公司 | Image processing anti-clamping device for automatic door |
| CN110939351A (en) * | 2019-10-28 | 2020-03-31 | 优创嘉(大连)科技有限公司 | Visual intelligent control method and visual intelligent control door |
-
2020
- 2020-11-20 CN CN202011311588.4A patent/CN112412242A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| DE19522760A1 (en) * | 1995-06-27 | 1997-04-10 | Dorma Gmbh & Co Kg | Automatic door operating system |
| CN101975009A (en) * | 2010-10-29 | 2011-02-16 | 无锡中星微电子有限公司 | Automatic door control device and method |
| CN104234575A (en) * | 2014-09-21 | 2014-12-24 | 王俭俭 | Control system and control method for automatic door |
| JP2017061830A (en) * | 2015-09-25 | 2017-03-30 | 寺岡オート・ドアシステム株式会社 | Automatic door system |
| CN106401367A (en) * | 2016-12-09 | 2017-02-15 | 贵州大学 | Automatic induction door based on image recognition and control method of automatic induction door |
| CN106910222A (en) * | 2017-02-15 | 2017-06-30 | 中国科学院半导体研究所 | Face three-dimensional rebuilding method based on binocular stereo vision |
| CN209924740U (en) * | 2019-01-31 | 2020-01-10 | 东莞市桥特电子科技有限公司 | Image processing anti-clamping device for automatic door |
| CN110939351A (en) * | 2019-10-28 | 2020-03-31 | 优创嘉(大连)科技有限公司 | Visual intelligent control method and visual intelligent control door |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113585913A (en) * | 2021-07-28 | 2021-11-02 | 广东蓝水花智能电子有限公司 | Automatic door and method for setting triggering area of automatic door |
| CN113585912A (en) * | 2021-07-28 | 2021-11-02 | 广东蓝水花智能电子有限公司 | Automatic door control method based on image processing technology and automatic door |
| CN113585912B (en) * | 2021-07-28 | 2022-09-09 | 广东蓝水花智能电子有限公司 | Automatic door control method based on image processing technology and automatic door |
| CN113585913B (en) * | 2021-07-28 | 2022-09-09 | 广东蓝水花智能电子有限公司 | Automatic door and method for setting triggering area of automatic door |
| CN117555258A (en) * | 2023-11-14 | 2024-02-13 | 济南巨光风田门业有限公司 | Electric upper door folding intelligent control management system based on artificial intelligence |
| CN117555258B (en) * | 2023-11-14 | 2024-04-26 | 济南巨光风田门业有限公司 | Electric upper door folding intelligent control management system based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112412242A (en) | Automatic door control and anti-pinch system based on binocular stereoscopic vision and method thereof | |
| CN210735884U (en) | Robot and elevator interaction system | |
| CN109878552B (en) | Foreign matter monitoring devices between track traffic platform door and train based on machine vision | |
| CN102036899B (en) | Video-based system and method of elevator door detection | |
| JP2005538278A (en) | Stereo door sensor | |
| CN105744232A (en) | Method for preventing power transmission line from being externally broken through video based on behaviour analysis technology | |
| CN107981790A (en) | Room area division methods and sweeping robot | |
| CN105867390A (en) | Method for controlling transport robot to autonomously enter elevator | |
| CA2628204A1 (en) | Automatic door assembly with video imaging device | |
| CN103696652A (en) | Electric folding door and protecting method and system thereof | |
| US11443517B1 (en) | Video-based parking garage monitoring | |
| CN203237899U (en) | Elevator door safety protection system and elevator | |
| KR20170052285A (en) | Intelligent camera and System for automatic opening door using Intelligent camera | |
| KR101066562B1 (en) | Screen door system using stereo image | |
| CN103661491B (en) | Platform shield door intelligent control system and method | |
| CN113114969B (en) | Intelligent video intercom monitoring system and intercom monitoring method for intelligent building based on big data | |
| KR102302767B1 (en) | System for Detecting Subway Entrance Door based on Image Deep Learning Reading | |
| CN117253203A (en) | Obstacle detecting system based on visual sensor | |
| CN111390918B (en) | Active control system of household intelligent robot | |
| CN115258862A (en) | Deep learning-based two-wheel vehicle elevator access prohibiting system and method | |
| CN208833926U (en) | A kind of shuttle operating area laser array safety detecting system | |
| CN113911885A (en) | Elevator anti-pinch method and system based on image processing | |
| CN211225909U (en) | Floor character superposition device based on intelligent perception technology | |
| EP2112318A2 (en) | Security system for an automation unit of motorised gates, doors or barriers | |
| CN208285448U (en) | Door gear based on the identification of image intelligent boarding intention |
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 |