CN213814722U

CN213814722U - AdaBoost algorithm-based highway tunnel fire initial-stage video flame detection device

Info

Publication number: CN213814722U
Application number: CN202120149788.8U
Authority: CN
Inventors: 叶津凌; 谢静思; 孟丛丛; 刘嘉华
Original assignee: Jiangxi Vocational and Technical College of Communication
Current assignee: Jiangxi Vocational and Technical College of Communication
Priority date: 2021-01-20
Filing date: 2021-01-20
Publication date: 2021-07-27
Anticipated expiration: 2031-01-20

Abstract

The utility model discloses a highway tunnel fire initial stage video flame detection device based on AdaBoost algorithm, including the detection device body, the detection device body is including detecting the box and detecting the base, the below that detects the base is equipped with the supporting legs, the below of supporting legs is equipped with the gyro wheel, it includes first cavity to detect the box, the second cavity, the third cavity, the fourth cavity, the fifth cavity, sixth cavity and seventh cavity, the side of detection device body is equipped with camera device one, camera device two, camera device three and camera device four, the inside of first cavity is equipped with step motor one, the inside of second cavity is equipped with lifting threaded rod, the inside of third cavity is equipped with data transmission device, execution device one, digital video recorder one, data processing apparatus, the inside of seventh cavity is equipped with AdaBoost algorithm controlling means and battery. The utility model discloses a set up camera device and mobilizable camera device of co-altitude not, improved the detection efficiency at conflagration initial stage.

Description

AdaBoost algorithm-based highway tunnel fire initial-stage video flame detection device

Technical Field

The utility model belongs to the technical field of fire identification, concretely relates to highway tunnel fire initial stage video flame detection device based on AdaBoost algorithm.

Background

With the application of security video monitoring systems in various fields and buildings in a large quantity, the fire image recognition technology is paid attention and researched by people, and compared with the traditional disaster detection technology, the visual detection has the advantages of large detection area, short response time, rich and visual information, low maintenance cost and the like.

Deep learning based on a convolutional neural network is successfully applied in the fields of character identification, face identification and the like, and a fire identification method based on the convolutional neural network is gradually applied, but because the deep learning needs a large number of data sets, the data sets used by the deep learning are large fire pictures searched in Google and hundred degrees, the difference from the actual application scene is large, image samples are few, the image background is single, and an interference source lacks diversity, so that the accuracy of fire detection cannot meet the requirement.

At present, the existing detection device can only detect and identify the fire through a single camera, so as to identify and respond to the fire, the method cannot identify a plurality of angles, and the situations that the fire points with different heights are not identified in place exist, so that a road tunnel fire initial stage video flame detection device based on an AdaBoost algorithm is urgently needed.

SUMMERY OF THE UTILITY MODEL

The utility model provides a based on AdaBoost algorithm highway tunnel conflagration initial stage video flame detection device through setting up camera device and the mobilizable camera device of co-altitude, has improved the detection efficiency in conflagration initial stage.

The technical scheme of the utility model is realized like this:

a highway tunnel fire initial-stage video flame detection device based on an AdaBoost algorithm comprises a detection device body and is characterized in that the detection device body comprises a detection box body and a detection base, supporting legs are arranged below the detection base, idler wheels are arranged below the supporting legs, the detection box body comprises a first cavity, a second cavity, a third cavity, a fourth cavity, a fifth cavity, a sixth cavity and a seventh cavity, a first camera device, a second camera device, a third camera device and a fourth camera device are arranged on the side surface of the detection device body, a first stepping motor is arranged inside the first cavity, a lifting threaded rod is arranged inside the second cavity, a first data transmission device, a first execution device, a first hard disk video recorder and a data processing device are arranged inside the third cavity, a second hard disk video recorder and a second execution device are arranged inside the fourth cavity, the hard disk video recorder III and the execution device III are arranged in the fifth cavity, the hard disk video recorder IV and the execution device IV are arranged in the sixth cavity, and the AdaBoost algorithm control device and the storage battery are arranged in the seventh cavity.

The utility model discloses an among the highway tunnel conflagration initial stage video flame detection device based on AdaBoost algorithm, the inside of detecting the base is equipped with places the cavity, the cavity is placed to being equipped with of placing the cavity, place cavity two, place cavity three, place cavity four, place cavity five, and place cavity one and place and be equipped with baffle one between the cavity two, place cavity two and place and be equipped with baffle two between the cavity three, it is equipped with baffle three to place cavity three and place cavity four, it is equipped with baffle four to place cavity four and place cavity five between, it is made to detect base, baffle one, baffle two, baffle three, baffle four to be metal material.

The utility model discloses an among the highway tunnel conflagration initial stage video flame detection device based on AdaBoost algorithm, be equipped with bee calling organ and light on the top of detection device body, the light is equipped with light one and light two, bee calling organ sets up between light one and light two.

The utility model discloses an among the highway tunnel conflagration initial stage video flame detection device based on AdaBoost algorithm, lifting threaded rod is equipped with screw thread section and straightway, the straightway inserts to the inside of first cavity, and is equipped with bevel gear one on the straightway, be equipped with bevel gear two on step motor one's the output shaft, bevel gear one and bevel gear two intermeshing.

The utility model discloses an among the highway tunnel conflagration initial stage video flame detection device based on AdaBoost algorithm, camera device one's bottom is equipped with the thread bush, and the bottom of thread bush is equipped with the dovetail, be equipped with the dovetail on the inside wall of second cavity, the thread bush is fixed on the screw thread section on the lifting screw pole, and the dovetail imbeds to the dovetail on.

The utility model discloses an among the highway tunnel fire initial stage video flame detection device based on AdaBoost algorithm, camera two sets up on the lateral wall of fourth cavity, camera three sets up on the lateral wall of fifth cavity, camera four sets up on the sixth cavity.

The utility model discloses a based on among AdaBoost algorithm highway tunnel conflagration initial stage video flame detection device, be electric connection between battery, light one, light two, camera device one, camera device two, AdaBoost algorithm controlling means, data processing apparatus, data transmission device, digital video recorder one, digital video recorder two, digital video recorder three, the digital video recorder four.

Implement the utility model discloses a this kind of based on AdaBoost algorithm highway tunnel conflagration initial stage video flame detection device has following beneficial effect: the present case sets up the detection box, and the inside that will detect the box is equipped with the lift threaded rod that can rotate, camera device one can carry out free lift according to step motor's operation, also can be through camera device two of co-altitude, camera device three and camera device four detect the characteristics at conflagration initial stage, this problem that traditional conflagration check point is fixed and can not detect different angles and co-altitude has just been solved, the gyro wheel setting that detects the base bottom makes the removal that can be portable carrying out check out test set, the overall process that artificial control conflagration initial stage detected like this, the device is operated very portably, the detection that is particularly suitable for the conflagration is used.

Drawings

Fig. 1 is a schematic diagram of the video flame detection device for the early stage of the fire of the highway tunnel based on the AdaBoost algorithm of the utility model;

FIG. 2 is an enlarged view of a portion of FIG. 1 at A;

FIG. 3 is an enlarged view of a portion of FIG. 1 at B;

fig. 4 is a partially enlarged schematic view of C in fig. 1.

In the figure: the detection device comprises a detection device body 1, a detection box body 2, a detection base 3, supporting legs 4, rollers 5, a first cavity 6, a second cavity 7, a third cavity 8, a fourth cavity 9, a fifth cavity 10, a sixth cavity 11, a seventh cavity 12, a first camera device 13, a second camera device 14, a third camera device 15, a fourth camera device 16, a first stepping motor 17, a lifting threaded rod 18, a data transmission device 19, a first execution device 20, a first hard disk video recorder 21, a data processing device 22, a second hard disk video recorder 23, a second execution device 24, a third hard disk video recorder 25, a third execution device 26, a fourth hard disk video recorder 27, a fourth execution device 28, an AdaBoost algorithm control device 29, a storage battery 30, a first placement cavity 31, a second placement cavity 32, a third placement cavity 33, a fourth placement cavity 34, a fifth placement cavity 35, a first partition 36, a second partition 37, a third partition 38, a fourth partition 39, a third partition, The device comprises a buzzer 40, a first illuminating lamp 41, a second illuminating lamp 42, a threaded section 43, a straight section 44, a first conical gear 45, a second conical gear 46, a threaded sleeve 47, a dovetail block 48 and a dovetail groove 49.

Detailed Description

The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

The AdaBoost algorithm-based highway tunnel fire initial-stage video flame detection device of the invention as shown in FIGS. 1 to 4 comprises a detection device body 1, wherein the detection device body 1 comprises a detection box body 2 and a detection base 3, a supporting foot 4 is arranged below the detection base 3, a roller 5 is arranged below the supporting foot 4, the roller 5 is convenient for the detection device body to move conveniently, the detection box body 2 comprises a first cavity 6, a second cavity 7, a third cavity 8, a fourth cavity 9, a fifth cavity 10, a sixth cavity 11 and a seventh cavity 12, a first camera device 13, a second camera device 14, a third camera device 15 and a fourth camera device 16 are arranged on the side surface of the detection device body 1, the first camera device 13, the second camera device 14, the third camera device 15 and the fourth camera device 16 are arranged to monitor a fire from different angles and different heights, the first cavity 6 is internally provided with a first stepping motor 17, the first stepping motor 17 rotates forwards and backwards to ensure that the first camera device 13 can freely ascend and descend, the second cavity 7 is internally provided with a lifting threaded rod 18, the third cavity 8 is internally provided with a data transmission device 19, an execution device 20, a first hard disk video recorder 21 and a data processing device 22, the data transmission device 19 and the data processing device 22 are both in the prior art, the fourth cavity 9 is internally provided with a second hard disk video recorder 23 and an execution device 24, the execution device is an existing execution system device, for the prior art, the fifth cavity 10 is internally provided with a third hard disk video recorder 25 and a third execution device 26, the sixth cavity 11 is internally provided with a fourth hard disk video recorder 27 and an execution device 28, the seventh cavity 12 is internally provided with an AdaBoost algorithm control device 29 and a storage battery 30, and the AdaBoost algorithm control device 29 is arranged to identify the color of an external fire, thereby ensuring higher identification efficiency, the detection base 3 is internally provided with a placing cavity, the placing cavity is provided with a placing cavity I31, a placing cavity II 32, a placing cavity III 33, a placing cavity IV 34 and a placing cavity IV 35, a partition plate I36 is arranged between the placing cavity I31 and the placing cavity II 32, a partition plate II 37 is arranged between the placing cavity II 32 and the placing cavity III 33, a partition plate III 38 is arranged between the placing cavity III 33 and the placing cavity IV 34, a partition plate IV 39 is arranged between the placing cavity IV 34 and the placing cavity IV 35, the detection base 3, the partition plate I36, the partition plate II 37, the partition plate III 38 and the partition plate IV 39 are all made of metal materials, the top of the detection device body 1 is provided with a buzzer 40 and an illuminating lamp, the buzzer 40 meets the alarm requirement, the illuminating lamp I41 and the illuminating lamp II 42 are arranged, the illuminating lamp I41 and the illuminating lamp II 42 meet the requirement of night identification, the buzzer 40 is arranged between the first illuminating lamp 41 and the second illuminating lamp 42, the lifting threaded rod 18 is provided with a threaded section 43 and a linear section 44, the linear section 44 is inserted into the first cavity 6, the linear section 44 is provided with a first conical gear 45, the output shaft of the first stepping motor 17 is provided with a second conical gear 46, the first conical gear 45 is meshed with the second conical gear 46, the first camera 13 can be lifted freely, the bottom of the first camera 13 is provided with a threaded sleeve 47, the bottom of the threaded sleeve 47 is provided with a dovetail block 48, the inner side wall of the second cavity 7 is provided with a dovetail groove 49, the threaded sleeve 47 is fixed on the threaded section 43 on the lifting threaded rod 18, the dovetail block 48 is embedded into the dovetail groove 49, the second camera 14 is arranged on the outer side wall of the fourth cavity 9, the third camera 15 is arranged on the outer side wall of the fifth cavity 10, the fourth camera 16 is arranged on the sixth cavity 11, the storage battery 30, the first illuminating lamp 41, the second illuminating lamp 42, the first camera device 13, the second camera device 14, the AdaBoost algorithm control device 29, the data processing device 22, the data transmission device 19, the first hard disk video recorder 21, the second hard disk video recorder 23, the third hard disk video recorder 25 and the fourth hard disk video recorder 27 are all electrically connected.

The utility model discloses a theory of operation does: firstly, move the detection device body to the position department that needs to carry out fire detection, then start step motor one for camera device body moves to the height that needs wait to detect, then start the battery, when camera device one, camera device two, camera device three and camera device four detected the fire point, transmit the buzzer behind camera device, data transmission device, the final controlling element with the signal of telecommunication by AdaBoost algorithm controlling means, send the chimes of doom by the buzzer, thereby the detection of conflagration has been accomplished. Thus, the object of the present invention can be achieved.

The above description is only a preferred embodiment of the present invention, and should not be taken as limiting the invention, and any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A highway tunnel fire initial-stage video flame detection device based on an AdaBoost algorithm comprises a detection device body and is characterized in that the detection device body comprises a detection box body and a detection base, supporting legs are arranged below the detection base, idler wheels are arranged below the supporting legs, the detection box body comprises a first cavity, a second cavity, a third cavity, a fourth cavity, a fifth cavity, a sixth cavity and a seventh cavity, a first camera device, a second camera device, a third camera device and a fourth camera device are arranged on the side surface of the detection device body, a first stepping motor is arranged inside the first cavity, a lifting threaded rod is arranged inside the second cavity, a first data transmission device, a first execution device, a first hard disk video recorder and a data processing device are arranged inside the third cavity, a second hard disk video recorder and a second execution device are arranged inside the fourth cavity, the hard disk video recorder III and the execution device III are arranged in the fifth cavity, the hard disk video recorder IV and the execution device IV are arranged in the sixth cavity, and the AdaBoost algorithm control device and the storage battery are arranged in the seventh cavity.

2. The AdaBoost algorithm-based highway tunnel fire initial-stage video flame detection device as claimed in claim 1, wherein a placing cavity is arranged inside the detection base, a first placing cavity, a second placing cavity, a third placing cavity, a fourth placing cavity and a fifth placing cavity are arranged in the placing cavity, a first partition plate is arranged between the first placing cavity and the second placing cavity, a second partition plate is arranged between the second placing cavity and the third placing cavity, a third partition plate is arranged between the third placing cavity and the fourth placing cavity, a fourth partition plate is arranged between the fourth placing cavity and the fifth placing cavity, and the detection base, the first partition plate, the second partition plate, the third partition plate and the fourth partition plate are all made of metal materials.

3. The AdaBoost algorithm-based highway tunnel fire initial-stage video flame detection device as claimed in claim 1, wherein a buzzer and a lighting lamp are arranged on the top of the detection device body, the lighting lamp is provided with a first lighting lamp and a second lighting lamp, and the buzzer is arranged between the first lighting lamp and the second lighting lamp.

4. The AdaBoost algorithm-based highway tunnel fire initial-stage video flame detection device according to claim 1, wherein the lifting threaded rod is provided with a threaded section and a straight section, the straight section is inserted into the first cavity, the straight section is provided with a first conical gear, an output shaft of the first stepping motor is provided with a second conical gear, and the first conical gear and the second conical gear are meshed with each other.

5. The AdaBoost algorithm-based highway tunnel fire initial-stage video flame detection device as claimed in claim 4, wherein a threaded sleeve is arranged at the bottom of the first camera device, a dovetail block is arranged at the bottom of the threaded sleeve, a dovetail groove is arranged on the inner side wall of the second cavity, the threaded sleeve is fixed on a threaded section on the lifting threaded rod, and the dovetail block is embedded into the dovetail groove.

6. The AdaBoost algorithm-based highway tunnel fire initial stage video flame detection device according to claim 1, wherein the second camera device is arranged on an outer side wall of the fourth cavity, the third camera device is arranged on an outer side wall of the fifth cavity, and the fourth camera device is arranged on the sixth cavity.

7. The road tunnel fire initial-stage video flame detection device based on the AdaBoost algorithm of any one of claims 1 to 6, wherein the storage battery, the first illuminating lamp, the second illuminating lamp, the first camera device, the second camera device, the AdaBoost algorithm control device, the data processing device, the data transmission device, the first hard disk video recorder, the second hard disk video recorder, the third hard disk video recorder and the fourth hard disk video recorder are all electrically connected.