Design And Implementation Of ARM-based Dam Mud Fluid Drainage Feature Detection System

With the deterioration of the environment,mud-rock flows frequently occur in various regions,resulting in the serious paralysis of traffic facilities such as highways and railways.It also causes a large number of casualties and great losses to the society.In order to reduce casualties and property damage,the government departments have built sand dams in most areas.When the bar is vented through the dam,the dynamic pressure generated by the overflow mud fluid will increase obviously,and the characteristics such as the transverse width of the body,the incidence angle and the body area of the downstream and the downstream will affect the structure of the dam body.It also will restrict the impact characteristics of the mud fluid,and the gutter bed cutting at the downstream of the bar.Due to the uncontrollable environmental factors when mud fluid is discharged,artificial real-time measurement is hindered.Therefore,how to use the convenient equipment and processing technology to detect the characteristics of mud fluid in the special environment has been an urgent problem and trend to be solved.Based on the indoor model,the system simulates the dam mud fluid drainage by using the soil-rock mixture quality.By using the image processing method,the thesis designs a convenient detection device,which performs feature region extraction and eigenvalue detection.In order to realize the detection system,it mainly studies the feature extraction and detection algorithm,system design scheme,system software and hardware and application program.The specific contents are as follows:First,according to the analysis of actual situation of the indoor model environment,the thesis puts forward the efficient and accurate is suitable for the characteristics of the testing process(such as a tub of removing,image preprocessing,feature region of the upper and lower edge,the edge of the area to remove and small area to remove algorithm).It also analyzes and verifies the algorithm used in the process to complete the feature extraction of the dam mud fluid drainage feature area,the lateral width of the feature area and the incident angle;Second,this thesis analyzes the overall structure and functional requirements of the system,according to the architecture and actual needs of the system hardware(such as processor,CMOS camera,touch screen,DC-DC DC converter module,storage module,communication module and level matching module,etc.)and the system software is selected to complete the overall design of the ARM-based dam mud fluid leakage feature detection system;Third,the system implements an embedded software platform on the designed system hardware.This thesis mainly includes the embedded cross-compilation environment needed for system development,and analyzes the startup principle and operation process and port of u-boot.It also discusses the source code structure and principle of embedded Linux,and studies how to implement embedded Linux kernel on ARM hardware platform.In this thesis,it makes a detailed study on the production and mounting methods of the file system,and introduces the OpenCV under the embedded system and the detailed source code compilation research.Finally,the image dependency library under the embedded system is generated and verified by program migration.Fourth,the GUI programming of Qt is used to realize the application of the detection system for the characteristics of the dam mud flow.This thesis studies the signal and slot mechanism in Qt programming technology,then analyzes the system requirements,designs and layouts the embedded system application interface.It also realizes the program to complete multiple interface switching and camera parameter measurement,correction and display.In this thesis,the feature area extraction and eigenvalue measurement,displaing data and data preservation and transmission functions of the dam mud fluid discharge image are obtained.The thesis also collects,previews and intercepts the dam mud fluid discharge process.Through the system design,the artificial repeated operation is reduced,and more dam mud fluid discharge characteristic value data can be stored in a short time.It is of great significance to provide theoretical support for the later research on the development form of the scour pit behind the debris flow dam,the design of the dam foundation and the selection of the post-dam protection project.