CN213813372U

CN213813372U - Pavement crack detection system

Info

Publication number: CN213813372U
Application number: CN202021722335.1U
Authority: CN
Inventors: 李鹏; 谭俨; 史峰
Original assignee: Binjiang College of Nanjing University of Information Engineering
Current assignee: Binjiang College of Nanjing University of Information Engineering
Priority date: 2020-08-17
Filing date: 2020-08-17
Publication date: 2021-07-27
Anticipated expiration: 2030-08-17

Abstract

The utility model discloses a pavement crack detection system, which comprises a positioning module, an image acquisition and display module, a main control module connected with the positioning module and the image acquisition and display module and used for acquiring crack positions and identifying crack conditions, and a remote service module connected with the main control module and used for storing crack images and carrying out secondary judgment on the crack images; the main control module is connected with the remote service module through a 5G communication module. The utility model carries out secondary recognition to the primary judged picture of the main control module through the neural network of the remote service module, further improving the accuracy of detection; the pavement crack detection system has the advantages that pavement videos or image data do not need to be transmitted to a server for judgment and processing, the main control module directly identifies cracks and crack types of collected pavement images, and pavement crack information can be timely, effectively and accurately detected, and the crack types can be judged.

Description

Pavement crack detection system

Technical Field

The utility model relates to a detecting system especially relates to a road surface crack detecting system.

Background

With the development of economy, highway construction is rapidly developed. Meanwhile, sudden increase, overload, heavy vehicles, driving habits and the like of the vehicles can damage the road and the ground, so that the ground is damaged due to overload and other reasons. Cracks are the most common and dangerous one of the road surface diseases, and if the cracks cannot be repaired in time, the further damage of the road surface can be caused to a great extent, so that the driving safety is influenced.

At present, a detection system for pavement cracks mainly comprises an image recognition method, an ultrasonic detection method and an optical detection method. The image recognition method is characterized in that a camera is used for shooting the ground, then shooting data are sent to a server, received data are recognized and judged by using the strong processing capacity of the server, time is consumed, huge storage space is consumed, and large errors exist when the positions of cracks are judged. Ultrasonic detection and optical detection are environmentally critical. It can be seen that there is still much room for development of pavement crack detection systems.

SUMMERY OF THE UTILITY MODEL

Utility model purpose: the utility model aims at providing a can be timely, effective, accurate detect road surface crack positional information and judge the road surface crack detecting system of crack type.

The technical scheme is as follows: the utility model discloses a pavement crack detection system, including orientation module, image acquisition and display module, still include with orientation module, image acquisition and display module are connected and are used for acquireing the main control module of crack position and discernment crack condition, with main control module is connected and is used for storing the crack image and carry out the long-range service module of secondary judgement to the crack image; the main control module is connected with the remote service module through a 5G communication module.

Preferably, the communication module communicates with the main control module through a network port or a USB interface.

Preferably, the positioning module is a Beidou positioning module.

Preferably, the main control module uses an EP4CE30F23C8N FPGA chip as a main control chip.

Preferably, the image acquisition and display module comprises an image acquisition module and an LCD display module.

Preferably, the image acquisition module is a CMOS image sensor.

Preferably, the positioning module has a positioning accuracy of less than 5m at a speed of less than 80 km/h.

Has the advantages that: compared with the prior art, the utility model, gain as follows and show the effect: 1. the secondary recognition is carried out on the primary judged picture of the main control module through the neural network of the remote service module, and the detection accuracy is further improved. 2. Need not to handle on passing video or image data to the server, the utility model discloses a main control module directly carries out the discernment of crack and crack kind to the crack image of gathering. 3. The utility model discloses a road surface crack detecting system based on vision and neural network provides a new means for road surface crack detection area. 4. The system appearance is small and exquisite portable, is fit for installing on all kinds of removal automobile bodies.

Drawings

Fig. 1 is a block diagram of the system structure of the present invention;

FIG. 2 is a schematic view of the working process of the present invention;

fig. 3 is a schematic view of the installation structure of the present invention.

Detailed Description

The technical solution of the present invention is further explained with reference to the drawings of the specification.

As shown in fig. 1, the utility model discloses a road surface crack detecting system includes host system 2, and host system 2 adopts the EP4CE30F23C8N FPGA chip of the cycle IV series of Altera as host system chip, and this chip has abundant inside programmable logic cell and memory cell, and these all provide the guarantee on the performance for entire system's function. The main control chip is mainly used for sending/receiving instructions and processing data, the FPGA chip is controlled to send instructions to each module in a programming mode, each module is controlled to collect information, and meanwhile, each module sends the collected information to the FPGA for processing. A positioning module 4 and an image acquisition and display module 6 which are respectively connected with the main control module 2.

The positioning module 4 is a Beidou positioning module 4, the module is obtained by adopting a domestic Taidou D303 Beidou positioning chip, and the positioning accuracy of the chip is less than 5m under the condition of being lower than 80km/h, so that the requirement of the positioning accuracy is completely met.

The image acquisition and display module 6 comprises an image acquisition module 5 and a display module 6. The image acquisition module 5 adopts a CMOS image sensor with the model of OV7725 to acquire road surface image information. The CMOS sensor is directly connected with the FPGA through an external I/O port of the FPGA, the sensor supplies power by using 3.3V, relevant parameters of the camera and an I2C protocol need to be configured before the sensor is used, and the camera can normally work after the configuration is completed. The display module 6 is an LCD (liquid crystal display) module 6 which is mainly used for displaying the crack types and the longitude and latitude information of the crack positions, and can realize the real-time information display on a lower computer.

The main control module 2 is connected with a remote service module 11 for storing the crack images and performing secondary identification on the crack images. The remote service module 11 comprises a remote server, the remote server is not physically connected with the modules, data exchange is carried out only through 5G communication, and meanwhile the remote server needs to carry out secondary identification on the crack image judged by the lower computer, so that the accuracy of system detection is improved.

The main control module 2 is connected with a remote server through a communication module 3 and is used for transmitting the crack information acquired by the main control module 2 to the remote server. The communication module 3 is a 5G communication module 3, the module is connected with the main control chip through an Ethernet port and is used for transmitting the crack images in real time, and the transmission efficiency is improved by using a 5G transmission technology, so that the pressure caused by image transmission is reduced in the operation process of the system.

As shown in fig. 2, after the image data is acquired by the camera in the image acquisition module 5, the data is sent to the SRAM after gray level conversion, median filtering, Sobel edge detection and morphological dilation corrosion, and the projection features are extracted, identified and classified. Meanwhile, the classification result is sent to the LCD display module 6 through the serial port for display. The Beidou positioning module 4 also transmits longitude and latitude information to the LCD through a serial port to be displayed, and the 5G transmission module in the system is used for communicating the connection between the lower computer and the remote server, and mainly uploads an image which is judged to be a crack by the lower computer to the remote server for secondary identification through a neural network. And the FPGA transmits the latitude and longitude information stored in the SRAM and the image information judged to be cracks to the 5G communication module 3 through the Ethernet port by controlling the FIFO to carry out wireless transmission of data.

As shown in FIG. 3, in actual use, can place host system 2, orientation module 4, communication module 3 in box 1, place in box 1 the utility model discloses a road surface crack detecting system provides power module 8 of electric power. A hollow metal support 9 is arranged above the box body 1, the hollow metal support 9 is connected with the image acquisition module 5 through an angle adjusting knob 10, and the angle of the image acquisition module 5 can be freely adjusted in the range of being parallel to the ground and perpendicular to the ground through the angle adjusting knob 10. Display module 6 installs in the box 1 outside, and display module 6 can closely laminate with box 1, also can install in other suitable positions through the extension line, if: when the device is installed on a long-distance bus, the display module 6 can be installed near a driving instrument panel through an extension line, and the display module 6 is directly and physically connected with the main control module 2. The box body 1 is connected with an antenna 7. The communication module 3, the positioning module 4, the image acquisition module 5, the display module 6 and the antenna 7 are respectively physically connected with the main control module 2 through data lines, and are controlled by the main control module 2 to execute various operations and receive and process information transmitted by each module and system. The remote server carries out remote communication with the modules through the communication module 3 without physical contact connection.

Claims

1. A pavement crack detection system comprises a positioning module (4) and an image acquisition and display module (6), and is characterized by further comprising a main control module (2) connected with the positioning module (4) and the image acquisition and display module (6) and used for acquiring the crack position and identifying the crack condition, and a remote service module (11) connected with the main control module (2) and used for storing a crack image and performing secondary judgment on the crack image; the main control module (2) is connected with the remote service module (11) through a 5G communication module.

2. The pavement crack detection system according to claim 1, characterized in that the communication module (3) communicates with the main control module (2) through a network port or a USB interface.

3. The pavement crack detection system according to claim 1, characterized in that the positioning module (4) is a Beidou positioning module (4).

4. The pavement crack detection system according to claim 1, characterized in that the main control module (2) uses an EP4CE30F23C8N FPGA chip as a main control chip.

5. The pavement crack detection system according to claim 1, characterized in that the image acquisition and display module (6) comprises an image acquisition module and an LCD display module (6).

6. The pavement crack detection system of claim 5, wherein the image capture module is a CMOS image sensor.

7. Pavement crack detection system according to claim 1, characterized in that the positioning accuracy of the positioning module (4) is less than 5m at a speed below 80 km/h.