Research On Face Recognition Based On MapReduce

With the advent of the era of big data and intelligence,how to efficiently dig out the valuable information from face image data has become a hot topic in the field of face recognition.With the rapid development of cloud computing and distributed batch processing bring new ideas to face recognition.Traditional face recognition technology is only for small range,stationary state,and single face recognition.Usually,when the amount of data increases,the real-time performance is very low,so it can not be used in a large amount of data.In this paper,the range of application and the real-time of face recognition are explored,and a new idea of combining the existing face recognition algorithm with the batch processing framework MapReduce is proposed.The main contributions of this paper are as follows:1)In view of the small range of traditional face recognition applications,we propose to use large data storage technology Hadoop HDFS and HBase to store data.This will all be built in pictures and recognition of face images stored in HBase,the only representative of built-in picture information text file stored in HDFS,and then make the face recognition can be applied to a wider range of places.2)Aiming at the problem of low real-time performance of traditional face recognition,the idea of combining face recognition algorithm with batch MapReduce in Hadoop is proposed.Firstly,using Map to calculate the face recognition of PC A algorithm for Euclidean distance,obtained by processing the intermediate results.And then treated with Reduce to intermediate result,fmaly the built-in image information corresponding to the minimum Euclidean distance as the ultimate results and storage.In order to test the performance of face recognition system for improving,multiple groups of different number of built-in face image data to evaluate the real-time and scope of improvement of the model of face recognition,and has got a good result.