CN109410497A - A kind of monitoring of bridge opening space safety and alarm system based on deep learning - Google Patents
A kind of monitoring of bridge opening space safety and alarm system based on deep learning Download PDFInfo
- Publication number
- CN109410497A CN109410497A CN201811383003.2A CN201811383003A CN109410497A CN 109410497 A CN109410497 A CN 109410497A CN 201811383003 A CN201811383003 A CN 201811383003A CN 109410497 A CN109410497 A CN 109410497A
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- bridge opening
- pyrotechnics
- deep learning
- opening space
- 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.)
- Granted
Links
- 238000013135 deep learning Methods 0.000 title claims abstract description 44
- 238000012544 monitoring process Methods 0.000 title claims abstract description 29
- 238000001514 detection method Methods 0.000 claims abstract description 60
- 230000009545 invasion Effects 0.000 claims abstract description 26
- 238000000034 method Methods 0.000 claims abstract description 10
- 238000007726 management method Methods 0.000 claims description 28
- 238000012549 training Methods 0.000 claims description 25
- 238000013523 data management Methods 0.000 claims description 8
- 238000004458 analytical method Methods 0.000 claims description 7
- 230000006399 behavior Effects 0.000 claims description 6
- 238000005516 engineering process Methods 0.000 claims description 6
- 238000003062 neural network model Methods 0.000 claims description 6
- 230000002159 abnormal effect Effects 0.000 claims description 5
- 230000001537 neural effect Effects 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 4
- 235000019504 cigarettes Nutrition 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 2
- 238000012545 processing Methods 0.000 abstract description 4
- 230000008901 benefit Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000013527 convolutional neural network Methods 0.000 description 2
- 230000007613 environmental effect Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000011065 in-situ storage Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 230000004308 accommodation Effects 0.000 description 1
- 230000004913 activation Effects 0.000 description 1
- 230000002776 aggregation Effects 0.000 description 1
- 238000004220 aggregation Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 230000000977 initiatory effect Effects 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B13/00—Burglar, theft or intruder alarms
- G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
- G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
- G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
- G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
- G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
- G08B13/19613—Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion
- G08B13/19615—Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion wherein said pattern is defined by the user
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- 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
-
- 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
-
- 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
- G06V40/20—Movements or behaviour, e.g. gesture recognition
- G06V40/23—Recognition of whole body movements, e.g. for sport training
- G06V40/25—Recognition of walking or running movements, e.g. gait recognition
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/12—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions
- G08B17/125—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions by using a video camera to detect fire or smoke
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Alarm Systems (AREA)
Abstract
The present invention provides a kind of monitoring of bridge opening space safety and alarm system based on deep learning, is related to intelligent security guard technical field, including video acquisition module, pedestrian's intrusion detection module, pyrotechnics detection module, information management module, alarm module；Video acquisition module is connected with pedestrian's intrusion detection module, pyrotechnics detection module input terminal respectively, and pedestrian's intrusion detection module, pyrotechnics detection module output end are connected with information management module input terminal respectively, and information management module output end is connected with alarm module.The present invention is based on deep learning combination recognition of face with match to bridge opening space specify region perform intrusion detection, based on deep learning combination image processing techniques to progress pyrotechnics identification in monitoring coverage area under bridge, bridge opening space video is acquired by monitor camera, detect pedestrian's invasion and pyrotechnics, and warning message is sent to patrol officer in time, take relevant measure to ensure bridge opening safety.
Description
Technical field
The invention belongs to intelligent security guard technical fields, and in particular to a kind of bridge opening space safety monitoring based on deep learning
With alarm system.
Background technique
It emerges one after another with the rapid development of economy, highway arterial highway bridge is overhead, there is also some while having a good transport service
Security risk, such as homeless staff occupy vacant bridge opening space as temporary accommodation, and neighbouring resident is even directly in bridge opening
Burning domestic garbage, easily initiation fire, endanger the quality framework of bridge pipeline facility and bridge itself.
Bridge opening spatial scene is analyzed using existing monitor video, due to utilizing common Face datection discrimination not
Height, majority of case need face face video camera and it is static can just identify for a moment, not only face recognition accuracy rate ratio
It is lower, and be difficult to accurately detect whether bridge opening space one skilled in the art invades.
It is detected using the color characteristic of flame and smog, is easy to be imaged by external environmental interference, such as night red,
Flame loses original color space, car light interference, haze interference etc., and computationally intensive in practical applications, scene bad adaptability is held
Easily fail to report wrong report phenomenon.
(1) the technical issues of solving
The present invention is difficult to what phenomena such as bridge opening space pedestrian is invaded and burnt away the refuse accurately was detected for the prior art
Defect problem is proposed a kind of monitoring of bridge opening space safety and alarm system based on deep learning, is combined based on deep learning
Recognition of face with match to bridge opening space specify region perform intrusion detection, based on deep learning combination image processing techniques to bridge
Carry out pyrotechnics identification in lower monitoring coverage area, bridge opening space video acquired by monitor camera, detection pedestrian invasion and
Pyrotechnics, and warning message is sent to patrol officer in time, take relevant measure to ensure bridge opening safety.
(2) technical solution
In order to achieve the above object, the present invention is achieved by the following technical programs:
A kind of monitoring of bridge opening space safety and alarm system based on deep learning, including video acquisition module, pedestrian enter
Invade detection module, pyrotechnics detection module, information management module, alarm module；Video acquisition module respectively with pedestrian's intrusion detection
Module, pyrotechnics detection module input terminal be connected, pedestrian's intrusion detection module, pyrotechnics detection module output end respectively with information management
Module input is connected, and information management module output end is connected with alarm module；
The video acquisition module acquires bridge opening video using monitoring camera；
Pedestrian's intrusion detection module captures the pedestrian of bridge opening space invasion by video acquisition module, using depth
Practise Face datection and the collected video detection pedestrian intrusion behavior of the matching analysis bridge opening space monitor camera；
The pyrotechnics detection module passes through deep learning residual error using deep learning neural network model training pyrotechnics data
Network training Image Feature Detection pyrotechnics；
The information management module establishes pedestrian and invades data management and pyrotechnics data management library, supervises to bridge opening space safety
It is related to data information during survey to be managed；
The alarm module to detect pedestrian invasion or pyrotechnic issue alarm signal.
Abnormal time, place, people information are sent to and patrol by an embodiment according to the present invention, the information management module
Inspection personnel alarm, while in map interface mark pedestrian invasion, the alert icon of pyrotechnics, and realize comprising alert event
Time, place, event type report, check historical events for subsequent, analyze data.
An embodiment according to the present invention, it includes following step that pedestrian's intrusion detection module, which carries out pedestrian's intrusion detection,
It is rapid:
S1, it establishes and model is invaded based on deep learning neural metwork training bridge opening space pedestrian；
S2, the pedestrian that the invasion of bridge opening space is captured by web camera, by collected data through deep learning face
Identification is handled.
An embodiment according to the present invention, step S1 training open source image set first, iteration 200000 times, obtains logical
Bridge opening space and pedestrian sample is added, refinement training is carried out, obtains on the basis of universal model with pedestrian's character representation model
Bridge opening pedestrian's IDS Framework.
An embodiment according to the present invention, the step S2 once identifies that someone invades in the space of bridge opening, then to inspection
Whether personnel send warning message, including spot, incident personal information, while being more by face matching judgment invader
Personnel in secondary invasion bridge opening will notify patrol officer to check information management record simultaneously if matching is the invader repeatedly occurred
Judge whether to settle down herein in conjunction with video pictures, if then it is notified to move away from as early as possible, if doubtful thief is equal to relevant departments and
Shi Fanying.
An embodiment according to the present invention, it is as follows that the pyrotechnics detection module carries out pyrotechnics testing process:
Roadside is acquired by web crawlers technology and burns image, and the normal scene image in bridge opening space is added, forms positive and negative sample
This collection, the training classifier under Caffe environment, identification is normal, lights a fire, three kinds of scenes of smoldering；
Using deep learning neural network model training pyrotechnics data, full articulamentum exports 3 classes, respectively indicates fire, cigarette, just
Normal scene；
When system detection is to pyrotechnics, time, place, people information are just sent to patrol officer, carried out timely and effective
Management, it is ensured that the safety in bridge opening space.
(3) beneficial effect
Beneficial effects of the present invention: it is a kind of based on deep learning bridge opening space safety monitoring and alarm system, have with
It is lower the utility model has the advantages that
(1) using deep learning Face datection and the collected video of the matching analysis bridge opening space monitor camera, detection
Pedestrian's intrusion behavior, relative to traditional shallow-layer learning art for detecting and matching face by simple eyes feature, no
The condition limitation of Face datection is addressed only, and improves Detection accuracy.
(2) pyrotechnics is detected using the method based on deep learning neural metwork training characteristics of image, not only solves flame
The extraneous rings such as original color, car light interference are lost vulnerable to the imaging of night red, flame when being detected with smog with color characteristic
The problem of border is interfered, also solves computationally intensive problem, improves scene adaptability and detection hit rate.
(3) present invention contains information management module, in map interface mark pedestrian invasion, the alert icon of pyrotechnics, and in fact
Existing historical data, current data report, time, place comprising alert event, event type check historical events for subsequent；
Information management module can not only save current data, but also can be by analysis of history data, frequently, repeatedly to pedestrian's invasion
There is the bridge opening monitoring dynamics of pyrotechnics phenomenon, recall monitor camera picture, analyze its reason, take appropriate measures,
Pyrotechnics caused by artificial to malice, punishes it, it is ensured that the safety in bridge opening space accordingly.
(4) present invention contains bridge opening abnormal alarm module, based on deep learning Face datection and matching and image classification
Technology analyzes the collected video of monitor camera, detects that pedestrian's invasion and pyrotechnics are issued to patrol officer in time
Spot can be sent to patrol officer by alarm signal, alarm module, reach in-situ processing exception in time convenient for patrol officer
Event.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is invention's principle block diagram；
Fig. 2 is flow chart of the present invention；
Fig. 3 is pyrotechnics overhaul flow chart；
Fig. 4 is management flow figure；
Fig. 5 is alarm flow figure.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
In conjunction with Fig. 1, a kind of monitoring of bridge opening space safety and alarm system based on deep learning, including video acquisition mould
Block, pedestrian's intrusion detection module, pyrotechnics detection module, information management module, alarm module；Video acquisition module respectively with pedestrian
Intrusion detection module, pyrotechnics detection module input terminal be connected, pedestrian's intrusion detection module, pyrotechnics detection module output end respectively with
Information management module input terminal is connected, and information management module output end is connected with alarm module.
Video acquisition module acquires bridge opening video using monitoring camera.
Pedestrian's intrusion detection module captures the pedestrian of bridge opening space invasion by video acquisition module, using deep learning people
Face detection and the collected video detection pedestrian intrusion behavior of the matching analysis bridge opening space monitor camera.
Pyrotechnics detection module passes through deep learning residual error network using deep learning neural network model training pyrotechnics data
Training image feature detects pyrotechnics.
Information management module establishes pedestrian and invades data management and pyrotechnics data management library, to the monitoring of bridge opening space safety
It is related to data information in the process to be managed.
Alarm module to detect pedestrian invasion or pyrotechnic occur issue alarm signal.
In conjunction with Fig. 2, specific flow chart, video acquisition module acquires bridge opening video by monitoring camera, by depth
The correlation model for practising residual error network training bridge opening space analyzes the video of bridge opening monitoring camera acquisition；It is loaded into row
It is empty to detect bridge opening using bridge opening pedestrian's IDS Framework and pyrotechnics classifier for contrived intrusion detection module, pyrotechnics detection module
Between whether have pedestrian invade and pyrotechnics；Information management module by abnormal time, place, people information be sent to patrol officer into
Row alarm, while map interface mark pedestrian invasion, pyrotechnics alert icon, and realize comprising alert event time,
The report of point, event type checks historical events for subsequent, analyzes data.
Frequent bridge opening, which is invaded, for pedestrian adds guardrail, monitoring dynamics, the invasion still occurred to continual warnings
Personnel take appropriate measures, to ensure the safety in bridge opening space；For the bridge opening for pyrotechnics phenomenon repeatedly occur, recalls monitoring and take the photograph
Camera picture analyzes Producing reason, takes appropriate measures, and caused pyrotechnics artificial to malice locates it accordingly
It penalizes, it is ensured that the safety in bridge opening space.
Pedestrian's intrusion detection module carry out pedestrian's intrusion detection the following steps are included:
S1, it establishes and model is invaded based on deep learning neural metwork training bridge opening space pedestrian；
Training open source image set first, iteration 200000 times, obtains general pedestrian's character representation model, in universal model base
On plinth, bridge opening space and pedestrian sample is added, carries out refinement training, obtains bridge opening pedestrian's IDS Framework；
S2, the pedestrian that the invasion of bridge opening space is captured by web camera, by collected data through deep learning face
Identification is handled.
Cross entropy loss function is used when carrying out Face datection:
Frame, which returns, uses quadratic sum loss function:
Facial modeling also uses quadratic sum loss function:
There are many different tasks on entire convolutional neural networks frame, therefore use following letter in multitask training
Number:
Wherein α indicates the importance of task.
Once identifying that someone invades in the space of bridge opening, then sends warning message, including spot, thing to patrol officer
Personal information etc. is sent out, while whether being the multiple personnel invaded in bridge opening by face matching judgment invader, if matching is more
The invader of secondary appearance will then notify patrol officer to check that information management records and video pictures is combined to judge whether to settle down herein,
If then it is notified to move away from as early as possible, reflect in time if doubtful thief is equal to relevant departments.
Since pyrotechnics external appearance characteristic changes greatly, wrong report, scene bad adaptability are easily failed to report under external environmental interference.In conjunction with
Fig. 3 pyrotechnics overhaul flow chart, pyrotechnics detection module carry out pyrotechnics detection.Roadside, which is acquired, by web crawlers technology burns image,
The normal scene image in bridge opening space is added, forms positive and negative sample set, the training classifier under Caffe environment identifies normal, raw
Fire, three kinds of scenes of smoldering；Using deep learning neural network model training pyrotechnics data, full articulamentum exports 3 classes, respectively indicates
Fire, cigarette, normal scene；When system detection is to pyrotechnics, time, place, people information are just sent to patrol officer, carry out and
When effectively manage, it is ensured that the safety in bridge opening space.
During handling image using deep learning convolutional neural networks, the spy of image is calculated using matrix convolution
Sign, virtual value convolution is defined as:
When calculating activation value by forward-propagating, the output of the upper one layer convolutional layer for input layer are as follows:
Wherein b(l)For bias unit.
The output of sub-sampling layer are as follows:
Using the method in average pond, the weight of each unit of the convolution kernel used is β(l+1), each single item convolution operation
A bias unit b is still added afterwards(l+1)。
Upper one layer be sub-sampling layer convolutional layer output are as follows:
Bridge opening space safety is related to the management of data information during monitoring, as shown in figure 4, information management module point
Pedestrian is not established invades data management and pyrotechnics data management library.Pedestrian invades the time of essential record time generation, place (i.e.
Bridge opening address), personage, invasion number, the same time invasion number.Wherein the record of time is conducive to analyze pedestrian's invasion
Time tendency, the record in place are conducive to analyze the information such as safety coefficient, the bridge opening geographical advantage in bridge opening, the record purpose of personage
It is to prevent the aggregation ground of working at a selected spot of the bad personnel such as thief, then to look for by recognition of face Information locating for invading frequent personnel
To this person, it is cross-examined etc., prevents its bad behavior and causes bridge opening space and neighbouring danger.For the information of pyrotechnics
Management essentially consists in the record time, place, steps up patrols to the bridge opening for frequently occurring pyrotechnics, recalls monitor camera picture, point
Its reason is analysed, is taken appropriate measures, caused pyrotechnics artificial to malice is punished it accordingly, avoided the occurrence of great
Loss, it is ensured that the safety in bridge opening space.The management of information and record are the important evidences of later period investigation and data analysis, to bridge
The safety in hole space has good prevention effect.
In conjunction with Fig. 5 alarm flow figure, based on deep learning Face datection and matching and image classification to monitor camera
Collected video is analyzed, and detects that pedestrian's invasion and pyrotechnics issue alarm signal to patrol officer in time, mould of alarming
Spot can be sent to patrol officer by block.
In conclusion the embodiment of the present invention, the monitoring of bridge opening space safety and alarm system based on deep learning, have with
Lower advantage:
(1) using deep learning Face datection and the collected video of the matching analysis bridge opening space monitor camera, detection
Pedestrian's intrusion behavior, relative to traditional shallow-layer learning art for detecting and matching face by simple eyes feature, no
The condition limitation of Face datection is addressed only, and improves Detection accuracy.
(2) pyrotechnics is detected using the method based on deep learning neural metwork training characteristics of image, not only solves flame
The extraneous rings such as original color, car light interference are lost vulnerable to the imaging of night red, flame when being detected with smog with color characteristic
The problem of border is interfered, also solves computationally intensive problem, improves scene adaptability and detection hit rate.
(3) present invention contains information management module, in map interface mark pedestrian invasion, the alert icon of pyrotechnics, and in fact
Existing historical data, current data report, time, place comprising alert event, event type check historical events for subsequent；
Information management module can not only save current data, but also can be by analysis of history data, frequently, repeatedly to pedestrian's invasion
There is the bridge opening monitoring dynamics of pyrotechnics phenomenon, recall monitor camera picture, analyze its reason, take appropriate measures,
Pyrotechnics caused by artificial to malice, punishes it, it is ensured that the safety in bridge opening space accordingly.
(4) present invention contains bridge opening abnormal alarm module, based on deep learning Face datection and matching and image classification
Technology analyzes the collected video of monitor camera, detects that pedestrian's invasion and pyrotechnics are issued to patrol officer in time
Spot can be sent to patrol officer by alarm signal, alarm module, reach in-situ processing exception in time convenient for patrol officer
Event.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (6)
1. a kind of monitoring of bridge opening space safety and alarm system based on deep learning, which is characterized in that including video acquisition mould
Block, pedestrian's intrusion detection module, pyrotechnics detection module, information management module, alarm module；Video acquisition module respectively with pedestrian
Intrusion detection module, pyrotechnics detection module input terminal be connected, pedestrian's intrusion detection module, pyrotechnics detection module output end respectively with
Information management module input terminal is connected, and information management module output end is connected with alarm module；
The video acquisition module acquires bridge opening video using monitoring camera；
Pedestrian's intrusion detection module captures the pedestrian of bridge opening space invasion by video acquisition module, using deep learning people
Face detection and the collected video detection pedestrian intrusion behavior of the matching analysis bridge opening space monitor camera；
The pyrotechnics detection module passes through deep learning residual error network using deep learning neural network model training pyrotechnics data
Training image feature detects pyrotechnics；
The information management module establishes pedestrian and invades data management and pyrotechnics data management library, to the monitoring of bridge opening space safety
It is related to data information in the process to be managed；
The alarm module to detect pedestrian invasion or pyrotechnic issue alarm signal.
2. a kind of monitoring of bridge opening space safety and alarm system, feature based on deep learning as described in claim 1 exists
In abnormal time, place, people information are sent to patrol officer and alarmed by the information management module, while in map
Interface marks the alert icon of pedestrian's invasion, pyrotechnics, and realize comprising time of alert event, place, event type report,
Historical events is checked for subsequent, analyzes data.
3. a kind of monitoring of bridge opening space safety and alarm system, feature based on deep learning as described in claim 1 exists
Pedestrian's intrusion detection is carried out in, pedestrian's intrusion detection module the following steps are included:
S1, it establishes and model is invaded based on deep learning neural metwork training bridge opening space pedestrian；
S2, the pedestrian that the invasion of bridge opening space is captured by web camera, by collected data through deep learning recognition of face
It is handled.
4. a kind of monitoring of bridge opening space safety and alarm system, feature based on deep learning as claimed in claim 3 exists
In step S1 training open source image set first iteration 200000 times, obtains general pedestrian's character representation model, general
On the basis of model, bridge opening space and pedestrian sample is added, carries out refinement training, obtains bridge opening pedestrian's IDS Framework.
5. a kind of monitoring of bridge opening space safety and alarm system, feature based on deep learning as claimed in claim 3 exists
In the step S2 once identifies that someone invades in the space of bridge opening, then sends warning message, including spot to patrol officer
Point, incident personal information, while whether being the multiple personnel invaded in bridge opening by face matching judgment invader, if matching is
The invader repeatedly occurred will then notify patrol officer to check that information management records and video pictures is combined to judge whether to settle down this
Place is reflected if doubtful thief waits to relevant departments in time if then it is notified to move away from as early as possible.
6. a kind of monitoring of bridge opening space safety and alarm system, feature based on deep learning as described in claim 1 exists
In it is as follows that the pyrotechnics detection module carries out pyrotechnics testing process:
Roadside is acquired by web crawlers technology and burns image, and the normal scene image in bridge opening space is added, forms positive and negative sample set,
The training classifier under Caffe environment, identification is normal, lights a fire, three kinds of scenes of smoldering；
Using deep learning neural network model training pyrotechnics data, full articulamentum exports 3 classes, respectively indicates fire, cigarette, normal field
Scape；
When system detection is to pyrotechnics, time, place, people information are just sent to patrol officer, timely and effectively managed
Reason, it is ensured that the safety in bridge opening space.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811383003.2A CN109410497B (en) | 2018-11-20 | 2018-11-20 | Bridge opening space safety monitoring and alarming system based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811383003.2A CN109410497B (en) | 2018-11-20 | 2018-11-20 | Bridge opening space safety monitoring and alarming system based on deep learning |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109410497A true CN109410497A (en) | 2019-03-01 |
| CN109410497B CN109410497B (en) | 2021-01-19 |
Family
ID=65474093
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811383003.2A Active CN109410497B (en) | 2018-11-20 | 2018-11-20 | Bridge opening space safety monitoring and alarming system based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109410497B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111432182A (en) * | 2020-04-29 | 2020-07-17 | 上善智城(苏州)信息科技有限公司 | Safety supervision method and system for oil discharge place of gas station |
| CN111553264A (en) * | 2020-04-27 | 2020-08-18 | 中国科学技术大学先进技术研究院 | Campus non-safety behavior detection and early warning method suitable for primary and secondary school students |
| CN113657298A (en) * | 2021-08-20 | 2021-11-16 | 软通动力信息技术(集团)股份有限公司 | Pedestrian intrusion identification method, device, equipment and medium based on large displacement tracking |
| CN114267082A (en) * | 2021-09-16 | 2022-04-01 | 南京邮电大学 | Bridge side falling behavior identification method based on deep understanding |
| CN115082834A (en) * | 2022-07-20 | 2022-09-20 | 成都考拉悠然科技有限公司 | Engineering vehicle black smoke emission monitoring method and system based on deep learning |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106355162A (en) * | 2016-09-23 | 2017-01-25 | 江西洪都航空工业集团有限责任公司 | Method for detecting intrusion on basis of video monitoring |
| CN106372576A (en) * | 2016-08-23 | 2017-02-01 | 南京邮电大学 | Deep learning-based intelligent indoor intrusion detection method and system |
| CN108389359A (en) * | 2018-04-10 | 2018-08-10 | 中国矿业大学 | A kind of Urban Fires alarm method based on deep learning |
| US20180260630A1 (en) * | 2017-03-10 | 2018-09-13 | Turing Video, Inc. | Activity recognition method and system |
| CN108777777A (en) * | 2018-05-04 | 2018-11-09 | 江苏理工学院 | A kind of monitor video crop straw burning method for inspecting based on deep neural network |
-
2018
- 2018-11-20 CN CN201811383003.2A patent/CN109410497B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106372576A (en) * | 2016-08-23 | 2017-02-01 | 南京邮电大学 | Deep learning-based intelligent indoor intrusion detection method and system |
| CN106355162A (en) * | 2016-09-23 | 2017-01-25 | 江西洪都航空工业集团有限责任公司 | Method for detecting intrusion on basis of video monitoring |
| US20180260630A1 (en) * | 2017-03-10 | 2018-09-13 | Turing Video, Inc. | Activity recognition method and system |
| CN108389359A (en) * | 2018-04-10 | 2018-08-10 | 中国矿业大学 | A kind of Urban Fires alarm method based on deep learning |
| CN108777777A (en) * | 2018-05-04 | 2018-11-09 | 江苏理工学院 | A kind of monitor video crop straw burning method for inspecting based on deep neural network |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111553264A (en) * | 2020-04-27 | 2020-08-18 | 中国科学技术大学先进技术研究院 | Campus non-safety behavior detection and early warning method suitable for primary and secondary school students |
| CN111553264B (en) * | 2020-04-27 | 2023-04-18 | 中科永安(安徽)科技有限公司 | Campus non-safety behavior detection and early warning method suitable for primary and secondary school students |
| CN111432182A (en) * | 2020-04-29 | 2020-07-17 | 上善智城(苏州)信息科技有限公司 | Safety supervision method and system for oil discharge place of gas station |
| CN113657298A (en) * | 2021-08-20 | 2021-11-16 | 软通动力信息技术(集团)股份有限公司 | Pedestrian intrusion identification method, device, equipment and medium based on large displacement tracking |
| CN114267082A (en) * | 2021-09-16 | 2022-04-01 | 南京邮电大学 | Bridge side falling behavior identification method based on deep understanding |
| CN114267082B (en) * | 2021-09-16 | 2023-08-11 | 南京邮电大学 | Bridge side falling behavior identification method based on depth understanding |
| CN115082834A (en) * | 2022-07-20 | 2022-09-20 | 成都考拉悠然科技有限公司 | Engineering vehicle black smoke emission monitoring method and system based on deep learning |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109410497B (en) | 2021-01-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109410497A (en) | A kind of monitoring of bridge opening space safety and alarm system based on deep learning | |
| Gong et al. | A Real‐Time Fire Detection Method from Video with Multifeature Fusion | |
| CN109858365B (en) | Special crowd gathering behavior analysis method and device and electronic equipment | |
| CN103108159B (en) | Electric power intelligent video analyzing and monitoring system and method | |
| CN101751744B (en) | Detection and early warning method of smoke | |
| CN106454250A (en) | Intelligent recognition and early warning processing information platform | |
| CN106650584B (en) | Flame detecting method and system | |
| CN110428522A (en) | A kind of intelligent safety and defence system of wisdom new city | |
| CN101609588A (en) | Full-automatic anti-intrusion intelligent video monitoring alarm system for unattended villa | |
| CN110619277A (en) | Multi-community intelligent deployment and control method and system | |
| CN103514694A (en) | Intrusion detection monitoring system | |
| CN111462451A (en) | Straw burning detection alarm system based on video information | |
| CN106647569A (en) | Intelligent inspection robot and method | |
| CN110867046A (en) | Intelligent car washer video monitoring and early warning system based on cloud computing | |
| CN106454253A (en) | Method and system for detecting area wandering | |
| CN109389794A (en) | A kind of Intellectualized Video Monitoring method and system | |
| CN112132048A (en) | Community patrol analysis method and system based on computer vision | |
| KR20190035187A (en) | Sound alarm broadcasting system in monitoring area | |
| CN110780356A (en) | Subway platform clearance foreign matter detecting system | |
| CN109001210B (en) | System and method for detecting aging and cracking of sealing rubber strip of civil air defense door | |
| CN203327163U (en) | Monitoring system | |
| CN210222962U (en) | Intelligent electronic fence system | |
| CN111753587A (en) | Method and device for detecting falling to ground | |
| CN116993265A (en) | Intelligent warehouse safety management system based on Internet of things | |
| CN115346170B (en) | Intelligent monitoring method and device for gas facility area |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |