CN105100757A - Detector with integrated structure - Google Patents
Detector with integrated structure Download PDFInfo
- Publication number
- CN105100757A CN105100757A CN201510595412.9A CN201510595412A CN105100757A CN 105100757 A CN105100757 A CN 105100757A CN 201510595412 A CN201510595412 A CN 201510595412A CN 105100757 A CN105100757 A CN 105100757A
- Authority
- CN
- China
- Prior art keywords
- prospect
- module
- detector
- pedestrian
- foreground
- 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
Landscapes
- Burglar Alarm Systems (AREA)
- Image Analysis (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention provides a detector with an integrated structure. The detector with the integrated structure comprises a detector, a fixing device, a light supplementing device and a rainproof and thermally-insulated shield. The detector and the light supplementing device are vertically arranged and fixed together via the fixing device. The detector comprises a detector body and a detector shell. The light supplementing device comprises a light supplementing device body and a light supplementing device shell. The detector shell and the light supplementing device shell respectively comprise an upper cover, an intermediate shell, an upper flange and a lower flange. The upper cover and the upper end of the intermediate shell are connected via threads. Meanwhile, the upper cover and the intermediate shell are also connected via nuts, the lower end of the intermediate shell and the upper flange have an integrated structure, and the lower flange is a lower cover which is connected with the intermediate shell via threads. Meanwhile, the upper and lower flanges are connected via nuts. The rainproof and thermally-insulated shield is fixed above the fixing device so that use demands of users are met.
Description
Technical field
The present invention relates to computer vision field, especially relate to a kind of detector with integral structure.
Background technology
Along with developing rapidly of science and technology, detector has been applied in every field.But, in some occasion, high requirement is proposed to the shell of detector and light compensating apparatus, the shell mechanism of existing detector and light compensating apparatus, cannot user demand be met.
In addition, in some occasion, as in tunnel, light is poor, and car is more, but often having pedestrian to avoid detouring to take a shortcut, taking a risk to walk from tunnel, this is very large traffic safety hidden danger.At present, the quantity of traffic administration personnel can not catch up with the quantity of means of transportation, someone monitors in real time can not to reach each traffic section, and, if monitor staff continued observation video more than 22 minutes, so can omit the site activity of 95%, mean that monitoring probability is down to 0.2%, namely rely on human eye detection reliability lower, so in the urgent need to intelligent monitoring device, assist traffic administration personnel to carry out traffic monitoring and control, this will have huge social benefit and economic benefit.
Summary of the invention
The object of the invention is to propose a kind of detector with integral structure, to solve the shell mechanism of existing detector and the light compensating apparatus existed in prior art, the problem of user demand cannot be met.
In order to achieve the above object, the present invention proposes a kind of detector with integral structure, comprise detector, fixture and light compensating apparatus, rainproof heat shield, described detector and light compensating apparatus are arranged above and below setting, be fixed together by described fixture, described detector comprises detector body and probe body, described light compensating apparatus comprises light compensating apparatus body and light compensating apparatus shell, described probe body and light compensating apparatus shell include upper cover, middle casing, upper flange, lower flange, described upper cover is connected with the threaded upper ends of middle casing, simultaneously, described upper cover is also connected by nut with middle casing, the lower end of described middle casing and described upper flange are integral type structure, described lower flange is a lower cover, itself and described middle casing are threaded connection, simultaneously, described upper lower flange is connected by nut, wherein said rainproof heat shield is fixed on above fixture, ventilation gap is left between described rainproof heat shield and described probe body.
Wherein, described fixture comprises left side baffle plate, right-hand apron and base, described left side baffle plate and right-hand apron symmetrical configuration, include horizontal canned paragraph from top to bottom, first vertical canned paragraph, first arcuate segment, second arcuate segment and the second vertical canned paragraph, described base is bilateral symmetry, upper end open, the width being wider than upper end open of lower end base plate, wherein, the left and right sides includes the 3rd vertical canned paragraph and tilting section, the upper end of described tilting section is fixedly connected with the lower end of the 3rd vertical canned paragraph, lower end is fixedly connected with one end of lower end base plate, described left side baffle plate and right-hand apron symmetry are placed, first vertical canned paragraph, second vertical canned paragraph, 3rd vertical canned paragraph is provided with fixing hole, the corresponding fixing hole of the left and right sides, fix by screw, the horizontal canned paragraph of both sides has at least one to be provided with fixing hole, described detector is fixed in the frame of described first arcuate segment formation, described light compensating apparatus is fixed in the frame of described second arcuate segment formation, described rainproof heat shield is connected with fixture by the fixing hole of described horizontal canned paragraph, ventilation gap is left between described rainproof heat shield and described probe body, second vertical canned paragraph of the described left and right sides is fixedly connected with by screw with the fixing hole of the 3rd vertical canned paragraph, described detector body is arranged in described probe body.
Wherein, described detector body comprises sheet metal component, circuit board, camera lens, buttock line, described circuit board is arranged on described sheet metal component, described camera lens comprises the first camera lens and the second camera lens, described first camera lens is arranged on B/W camera front end, described first camera lens front end is provided with a filter by filter adapter ring, described second camera lens is arranged on colour TV camera front end, described first camera lens is all connected with described circuit board with the second camera lens, described light compensating apparatus body, comprises six electroluminescent lamps arranged in the form of a ring.
Wherein, comprise tunnel pedestrian's intrusion detection module, described tunnel pedestrian's intrusion detection module comprises video acquisition unit, intelligent detection unit, memory cell, and video acquisition unit is connected with intelligent detection unit, display unit, memory cell; Intelligent detection unit is connected with video acquisition unit, display unit, memory cell, alarm unit; Intelligent detection unit and memory cell, display unit carry out Signal transmissions by fiber optic cables; Intelligent detection unit exports I/O signal to alarm unit.
Wherein, described video acquisition unit comprises colour TV camera, infrared illumination lamp, video compression coding module; Described intelligent detection unit comprises Background learning module, foreground detection module, invasion region setting module, invasion judge module and alarm module, picture compression coding module; Described memory cell comprises original video module and invasion picture module.
Wherein, described intelligent detection unit, by Background learning module, set up image background model, and along with scene, the change of light etc., intelligence is carried out study and is upgraded, then foreground detection module is carried out, namely to people emerging in scene or thing, carry out position and feature detection, and real time record upgrades foreground information, namely invasion region setting module sets the region that in tunnel, pedestrian may occur, and then enter invasion judge module, by foreground information and invasion regional location and the comparison of the time of invasion, judge whether to belong to tunnel pedestrian's intrusion event, if belong to pedestrian's invasion, alarm module exports I/O alarm signal to alarm unit, simultaneously, locking pedestrian positional information, be transferred to display unit, and invasion picture module tunnel pedestrian's intrusion event picture is transferred to after picture compression coding module 26 carries out compressed encoding in memory cell, so that follow-up verification.
Wherein, Background learning module in described intelligent detection unit, by carrying out Gauss's modeling to the image collected, utilize the Y of image, U, V component sets up the model library of image, adopt 2*2 region modeling point, each modeling point comprises four Gauss models, Gauss model comprises model gray scale, model variance, Model Weight, for the first two field picture, carry out model parameter initialization, for the image that subsequent transmission is come, mate with four Gauss models at corresponding pixel points place respectively, if coupling, then corresponding Gauss model weights increase, simultaneously and Gauss model is upgraded, if do not mate, then replace the minimum Gauss model of weight, when occurring that weights are greater than modeling threshold value in modeling point four Gauss models, current pixel point modeling success, the successful pixel number of modeling in statistics present image, when pixel modeling success count is greater than setting threshold, then start foreground detection, otherwise, proceed Background learning, modeling threshold value generally gets 1/5 ~ 1/3 of modeling point sum.
Wherein, foreground detection module in described intelligent detection unit, be detect emerging people or thing in scene, carry out Gauss's modeling to the image that transmission comes, each modeling point comprises a Gauss model, the Gauss model of modeling point Gauss model with corresponding background modeling point place is compared, if do not mated, then continue to compare, if all do not mated with modeling point four neighborhood Gauss model, then be defined as foreground point, otherwise be background dot; Terminate if current frame image modeling point detects, then expansion process is carried out to foreground point, belong to the pixel of same prospect with UNICOM, simultaneously, prospect is marked, and calculates the filling rate of foreground blocks, area, center, depth-width ratio information, stored in prospect historical data base.
Wherein, in described intelligent detection unit, invade region setting module, refer to the region forbidding occurring pedestrian in setting tunnel, by setting four region lines along tunnel edge, delineation invasion region.
Wherein, in described intelligent detection unit, invade judge module, refer to and invasion judgement is carried out to the prospect detected, due to judgement is the pedestrian information of invading, pedestrian's judgement is carried out to prospect, namely pedestrian is determine whether by foreground area, filling rate, depth-width ratio information, filling rate is the ratio of foreground pixel point and the prospect frame comprising the whole pixel of current prospect, and depth-width ratio is the ratio of the height and the width when the prospect frame comprising the whole pixel of current prospect, in tunnel pedestrian's intrusion detection, video acquisition unit is just installed tunnel, according to pedestrian's distance, the foreground area detected varies in size, the foreground area utilizing distance to detect is different, the i.e. difference of the depth of field, image is divided, be divided into three regions, each region prospect setting area maximin, when only having pedestrian to meet area constraints and depth-width ratio in corresponding region, then judge that current prospect is as pedestrian, this avoid the interference of too small fruit stone etc. simultaneously, it also avoid the mistake identification of dilly, improve recognition accuracy, after pedestrian's judgement terminates, center according to prospect changes, matched jamming is carried out to effective foreground information that prospect is behaved, and be less than rate of change threshold value further by the area change of pedestrian's prospect, the change of prospect average gray value is less than threshold value and carries out same prospect judgement restriction, if same prospect becomes, the match is successful, with the average gray value of prospect in current foreground information real-time update historical data base, filling rate, area, center, depth-width ratio information, the effective prospect disappeared in the visual field is removed, to the average gray value of emerging effective prospect, filling rate, area, center, depth-width ratio information is stored in prospect historical data base, alarm module in described intelligent detection unit, refers to when prospect pedestrian appears at protecting tunnel region, then sends I/O alarm signal, and meanwhile, alarm picture is stored in the invasion picture module in memory cell by picture compression coding module, in described intelligent detection unit, picture compression coding module is used for alarm picture compressed encoding, is convenient to store.
The present invention, compared with prior art, by the shell of detector and light compensating apparatus is set to integral structure, sealing strengthens greatly, meets the use of client.
Accompanying drawing explanation
Fig. 1 is structure first schematic diagram of the present invention;
Fig. 2 is structure second schematic diagram of the present invention;
The structured flowchart of Fig. 3 tunnel of the present invention pedestrian's intrusion detection module;
Fig. 4 is tunnel pedestrian's intrusion detection modules Background study module block diagram of the present invention;
Fig. 5 is tunnel pedestrian's intrusion detection module foreground detection module frame chart of the present invention;
Fig. 6 is pedestrian's intrusion detection module in tunnel of the present invention invasion judge module block diagram;
Fig. 7 is tunnel pedestrian's intrusion detection alarm condition key diagram of the present invention.
Embodiment
With reference to the accompanying drawings and the present invention is described in detail in conjunction with instantiation.Be to be noted that described example is only for the ease of the understanding of the present invention, therefore do not limit protection scope of the present invention.
Embodiment 1
As Fig. 1, shown in Fig. 2, the invention discloses a kind of detector with integral structure, comprise detector 1, fixture 3 and light compensating apparatus 2, rainproof heat shield, described detector 1 and light compensating apparatus 2 are arranged above and below setting, be fixed together by described fixture 3, described detector 1 comprises detector body and probe body, described light compensating apparatus 2 comprises light compensating apparatus body and light compensating apparatus shell, described probe body and light compensating apparatus shell include upper cover, middle casing, upper flange, lower flange, described upper cover is connected with the threaded upper ends of middle casing, simultaneously, described upper cover is also connected by nut with middle casing, the lower end of described middle casing and described upper flange are integral type structure, described lower flange is a lower cover, itself and described middle casing are threaded connection, simultaneously, described upper lower flange is connected by nut, wherein said rainproof heat shield is fixed on above fixture, ventilation gap is left between described rainproof heat shield and described probe body.
Wherein, described fixture comprises left side baffle plate, right-hand apron and base, described left side baffle plate and right-hand apron symmetrical configuration, include horizontal canned paragraph from top to bottom, first vertical canned paragraph, first arcuate segment, second arcuate segment and the second vertical canned paragraph, described base is bilateral symmetry, upper end open, the width being wider than upper end open of lower end base plate, wherein, the left and right sides includes the 3rd vertical canned paragraph and tilting section, the upper end of described tilting section is fixedly connected with the lower end of the 3rd vertical canned paragraph, lower end is fixedly connected with one end of lower end base plate, described left side baffle plate and right-hand apron symmetry are placed, first vertical canned paragraph, second vertical canned paragraph, 3rd vertical canned paragraph is provided with fixing hole, the corresponding fixing hole of the left and right sides, fix by screw, the horizontal canned paragraph of both sides has at least one to be provided with fixing hole, described detector is fixed in the frame of described first arcuate segment formation, described light compensating apparatus is fixed in the frame of described second arcuate segment formation, described rainproof heat shield is connected with fixture by the fixing hole of described horizontal canned paragraph, ventilation gap is left between described rainproof heat shield and described probe body, second vertical canned paragraph of the described left and right sides is fixedly connected with by screw with the fixing hole of the 3rd vertical canned paragraph, described detector body is arranged in described probe body.
Wherein, described detector body comprises sheet metal component, circuit board, camera lens, buttock line, described circuit board is arranged on described sheet metal component, described camera lens comprises the first camera lens and the second camera lens, described first camera lens is arranged on B/W camera front end, described first camera lens front end is provided with a filter by filter adapter ring, described second camera lens is arranged on colour TV camera front end, described first camera lens is all connected with described circuit board with the second camera lens, described light compensating apparatus body, comprises six electroluminescent lamps arranged in the form of a ring.
Embodiment 2
As shown in Figure 3, Figure 4, Figure 5, Figure 6, Figure 7, as shown in Figure 3, Figure 4, Figure 5, Figure 6, Figure 7, present embodiments provide a kind of tunnel pedestrian's intrusion detection module, comprise video acquisition unit 1, intelligent detection unit 2, memory cell 5, display unit 3, alarm unit 4.
Described video acquisition unit 1 is connected with intelligent detection unit 2, display unit 3, memory cell 5; Intelligent detection unit 2 is connected with video acquisition unit 1, display unit 3, memory cell 5, alarm unit 4; Video acquisition unit 1 carries out the transmission of real-time video information with display unit 3 and memory cell 5 by fiber optic cables; Video acquisition unit 1 and intelligent detection unit 2 carry out transmission of video by netting twine; Intelligent detection unit 2 and memory cell 5, display unit 3 carry out Signal transmissions by fiber optic cables; Intelligent detection unit 2 exports I/O signal to alarm unit 4.
Described video acquisition unit 1 comprises colour TV camera 11, infrared illumination lamp 12, video compression coding module 13.
Described intelligent detection unit 2 comprises Background learning module 21, foreground detection module 22, invasion region setting module 23, invasion judge module 24 and alarm module 25, picture compression coding module 26.
Described memory cell 5 comprises original video module 51 and invasion picture module 52.
Described display unit 3 is video wall or large-screen, for showing real-time video and intrusion detection result.
Described alarm unit 4 mainly alarm.
Described video acquisition unit 1 gathers real-time scene video by colour TV camera 11, and pass to intelligent detection unit 2, simultaneously, after the video gathered carries out compressed encoding by video compression coding module 13, by Optical Fiber Transmission to the original video module 51 in memory cell 5, put on record for store video.Colour TV camera 11 in described video acquisition unit 1 has day and night translation function, and day mode collects coloured image, when surrounding environment brightness is lower than a certain threshold value, opens drainage pattern at night, utilizes infrared illumination lamp 12 to throw light on, and then gather image.
Described intelligent detection unit 2, by Background learning module 21, set up image background model, and along with scene, the change of light etc., intelligence is carried out study and is upgraded, then foreground detection module 22 is carried out, namely to people emerging in scene or thing, carry out position and feature detection, and real time record upgrades foreground information, namely invasion region setting module 23 sets the region that in tunnel, pedestrian may occur, and then enter invasion judge module 24, by foreground information and invasion regional location and the comparison of the time of invasion, judge whether to belong to tunnel pedestrian's intrusion event, if belong to pedestrian's invasion, alarm module 25 exports I/O alarm signal to alarm unit 4, simultaneously, locking pedestrian positional information, be transferred to display unit 3, and the invasion picture module 52 tunnel pedestrian's intrusion event picture is transferred to after picture compression coding module 26 carries out compressed encoding in memory cell 5, so that follow-up verification.
Background learning module 21 in described intelligent detection unit 2, by carrying out Gauss's modeling to the image collected, utilize the Y of image, U, V component sets up the model library of image, the present invention adopts 2*2 region modeling point, each modeling point comprises four Gauss models, Gauss model mainly comprises model gray scale, model variance, Model Weight, for the first two field picture, carry out model parameter initialization, for the image that subsequent transmission is come, mate with four Gauss models at corresponding pixel points place respectively, if coupling, then corresponding Gauss model weights increase, simultaneously and Gauss model is upgraded, if do not mate, then replace the minimum Gauss model of weight, when occurring that weights are greater than modeling threshold value in modeling point four Gauss models, current pixel point modeling success, the successful pixel number of modeling in statistics present image, when pixel modeling success count is greater than setting threshold, then start foreground detection, otherwise, proceed Background learning, modeling threshold value generally gets 1/5 ~ 1/3 of modeling point sum.
Foreground detection module 22 in described intelligent detection unit 2, detect emerging people or thing in scene, Gauss's modeling is carried out to the image that transmission comes, each modeling point comprises a Gauss model, compares, the Gauss model of modeling point Gauss model with corresponding background modeling point place if do not mated, then continue to compare with modeling point four neighborhood Gauss model, if all do not mated, be then defined as foreground point, otherwise be background dot.Terminate if current frame image modeling point detects, then expansion process is carried out to foreground point, belong to the pixel of same prospect with UNICOM, simultaneously, prospect is marked, and calculates the filling rate of foreground blocks, area, center, depth-width ratio information, stored in prospect historical data base.
Invade region setting module 23 in described intelligent detection unit 2, refer to the region forbidding occurring pedestrian in setting tunnel, by setting four region lines along tunnel edge, delineation invasion region.
Invade judge module 24 in described intelligent detection unit 2, refer to and invasion judgement is carried out to the prospect detected.Due to judgement is the pedestrian information of invading, pedestrian's judgement is carried out to prospect, namely pedestrian is determine whether by foreground area, filling rate, depth-width ratio information, filling rate is the ratio of foreground pixel point and the prospect frame comprising the whole pixel of current prospect, and depth-width ratio is the ratio of the height and the width when the prospect frame comprising the whole pixel of current prospect.In tunnel pedestrian's intrusion detection, video acquisition unit 1 is just installed tunnel, according to pedestrian's distance, the foreground area detected varies in size, the foreground area utilizing distance to detect is different, the i.e. difference of the depth of field, image is divided, be divided into three regions, each region prospect setting area maximin, when only having pedestrian to meet area constraints and depth-width ratio in corresponding region, then judge that current prospect is as pedestrian, this avoid the interference of too small fruit stone etc. simultaneously, it also avoid the mistake identification of dilly, improve recognition accuracy.After pedestrian's judgement terminates, center according to prospect changes, matched jamming is carried out to effective foreground information that prospect is behaved, and be less than rate of change threshold value further by the area change of pedestrian's prospect, the change of prospect average gray value is less than threshold value and carries out same prospect judgement restriction, if same prospect becomes, the match is successful, with the average gray value of prospect in current foreground information real-time update historical data base, filling rate, area, center, depth-width ratio information, the effective prospect disappeared in the visual field is removed, to the average gray value of emerging effective prospect, filling rate, area, center, depth-width ratio information is stored in prospect historical data base.
Alarm module 25 in described intelligent detection unit 2, refers to when prospect pedestrian appears at protecting tunnel region, then sends I/O alarm signal, and meanwhile, alarm picture to be stored in the invasion picture module in memory cell 5 52 by picture compression coding module 26.
In described intelligent detection unit 2, picture compression coding module 26 is for by alarm picture compressed encoding, is convenient to store.
In described memory cell 5, original video module 51 transmits for receiver, video collecting unit 1 the video information of coming, and the tunnel pedestrian that invasion picture module 52 receives intelligent detection unit 2 detection invades result picture, for preserving and follow-up verification.
Described display unit 3 is video wall or large-screen, and the real-time scene video of receiver, video collecting unit 1 also shows, and is connected with intelligent detection unit 2 simultaneously, display Intelligent Measurement result, and shows tunnel pedestrian and invade information.
Described alarm unit 4 receives intelligent detection unit 1 and transmits the alarm signal of coming, if there is tunnel pedestrian's intrusion event, then reports to the police, and assists traffic control personnel to process in time, avoids traffic accident occurs.
For tunnel intrusion alarm situation explanation.In figure, two vertical direction straight lines represent the invasion zone boundary set along tunnel; horizontal two straight lines represent the invasion zone boundary of setting; when pedestrian appears at protection zone A; intelligent detection unit 2 detects that pedestrian invades information and reports to the police; when pedestrian appears at protection zone B, do not report to the police.
The above; be only the embodiment in the present invention, but protection scope of the present invention is not limited thereto, any researcher being familiar with this technology is in the technical scope disclosed by the present invention; according to the multiple change that actual conditions are made, all should be encompassed within protection scope of the present invention.Therefore, protection scope of the present invention should be as the criterion with the protection range of claims.
Claims (10)
1. the detector with integral structure, it is characterized in that, comprise detector, fixture and light compensating apparatus, rainproof heat shield, described detector and light compensating apparatus are arranged above and below setting, be fixed together by described fixture, described detector comprises detector body and probe body, described light compensating apparatus comprises light compensating apparatus body and light compensating apparatus shell, described probe body and light compensating apparatus shell include upper cover, middle casing, upper flange, lower flange, described upper cover is connected with the threaded upper ends of middle casing, simultaneously, described upper cover is also connected by nut with middle casing, the lower end of described middle casing and described upper flange are integral type structure, described lower flange is a lower cover, itself and described middle casing are threaded connection, simultaneously, described upper lower flange is connected by nut, wherein said rainproof heat shield is fixed on above fixture, ventilation gap is left between described rainproof heat shield and described probe body.
2. a kind of detector with integral structure according to claim 1, it is characterized in that, described fixture comprises left side baffle plate, right-hand apron and base, described left side baffle plate and right-hand apron symmetrical configuration, include horizontal canned paragraph from top to bottom, first vertical canned paragraph, first arcuate segment, second arcuate segment and the second vertical canned paragraph, described base is bilateral symmetry, upper end open, the width being wider than upper end open of lower end base plate, wherein, the left and right sides includes the 3rd vertical canned paragraph and tilting section, the upper end of described tilting section is fixedly connected with the lower end of the 3rd vertical canned paragraph, lower end is fixedly connected with one end of lower end base plate, described left side baffle plate and right-hand apron symmetry are placed, first vertical canned paragraph, second vertical canned paragraph, 3rd vertical canned paragraph is provided with fixing hole, the corresponding fixing hole of the left and right sides, fix by screw, the horizontal canned paragraph of both sides has at least one to be provided with fixing hole, described detector is fixed in the frame of described first arcuate segment formation, described light compensating apparatus is fixed in the frame of described second arcuate segment formation, described rainproof heat shield is connected with fixture by the fixing hole of described horizontal canned paragraph, ventilation gap is left between described rainproof heat shield and described probe body, second vertical canned paragraph of the described left and right sides is fixedly connected with by screw with the fixing hole of the 3rd vertical canned paragraph, described detector body is arranged in described probe body.
3. a kind of detector with integral structure according to claim 1, it is characterized in that, described detector body comprises sheet metal component, circuit board, camera lens, buttock line, described circuit board is arranged on described sheet metal component, described camera lens comprises the first camera lens and the second camera lens, described first camera lens is arranged on B/W camera front end, described first camera lens front end is provided with a filter by filter adapter ring, described second camera lens is arranged on colour TV camera front end, described first camera lens is all connected with described circuit board with the second camera lens, described light compensating apparatus body, comprise six electroluminescent lamps arranged in the form of a ring.
4. a kind of detector with integral structure according to claim 1, it is characterized in that, comprise tunnel pedestrian's intrusion detection module, described tunnel pedestrian's intrusion detection module comprises video acquisition unit, intelligent detection unit, memory cell, and video acquisition unit is connected with intelligent detection unit, display unit, memory cell; Intelligent detection unit is connected with video acquisition unit, display unit, memory cell, alarm unit; Intelligent detection unit and memory cell, display unit carry out Signal transmissions by fiber optic cables; Intelligent detection unit exports I/O signal to alarm unit.
5. a kind of detector with integral structure according to claim 4, is characterized in that, described video acquisition unit comprises colour TV camera, infrared illumination lamp, video compression coding module; Described intelligent detection unit comprises Background learning module, foreground detection module, invasion region setting module, invasion judge module and alarm module, picture compression coding module; Described memory cell comprises original video module and invasion picture module.
6. a kind of detector with integral structure according to claim 5, it is characterized in that, described intelligent detection unit, by Background learning module, set up image background model, and along with scene, the change of light etc., intelligence is carried out study and is upgraded, then foreground detection module is carried out, namely to people emerging in scene or thing, carry out position and feature detection, and real time record upgrades foreground information, namely invasion region setting module sets the region that in tunnel, pedestrian may occur, and then enter invasion judge module, by foreground information and invasion regional location and the comparison of the time of invasion, judge whether to belong to tunnel pedestrian's intrusion event, if belong to pedestrian's invasion, alarm module exports I/O alarm signal to alarm unit, simultaneously, locking pedestrian positional information, be transferred to display unit, and invasion picture module tunnel pedestrian's intrusion event picture is transferred to after picture compression coding module 26 carries out compressed encoding in memory cell, so that follow-up verification.
7. a kind of detector with integral structure according to claim 5, it is characterized in that, Background learning module in described intelligent detection unit, by carrying out Gauss's modeling to the image collected, utilize the Y of image, U, V component sets up the model library of image, adopt 2*2 region modeling point, each modeling point comprises four Gauss models, Gauss model comprises model gray scale, model variance, Model Weight, for the first two field picture, carry out model parameter initialization, for the image that subsequent transmission is come, mate with four Gauss models at corresponding pixel points place respectively, if coupling, then corresponding Gauss model weights increase, simultaneously and Gauss model is upgraded, if do not mate, then replace the minimum Gauss model of weight, when occurring that weights are greater than modeling threshold value in modeling point four Gauss models, current pixel point modeling success, the successful pixel number of modeling in statistics present image, when pixel modeling success count is greater than setting threshold, then start foreground detection, otherwise, proceed Background learning, modeling threshold value generally gets 1/5 ~ 1/3 of modeling point sum.
8. a kind of detector with integral structure according to claim 5, it is characterized in that, foreground detection module in described intelligent detection unit, detect emerging people or thing in scene, Gauss's modeling is carried out to the image that transmission comes, each modeling point comprises a Gauss model, the Gauss model of modeling point Gauss model with corresponding background modeling point place is compared, if do not mated, then continue to compare with modeling point four neighborhood Gauss model, if all do not mated, be then defined as foreground point, otherwise be background dot; Terminate if current frame image modeling point detects, then expansion process is carried out to foreground point, belong to the pixel of same prospect with UNICOM, simultaneously, prospect is marked, and calculates the filling rate of foreground blocks, area, center, depth-width ratio information, stored in prospect historical data base.
9. a kind of detector with integral structure according to claim 5, it is characterized in that, in described intelligent detection unit, invade region setting module, refer to the region forbidding occurring pedestrian in setting tunnel, by setting four region lines along tunnel edge, delineation invasion region.
10. a kind of detector with integral structure according to claim 5, is characterized in that, invade judge module in described intelligent detection unit, refers to and carries out invasion judgement to the prospect detected, due to judgement is the pedestrian information of invading, pedestrian's judgement is carried out to prospect, namely pedestrian is determine whether by foreground area, filling rate, depth-width ratio information, filling rate is the ratio of foreground pixel point and the prospect frame comprising the whole pixel of current prospect, and depth-width ratio is the ratio of the height and the width when the prospect frame comprising the whole pixel of current prospect, in tunnel pedestrian's intrusion detection, video acquisition unit is just installed tunnel, according to pedestrian's distance, the foreground area detected varies in size, the foreground area utilizing distance to detect is different, the i.e. difference of the depth of field, image is divided, be divided into three regions, each region prospect setting area maximin, when only having pedestrian to meet area constraints and depth-width ratio in corresponding region, then judge that current prospect is as pedestrian, this avoid the interference of too small fruit stone etc. simultaneously, it also avoid the mistake identification of dilly, improve recognition accuracy, after pedestrian's judgement terminates, center according to prospect changes, matched jamming is carried out to effective foreground information that prospect is behaved, and be less than rate of change threshold value further by the area change of pedestrian's prospect, the change of prospect average gray value is less than threshold value and carries out same prospect judgement restriction, if same prospect becomes, the match is successful, with the average gray value of prospect in current foreground information real-time update historical data base, filling rate, area, center, depth-width ratio information, the effective prospect disappeared in the visual field is removed, to the average gray value of emerging effective prospect, filling rate, area, center, depth-width ratio information is stored in prospect historical data base, alarm module in described intelligent detection unit, refers to when prospect pedestrian appears at protecting tunnel region, then sends I/O alarm signal, and meanwhile, alarm picture is stored in the invasion picture module in memory cell by picture compression coding module, in described intelligent detection unit, picture compression coding module is used for alarm picture compressed encoding, is convenient to store.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510595412.9A CN105100757B (en) | 2014-12-12 | 2015-09-18 | Detector with integral structure |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410757405X | 2014-12-12 | ||
CN201410757405.XA CN104469309A (en) | 2014-12-12 | 2014-12-12 | Tunnel pedestrian intrusion detection device and method |
CN201510595412.9A CN105100757B (en) | 2014-12-12 | 2015-09-18 | Detector with integral structure |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105100757A true CN105100757A (en) | 2015-11-25 |
CN105100757B CN105100757B (en) | 2018-09-18 |
Family
ID=52914584
Family Applications (3)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410757405.XA Pending CN104469309A (en) | 2014-12-12 | 2014-12-12 | Tunnel pedestrian intrusion detection device and method |
CN201510597878.2A Active CN105096603B (en) | 2014-12-12 | 2015-09-18 | Round-the-clock detection device |
CN201510595412.9A Active CN105100757B (en) | 2014-12-12 | 2015-09-18 | Detector with integral structure |
Family Applications Before (2)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410757405.XA Pending CN104469309A (en) | 2014-12-12 | 2014-12-12 | Tunnel pedestrian intrusion detection device and method |
CN201510597878.2A Active CN105096603B (en) | 2014-12-12 | 2015-09-18 | Round-the-clock detection device |
Country Status (1)
Country | Link |
---|---|
CN (3) | CN104469309A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105096603A (en) * | 2014-12-12 | 2015-11-25 | 天津艾思科尔科技有限公司 | All-weather detection device |
Families Citing this family (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105469604A (en) * | 2015-12-09 | 2016-04-06 | 大连海事大学 | An in-tunnel vehicle detection method based on monitored images |
CN105657266B (en) * | 2016-01-06 | 2019-04-09 | 山东康威通信技术股份有限公司 | The detection and tracking of a kind of tunnel mobile video to invasion personnel |
CN105632178B (en) * | 2016-03-02 | 2018-02-16 | 江苏中兴新泰物联网科技园有限公司 | A kind of wisdom traffic safety monitoring warning device |
CN107025436B (en) * | 2017-03-13 | 2020-10-16 | 西安电子科技大学 | Reliability-based self-updating personnel intrusion detection method |
CN107085729B (en) * | 2017-03-13 | 2021-06-22 | 西安电子科技大学 | Bayesian inference-based personnel detection result correction method |
CN107323114B (en) * | 2017-06-22 | 2019-08-16 | 珠海汇金科技股份有限公司 | Intrusion detection method, system and the print control instrument of print control instrument |
CN107451570A (en) * | 2017-08-04 | 2017-12-08 | 中控智慧科技股份有限公司 | A kind of method and device for starting palm pattern recognition device |
JP6875255B2 (en) * | 2017-11-09 | 2021-05-19 | 株式会社日立ビルシステム | Video monitoring system and video monitoring device |
CN108564751A (en) * | 2018-04-28 | 2018-09-21 | 国网重庆市电力公司 | The monitoring method of cable tunnel anti-intrusion, apparatus and system |
CN109166284A (en) * | 2018-09-11 | 2019-01-08 | 广东省电子技术研究所 | A kind of unlawful practice alarm system and unlawful practice alarm method |
CN109488383B (en) * | 2018-12-27 | 2020-11-03 | 江苏三恒科技股份有限公司 | Intelligent video analysis system and application method thereof |
CN110427865B (en) * | 2019-07-29 | 2023-08-25 | 三峡大学 | Human behavior video feature picture extraction and reconstruction method for high-voltage forbidden region |
CN111310733B (en) * | 2020-03-19 | 2023-08-22 | 成都云盯科技有限公司 | Personnel access detection method, device and equipment based on monitoring video |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004085810A (en) * | 2002-08-26 | 2004-03-18 | Nikon Corp | Auxiliary light projection device, camera and camera system |
CN202759522U (en) * | 2012-08-30 | 2013-02-27 | 常州硕欣电子科技有限公司 | High-speed spherical camera |
CN103200367A (en) * | 2013-04-15 | 2013-07-10 | 华为技术有限公司 | Video monitoring equipment and supplementary lighting device |
CN103489202A (en) * | 2013-01-15 | 2014-01-01 | 上海盈觉智能科技有限公司 | Intrusion detection method based on videos |
CN105096603A (en) * | 2014-12-12 | 2015-11-25 | 天津艾思科尔科技有限公司 | All-weather detection device |
CN205142403U (en) * | 2015-09-18 | 2016-04-06 | 天津艾思科尔科技有限公司 | Waterproof sealing type detecting device |
Family Cites Families (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1797496A (en) * | 2004-12-30 | 2006-07-05 | 汉王科技股份有限公司 | Movable type equipment for monitoring vehicles |
JP4894472B2 (en) * | 2006-11-15 | 2012-03-14 | 株式会社デンソー | Notification system |
JP4631036B2 (en) * | 2007-10-29 | 2011-02-16 | 東日本旅客鉄道株式会社 | Passer-by behavior analysis device, passer-by behavior analysis method, and program thereof |
CN202404757U (en) * | 2012-01-12 | 2012-08-29 | 石家庄汉邦科技有限公司 | Handheld portable green-channel vehicle detector |
CN203657881U (en) * | 2013-12-15 | 2014-06-18 | 郑州新力光电技术有限公司 | Light supplement detection device for high-capacity tank |
CN205142404U (en) * | 2015-09-18 | 2016-04-06 | 天津艾思科尔科技有限公司 | Detector with light filling device |
-
2014
- 2014-12-12 CN CN201410757405.XA patent/CN104469309A/en active Pending
-
2015
- 2015-09-18 CN CN201510597878.2A patent/CN105096603B/en active Active
- 2015-09-18 CN CN201510595412.9A patent/CN105100757B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2004085810A (en) * | 2002-08-26 | 2004-03-18 | Nikon Corp | Auxiliary light projection device, camera and camera system |
CN202759522U (en) * | 2012-08-30 | 2013-02-27 | 常州硕欣电子科技有限公司 | High-speed spherical camera |
CN103489202A (en) * | 2013-01-15 | 2014-01-01 | 上海盈觉智能科技有限公司 | Intrusion detection method based on videos |
CN103200367A (en) * | 2013-04-15 | 2013-07-10 | 华为技术有限公司 | Video monitoring equipment and supplementary lighting device |
CN105096603A (en) * | 2014-12-12 | 2015-11-25 | 天津艾思科尔科技有限公司 | All-weather detection device |
CN105096603B (en) * | 2014-12-12 | 2018-01-09 | 天津艾思科尔科技有限公司 | Round-the-clock detection device |
CN205142403U (en) * | 2015-09-18 | 2016-04-06 | 天津艾思科尔科技有限公司 | Waterproof sealing type detecting device |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105096603A (en) * | 2014-12-12 | 2015-11-25 | 天津艾思科尔科技有限公司 | All-weather detection device |
CN105096603B (en) * | 2014-12-12 | 2018-01-09 | 天津艾思科尔科技有限公司 | Round-the-clock detection device |
Also Published As
Publication number | Publication date |
---|---|
CN105096603A (en) | 2015-11-25 |
CN105100757B (en) | 2018-09-18 |
CN105096603B (en) | 2018-01-09 |
CN104469309A (en) | 2015-03-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105100757A (en) | Detector with integrated structure | |
CN111144291B (en) | Video monitoring area personnel intrusion discrimination method and device based on target detection | |
CN101918980B (en) | Runway surveillance system and method | |
CN107331097A (en) | The periphery intrusion preventing apparatus and method merged based on target position information | |
CN103986906B (en) | Door opening and closing detection method based on monitoring videos | |
KR101514696B1 (en) | Safety Accident Management System Of Bridge | |
CN104394361A (en) | Pedestrian crossing intelligent monitoring device and detection method | |
CN207164939U (en) | Periphery intrusion preventing device based on target position information fusion | |
CN210899299U (en) | Tunnel monitoring system | |
CN103426267A (en) | Video safety fence system based on video analysis technology | |
CN206136125U (en) | Radar video fuses alarm device , system and equipment | |
CN108764166B (en) | System and method for detecting position state of civil air defense door based on supporting rod | |
CN113206978B (en) | Intelligent monitoring and early warning system and method for security protection of oil and gas pipeline station | |
CN208208040U (en) | The anti-intrusion monitoring system of runway | |
CN104601966A (en) | Visual analysis visitors flow rate statistical method and visual analysis visitors flow rate statistical equipment based on laser detection lines | |
CN103592901A (en) | Holder control airport monitoring system | |
CN116311727A (en) | Intrusion response method, device, equipment and readable storage medium | |
CN210222962U (en) | Intelligent electronic fence system | |
CN103152558A (en) | Intrusion detection method based on scene recognition | |
CN115471965A (en) | Power transmission line area protection system and protection method | |
CN105208330A (en) | RFID person passageway monitoring equipment and method | |
KR102434154B1 (en) | Method for tracking multi target in traffic image-monitoring-system | |
CN203070472U (en) | Intelligent transformer station video security fence system | |
CN113222548A (en) | Community property eagle eye monitoring system and use method | |
KR102204015B1 (en) | Lifting gate based on image and vehicle control method through the same |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |