Design And Implement Of Embedded Mountainlandslide Monitoring System Based On Digital Image Process

China is a mountainous and mountain landslide disaster frequently happens in the country.Highly devastating landslide is a seriously threat to the safety of China's transport,people's lives and property.The landslide monitoring method,based on the digital image processing,has the characteristics of non-contact,which has been widely used in the field of landslide monitoring in recent years.Traditional monitoring methods,such as satellite remote sensing and close-range view photography and so on,have poor real-time and high cost.An embedded mountain landslide monitoring system based on the digital image processing principle·is designed in this paper,which improves the real-time performance,reduces the cost and makes the traditional monitoring methods better.The real-time monitoring of landslide is completed in this paper through the digital image acquisition,automatic identification and data uploading,which has great theoretical and practical significance for disaster prevention and reduction.The landslide monitoring system is designed in the paper based on sensor technology,embedded computer technology,modern communication technology,digital image processing principle,machine learning and pattern recognition principle.The system includes digital image acquisition module,digital image display module,master and digital image processing module,GPRS wireless network transmission module.OV5640 camera module and cloud platform module are used in the digital image acquisition module to carry out the digital image acquisition of the mountain from multi angles;The digital image display module can display the mountain image on the LCD digital image display module in real time;Master and digital image processing module takes the ARM processor,with the FPU and DSP instruction set Cortex-M4 architecture,as the core,and designs the corresponding.clock circuit,reset circuit,JTAG interface circuit,and the digital image collected by the digital image acquisition module is compressed stored in the SDRAM,and then according to the digital image processing algorithm,the feature segmentation of the image can be completed.Firstly,the color feature of the digital image is extracted in the digital image processing by the RGB domain color feature segmentation and the mean,variance and energy of the HSI field color histogram.Secondly,the gray level co-occurrence matrix and the corresponding algorithm are used to obtain the entropy,energy,inertial moments and local smoothness and other texture features.Finally,the mountain image features are input into the landslide state recognition model,and then the landslide status are accessible.The model is based on the support vector machine(SVM)machine learning algorithm,and the samples are constructed by normal mountains and landslide images library training;SIM900A module is used in GPRS wireless network transmission module,the color features,texture features and landslide status of the digital images can be sent to the server by the GPRS wireless network with the module,which can facilitate follow-up data research and information processing.With the system designed in the paper,the digital image acquisition,processing and data transmission are quick and efficient,and the landslide status can be monitored in real-time.The system designed in the paper is proved to be correct and effective by laboratory simulation test.The embedded mountain landslide monitoring system based on digital image processing designed in the paper can collect and monitor the landslide status in real time,stable and efficient under non-contact condition.The system has the advantages of good stability,low cost,simple structure and easy maintenance.The design goal is achieved after the theoretical research,algorithm design,hardware and software design and experimental debugging.The realization of the system provides some reference for the further study of landslide monitoring.