CN209388464U - A kind of mechanical garage personnel are strayed into monitoring identification early warning system - Google Patents
A kind of mechanical garage personnel are strayed into monitoring identification early warning system Download PDFInfo
- Publication number
- CN209388464U CN209388464U CN201920116280.0U CN201920116280U CN209388464U CN 209388464 U CN209388464 U CN 209388464U CN 201920116280 U CN201920116280 U CN 201920116280U CN 209388464 U CN209388464 U CN 209388464U
- Authority
- CN
- China
- Prior art keywords
- personnel
- strayed
- video camera
- early warning
- line
- 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.)
- Active
Links
Landscapes
- Alarm Systems (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The utility model discloses a kind of mechanical garage personnel to be strayed into monitoring identification early warning system, including video camera, yellow demarcartion line, RS485 communication bus and at least one combined aural and visual alarm identified for monitored by personnel, yellow demarcartion line is parallel with the boundary line of parking stall inlet, and the two spacing is 1m, the straight line of the boundary line and parking stall two sides of yellow demarcartion line and parking stall inlet constitutes personnel and is strayed into monitoring region;Video camera is connect by RS485 bus with combined aural and visual alarm, and video camera has been internally integrated controller, and video camera can shoot entire personnel and be strayed into monitoring region.Alarm signal is issued when personnel will enter garage; pass through RS485 bus starting combined aural and visual alarm; sound-light alarm and voice prompting are carried out to the personnel that will enter garage; realized with self safeguard protection for the personnel of improving; personnel are avoided to be strayed into the generation of garage accident from the root, the essential safety for improving mechanical garage is horizontal.
Description
Technical field
The utility model belongs to mechanical parking equipment technical field, and in particular to a kind of mechanical garage personnel are strayed into prison
Survey identification early warning system.
Background technique
As what mechanical parking equipment used popularizes, associated accident is also increasing, and accident occurs in order to prevent,
Provided with some safeguard procedures.Although someone's vehicle is strayed into the regulation of detection device in mechanical parking equipment standard, substantially
Type, including infrared or laser scan type are blocked for light beam, there are problems that blind area or out-of-service time section, i.e., in laser or infrared beam
Scan less than there are check frequencies in region, even if there is personnel to be strayed into the region, detection device can not also detect that personnel are strayed into
Behavior, and increasing scanning light beam quantity can be such that detection device cost obviously increases, and cannot be fully solved light beam scan blind spot
Problem;Second is that if personnel are to be strayed into mechanical parking equipment at vehicles while passing workspace, laser or infrared beam quilt at this time
Automobile blocks, detection device failure, can not also detect that the personnel in automobile disengaging this period of mechanical parking equipment miss
Enter behavior, there are problems that the out-of-service time section, this be also in December, 2,017 one woman of Jiangsu Prov. People's Hospital stereo garage bow
It sees the mobile phone and is strayed into one of the main reason for being wounded accident, which is strayed into garage when vehicle is outputed, at this time garage roller shutter
Laser or infrared detection system at door are not acted upon by occlusion, are failed the woman of detection identification in time and are strayed into garage behavior simultaneously
Stop garage, with arriving negative one layer under the lifting platform of normal pick-up state, after setting by the vehicle of automatic running and carrying vehicle
Standby " wounding ".
Summary of the invention
The purpose of this utility model is to provide a kind of mechanical garage personnel to be strayed into monitoring identification early warning system, solves existing
Some mechanical garages block the anti-people of type using light beam and are strayed into detection device, there is technical issues that blind area and out-of-service time.
In order to solve the above-mentioned technical problem the utility model, adopts the following technical scheme that
A kind of mechanical garage personnel are strayed into monitoring identification early warning system, including identified for monitored by personnel video camera,
Yellow demarcartion line, RS485 communication bus and at least one combined aural and visual alarm, the yellow demarcartion line are arranged in front of parking stall
On ground, yellow demarcartion line is parallel with the boundary line of parking stall inlet, and the two spacing is 1m, yellow demarcartion line and parking stall
The boundary line of inlet and the straight line of parking stall two sides constitute personnel and are strayed into monitoring region;
The video camera is connect by RS485 bus with combined aural and visual alarm, and video camera has been internally integrated controller, video camera
Entire personnel can be shot and be strayed into monitoring region.
It being further improved, the both ends that the personnel are strayed into monitoring region are each provided with a monitored by personnel and identify video camera,
Two video cameras pass through RS485 bus and connect with combined aural and visual alarm.
It is further improved, there is the mechanical garage on multiple parking stalls for being arranged side by side, be strayed into monitoring region along personnel
Length direction is arranged a personnel every 30m and is strayed into monitoring identification early warning submodule, and each personnel are strayed into monitoring identification early warning
Module includes two video cameras and a combined aural and visual alarm, and two video cameras pass through RS485 bus and connect with combined aural and visual alarm,
It is mutually indepedent that each personnel are strayed into monitoring identification early warning submodule.
It is further improved, the video camera has been internally integrated FPGA controller, the model YS- of the combined aural and visual alarm
01X highlights LED strobe light emission using the big decibel alarming horn of 120DB, may customize alarm voice, can be controlled by RS485 communication
Agreement, the MT9V034 of the model Aptina company of video camera, using cmos image sensor, exporting image full-size is
752*480pixels, maximum frame per second are 60fps, and video camera is connect by FMC interface with FPGA controller, FPGA model
EP2C70F672C8, FPGA controller include CMOS video acquisition module, control module, Nios II processor, Avalon Bus
Bus, human body target detection module, detection algorithm interface, dma controller, ROM, parallel storage, SDRAM storage control in piece
Outer SDRAM, RS485 interface of module, piece and HDMI interface.
It is strayed into the detection method of monitoring identification early warning system based on above-mentioned mechanical garage personnel, includes the following steps:
1), system electrification, under the Nios II processor control of FPGA controller, the initialization of human body target detection module,
CMOS video acquisition and control module automatically configure the video camera being attached thereto, and monitored by personnel identifies video camera in real time to shooting people
Member is strayed into monitoring region and is shot, and sends the picture of shooting to FPGA controller;
2), CMOS video acquisition module is successively acquired each frame picture sent from video camera by 640*480pixels size
Data, while it being sent to the read port of parallel storage and dma controller inside FPGA, when being transferred to FPGA internal storage
Automatic gray processing, and stored by particular order;In addition while by color image information it is transferred to dma controller and in its control
Under be transferred to SDRAM storage;
3), after FPGA internal storage deposit picture, human body target detection module knows algorithm to camera shooting using human body target
Machine acquisition picture detected, from picture memory take out corresponding position data calculate detection window HOG feature to
Amount, recalls SVM classifier stored in ROM and classifies to feature vector, realize the human testing of sliding window;
4) after, detecting human body target, Nios II processor issues alarm signal and gives RS485 interface, total by RS485
Line starts combined aural and visual alarm, carries out acousto-optic warning to the personnel that will enter garage.
It is further improved, for convenience of system debug, can also configure display, display is controlled by HDMI interface and FPGA
Device connection, FPGA controller, which will test result and be output to display by HDMI interface, to be shown.Human body target detection module
The dimensional information of human body target window and location information are sent to Nios II processor, Nios II processor is in memory
Draw corresponding box, output display result.
It is further improved, in the step 2), using histograms of oriented gradients (Histograms of Oriented
Gradients, HOG) feature combine linear support vector machine classifier (Support Vector Machine, SVM) algorithm pair
Human body target carries out detection identification, and the HOG feature for extracting current video frame includes the following steps:
2.1) detection window, is set, greyscale transform process is carried out to the video frame of acquisition;
2.2) normalization of color space, is carried out to input picture using Gramma correction method;
2.3) gradient of each pixel of image, is calculated;
2.4), creating unit lattice construct gradient orientation histogram for each cell;
2.5), blocking (p >=2) p*p cell are combined, the histograms of oriented gradients (HOG) of block is normalized
Processing, the influence shone with weakened light;
2.6) it, collects all pieces in detection window of HOG feature and forms the HOG feature vector for indicating the video frame, supply
Classification uses;
2.7) it, is instructed using the INRIA data set comprising different types of human body target picture as support vector machines (SVM)
Practice the database of study, extracts the HOG feature and corresponding label (+1 or -1) of the positive negative sample of database, be input to supporting vector
It is trained study in machine, obtains the classifier based on human body target detection identification: using MATLAB's when specific implementation
Svmtrain function in the tool box libsvm is trained the database of INRIA, obtains the classification factor vector of classifier
And classification thresholds;
2.8) Classification and Identification, is carried out using feature vector of the classifier to current video frame, determines whether that someone is strayed into
Monitor region.
Compared with prior art, the beneficial effects of the utility model are as follows:
Mechanical garage personnel are strayed into monitoring identification early warning system, and the personnel of being specific to are strayed into this dangerous movement of garage
And design it is a set of can be to the system for being strayed into garage behavior and making acousto-optic warning.The system does not influence the original sound and light signal in garage,
When the personnel of having detected, which enter yellow demarcartion line region, to be strayed into garage, monitored by personnel identifies that video camera can be with automatic identification
It takes pictures and starts acousto-optic warning, cause will to be strayed into garage personnel's note that it is made to fall back on yellow line with exterior domain, it is possible to reduce no
Necessary personal injury, unnecessary safety accident caused by avoiding because of human factor, while garage administrative staff can be assisted complete
The management cost using unit has effectively been saved in the daily nurse management of forming apparatus.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of the utility model patent.
Fig. 2 is controller module block diagram.
Specific embodiment
To keep the purpose of this utility model and technical solution clearer, below in conjunction with the utility model embodiment to this
The technical solution of utility model is clearly and completely described.
Embodiment one:
As shown in Figure 1, 2, a kind of mechanical garage personnel are strayed into monitoring identification early warning system, including know for monitored by personnel
Other video camera 1, yellow demarcartion line 2,4, two combined aural and visual alarms 3 of RS485 communication bus and a display, the yellow are alert
It guards against line 2 to be arranged on the ground in front of parking stall, yellow demarcartion line 2 is parallel with the boundary line of parking stall inlet, and between the two
Personnel, which are constituted, away from the straight line for 1m, the boundary line and parking stall two sides of yellow demarcartion line 2 and parking stall inlet is strayed into monitoring section
Domain 5;Two video cameras 1 are connect by RS485 bus 4 with combined aural and visual alarm 3.Video camera has been internally integrated controller, camera shooting
Machine 1 can shoot entire personnel and be strayed into monitoring region 5.
In the present embodiment, the video camera has been internally integrated FPGA controller, the model of the combined aural and visual alarm 3
YS-01X highlights LED strobe light emission using the big decibel alarming horn of 120DB, may customize alarm voice, and it is logical can be controlled by RS485
Agreement is interrogated, the MT9V034 of the model Aptina company of video camera 1 exports image full-size using cmos image sensor
For 752*480pixels, maximum frame per second is 60fps, and video camera 1 is connect by FMC interface with FPGA controller, FPGA model
EP2C70F672C8, FPGA controller include CMOS video acquisition module, control module, Nios II processor, Avalon Bus
Bus, human body target detection module, detection algorithm interface, dma controller, ROM, parallel storage, SDRAM storage control in piece
Outer SDRAM, RS485 interface of module, piece and HDMI interface.
The personnel for entering yellow demarcartion line region (5) are carried out certainly by being embedded in human body image recognizer in video camera
Dynamic detection identification, issues alarm signal when personnel will enter garage, right by RS485 bus starting combined aural and visual alarm (3)
The personnel that garage will be entered carry out sound-light alarm and voice prompting, are realized with self safeguard protection for the personnel of improving, from root
On avoid personnel from being strayed into the generation of garage accident, the essential safety for improving mechanical garage is horizontal.
Embodiment two:
In the present embodiment, there is the mechanical garage on multiple parking stalls for being arranged side by side, be strayed into monitoring region 5 along personnel
Length direction every 30m be arranged a personnel be strayed into monitoring identification early warning submodule, each personnel be strayed into monitoring identification early warning
Submodule includes two video cameras 1 and a combined aural and visual alarm 3, and two video cameras 1 pass through RS485 bus 4 and sound-light alarm
Device 3 connects, and it is mutually indepedent that each personnel are strayed into monitoring identification early warning submodule.
Other parts are identical with embodiment one.
Embodiment three:
It is strayed into the detection method of monitoring identification early warning system based on mechanical garage personnel, includes the following steps:
1), system electrification, under the Nios II processor control of FPGA controller, the initialization of human body target detection module,
CMOS video acquisition and control module automatically configure the video camera being attached thereto, and monitored by personnel identifies video camera in real time to shooting people
Member is strayed into monitoring region and is shot, and sends the picture of shooting to FPGA controller;
2), CMOS video acquisition module is successively acquired each frame picture sent from video camera by 640*480pixels size
Data, while it being sent to the read port of parallel storage and dma controller inside FPGA, when being transferred to FPGA internal storage
Automatic gray processing, and stored by particular order;In addition while by color image information it is transferred to dma controller and in its control
Under be transferred to SDRAM storage;
3), after FPGA internal storage deposit picture, human body target detection module knows algorithm to camera shooting using human body target
Machine acquisition picture detected, from picture memory take out corresponding position data calculate detection window HOG feature to
Amount, recalls SVM classifier stored in ROM and classifies to feature vector, realize the human testing of sliding window;
4) after, detecting human body target, Nios II processor issues alarm signal and gives RS485 interface, total by RS485
Line starts combined aural and visual alarm, carries out acousto-optic warning to the personnel that will enter garage.For convenience of system debug, display can also configure
Device, display are connect by HDMI interface with FPGA controller, and FPGA controller will test result and is output to by HDMI interface
Display is shown.The dimensional information of human body target window and location information are sent to Nios by human body target detection module
II processor, Nios II processor draw corresponding box, output display result in memory.
In the present embodiment, in the step 2), using histograms of oriented gradients (HOG) feature combination linear support vector
Machine classifier (SVM) algorithm carries out detection identification to human body target, and the HOG feature for extracting current video frame includes the following steps:
2.1) detection window, is set, greyscale transform process is carried out to the video frame of acquisition;
2.2) normalization of color space, is carried out to input picture using Gramma correction method;
2.3) gradient of each pixel of image, is calculated;
2.4), creating unit lattice construct gradient orientation histogram for each cell;
2.5), blocking (p >=2) p*p cell are combined, the histograms of oriented gradients (HOG) of block is normalized
Processing, the influence shone with weakened light;
2.6) it, collects all pieces in detection window of HOG feature and forms the HOG feature vector for indicating the video frame, supply
Classification uses;
2.7) it, is instructed using the INRIA data set comprising different types of human body target picture as support vector machines (SVM)
Practice the database of study, extracts the HOG feature and corresponding label (+1 or -1) of the positive negative sample of database, be input to supporting vector
It is trained study in machine, obtains the classifier based on human body target detection identification: using MATLAB's when specific implementation
Svmtrain function in the tool box libsvm is trained the database of INRIA, obtains the classification factor vector of classifier
And classification thresholds;
2.8) Classification and Identification, is carried out using feature vector of the classifier to current video frame, determines whether that someone is strayed into
Monitor region.
Do not done in the utility model illustrate be the prior art or can be realized by the prior art, Er Qieben
Specific implementation case described in utility model is only the preferable case study on implementation of the utility model, is not used to limit practical
Novel practical range.Equivalent changes and modifications made by i.e. all contents according to present utility model application the scope of the patents, should all make
For the technology scope of the utility model.
Claims (5)
1. a kind of mechanical garage personnel are strayed into monitoring identification early warning system, which is characterized in that including being identified for monitored by personnel
Video camera (1), yellow demarcartion line (2), RS485 communication bus (4) and at least one combined aural and visual alarm (3), the yellow is alert
It guards against line (2) to be arranged on the ground in front of parking stall, yellow demarcartion line (2) is parallel with the boundary line of parking stall inlet, and two
Person's spacing is 1m, and yellow demarcartion line (2) constitutes personnel with the boundary line of parking stall inlet and the straight line of parking stall two sides and is strayed into
It monitors region (5);
The video camera (1) is connect by RS485 bus (4) with combined aural and visual alarm (3), and video camera has been internally integrated controller,
Video camera (1) can shoot entire personnel and be strayed into monitoring region (5).
2. mechanical garage personnel according to claim 1 are strayed into monitoring identification early warning system, which is characterized in that the people
The both ends that member is strayed into monitoring region (5) are each provided with a video camera (1), two video cameras (1) by RS485 bus (4) with
Combined aural and visual alarm (3) connection.
3. mechanical garage personnel according to claim 1 are strayed into monitoring identification early warning system, which is characterized in that for simultaneously
Row is provided with the mechanical garage on multiple parking stalls, and the length direction for being strayed into monitoring region (5) along personnel is arranged one every 30m
Personnel be strayed into monitoring identification early warning submodule, each personnel be strayed into monitoring identification early warning submodule include two video cameras (1) and
One combined aural and visual alarm (3), two video cameras (1) are connect by RS485 bus (4) with combined aural and visual alarm (3), each personnel
It is mutually indepedent to be strayed into monitoring identification early warning submodule.
4. mechanical garage personnel according to any one of claim 1-3 are strayed into monitoring identification early warning system, feature
It is, the video camera has been internally integrated FPGA controller, and the model YS-01X of the combined aural and visual alarm (3) is controlled by
RS485 communications protocol, the MT9V034 of the model Aptina company of video camera (1), using cmos image sensor, output figure
Picture full-size is 752*480pixels, and maximum frame per second is 60fps, and video camera (1) passes through FMC interface and FPGA controller company
It connects, FPGA model EP2C70F672C8, FPGA controller includes CMOS video acquisition module, control module, Nios II processing
Device, Avalon Bus bus, human body target detection module, detection algorithm interface, dma controller, ROM, parallel storage in piece,
Outer SDRAM, RS485 interface of SDRAM storage control module, piece and HDMI interface.
5. mechanical garage personnel according to claim 4 are strayed into monitoring identification early warning system, which is characterized in that for convenience
System debug can also configure display, and display is connect by HDMI interface with FPGA controller, and FPGA controller will test knot
Fruit is output to display by HDMI interface and is shown.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201920116280.0U CN209388464U (en) | 2019-01-23 | 2019-01-23 | A kind of mechanical garage personnel are strayed into monitoring identification early warning system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201920116280.0U CN209388464U (en) | 2019-01-23 | 2019-01-23 | A kind of mechanical garage personnel are strayed into monitoring identification early warning system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN209388464U true CN209388464U (en) | 2019-09-13 |
Family
ID=67853870
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201920116280.0U Active CN209388464U (en) | 2019-01-23 | 2019-01-23 | A kind of mechanical garage personnel are strayed into monitoring identification early warning system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN209388464U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109727426A (en) * | 2019-01-23 | 2019-05-07 | 南京市特种设备安全监督检验研究院 | A kind of mechanical garage personnel are strayed into monitoring identification early warning system and detection method |
-
2019
- 2019-01-23 CN CN201920116280.0U patent/CN209388464U/en active Active
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109727426A (en) * | 2019-01-23 | 2019-05-07 | 南京市特种设备安全监督检验研究院 | A kind of mechanical garage personnel are strayed into monitoring identification early warning system and detection method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103839373B (en) | A kind of unexpected abnormality event Intelligent Recognition alarm device and warning system | |
CN109727426A (en) | A kind of mechanical garage personnel are strayed into monitoring identification early warning system and detection method | |
CN101593425B (en) | Machine vision based fatigue driving monitoring method and system | |
CN104155959B (en) | Transformer station's unattended integral monitoring system | |
CN104847211B (en) | Auxiliary system for safety of platform safety doors and train door sections and implementation method thereof | |
CN103714631B (en) | ATM cash dispenser intelligent monitor system based on recognition of face | |
CN109635631A (en) | A kind of fire watch room personnel recognition methods on duty based on artificial intelligence | |
CN109657555A (en) | It is a kind of to get off the detection device and method of omission for school bus student | |
CN104499864A (en) | Anti-collision anti-jamming bus safety door system | |
CN108447194A (en) | A kind of ATM protective cabins intelligent protection system and method | |
CN1967564A (en) | Method and device for detecting and identifying human face applied to set environment | |
CN103400430A (en) | Metro door safety monitoring system based on video light curtain | |
CN106022235A (en) | Missing child detection method based on human body detection | |
CN109190600A (en) | A kind of driver's monitoring system of view-based access control model sensor | |
CN205540806U (en) | Three -dimensional facial image recognition device | |
CN209388464U (en) | A kind of mechanical garage personnel are strayed into monitoring identification early warning system | |
CN102445453A (en) | Automatic detection device and identification method for integrality of guardrail of high-speed railway line | |
CN101777223B (en) | Financial self-service terminal and control method of safety zone thereof | |
CN108960131A (en) | The anti-people of mechanical garage is strayed into detection method | |
Batchu et al. | Driver drowsiness detection to reduce the major road accidents in automotive vehicles | |
CN104658105B (en) | Intelligent video analysis multi-person early-warning device and method | |
KR20220113631A (en) | Dangerous situation detection device and dangerous situation detection method | |
CN206363421U (en) | Prison intelligent card port management system based on face recognition technology | |
CN107742103A (en) | A kind of video frequency monitoring method and system | |
CN206400760U (en) | A kind of intelligent vehicle safety check clearance system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
GR01 | Patent grant | ||
GR01 | Patent grant |