Research On Digging Event Detection Based On Acceleration Sensor

Ancient tombs are the treasures of our country's cultural heritage.They play a significant role in studying many aspects of ancient society,restoring historical facts,understanding and inheriting the outstanding cultural and cultural relics of our country.It's difficult to protect because of its bad location,quantity and wide distribution.At present,the acts of grave robbery in our country are very rampant,causing huge losses to the cultural value and economic value of the ancient tombs.It has been discovered that most of the lawless elements steal the cultural relics by digging into the surface and destroying the grave.In view of this situation,this paper presents a mining event detection algorithm running on embedded devices,which can monitor real-time excavation events around the tomb and detect the tomb robbery behavior in timeThe specific work of our paper is as follows:At first,this article elaborates the current research status and the key technologies of anti-tomb system at home and abroad,and analyzes its current situation and future.It is found that there are many problems in the study of detection of excavation events at home and abroad.All kinds of anti-tomb system have not realized the complete behavior detection function.Then,this paper describes the role and significance of the application of mining event detection algorithm in the process of ancient tombs protection.A high precision mining behavior detection algorithm based on acceleration information is proposed.Finally,taking the detection precision of the mining event as the index,the simulation experiment of the mining event detection algorithm based on the acceleration sensor is carried out.The simulation data show that the acceleration event detection algorithm based on accelerometer has high detection accuracy,and it can be applied to the practice of tomb detection and protection.At last,in this paper,a power optimization scheme based on Cortex-M4 microprocessor is proposed,which reduces frequent I/O operations with ADXL345 sensors,improves the utilization ratio of CPU by using asynchronous I/O operation,and reduces the complexity of the algorithm with naive impulse detection.Finally,the performance and prediction accuracy before and after optimization are compared by simulation experiments.The results show that the optimized algorithm keeps the accuracy while reducing the power consumption greatly.