Wake Mechanism Of Wireless Sensor Networks Based On Machine Vision

In recent years,with the continuous progress of information technology,image based wireless sensor network application technology has become a hotspot research topic.In the image and video data transmission,pay data transmission energy cost will be greatly improved,the traditional sleep scheduling mechanism to reduce energy consumption will no longer be effective,how to ensure the sensor to minimize the transmission cost can satisfy the value of image data transmission to the Sink node timely and effectively becomes the focus of this research.This paper focus on regional monitoring as the application background,wake-up mechanism of image retrieval based on research of wireless sensor network application.The traditional approach is to man-made sleeping intervention to reduce the consumption of the transmission of sensor network artificially,retrieval approach is to awaken data independent judgment by the sensor network itself,in order to calculate the cost for storage and transmission of energy consumption.The proposed sensor wakeup mechanism based on image retrieval provides a new technology application,and applies depth learning and machine vision technology to wireless sensor networks.The main work is as follows:First of all,the Hash algorithm,put forward preliminary sensor node wake-up mechanism,nodes collect data after processing into Hash series,with the background image Hash sequence similarity matching preliminary determine whether a target,determine the target appears,the sensor nodes wake up the communication module,the fingerprint sequence is transmitted to SINK Hash node.Secondly,study the image difference method,using the difference method to complete the acquisition of the target image data segmentation,using the 3DMAX software to simulate the attitude and target scene,complete the establishment of image database to target different attitude and scenarios,and extraction of image features,to establish the GIST feature library built based on target image database and Cifar-10 image library.Third,the research of image indexing technology,focus on the LSH and ITQ image retrieval algorithm,clear evaluation criteria,finally completes the analysis of the above three parts of the simulation,and by comparing the results,analysis of the application in wireless sensor network energy saving aspects.