Font Size: a A A

Design And Implementation Of Child Face Recognition And Growth Prediction System

Posted on:2021-03-31Degree:MasterType:Thesis
Country:ChinaCandidate:F ZhouFull Text:PDF
GTID:2428330611962830Subject:Software engineering
Abstract/Summary:PDF Full Text Request
With the increasing national comprehensive strength of our country,the people's material and cultural life is getting richer and richer,family life is getting happier and happier,and the efforts and trust of parents are increasingly in their children.However,since the 1980 s,there have been many incidents of missing children,which has caused countless families to break down and caused a large number of parents to suffer.In view of the special national conditions in China,there are large number of people,and the uneven quality of the people,once the missing child is not found in the short time,the possibility of finding them in the long time is almost zero,because the growth of children's facial changes is huge,and once there is no accurate face Information,looking for them is like finding a needle in a haystack.At the same time every year,the Ministry of Public Security of China has successfully rescued some missing children,but due to the stagnation of information,these missing children are not well helped to find their native families.At present,China has a single method for finding lost children,and its efficiency is low.Posting information on the Internet needs to distinguish it between true and false in order to prevent being deceived,it is not easy to find a lot of manpower,material and financial resources in order to find the missing children offline.Moreover,charity pays less attention to the problem of missing children,but the addition of commercial platforms is easy to make people doubt that it has ulterior motives.Based on the social issues raised above,this article aims to use the existing face recognition and face prediction algorithms to build a child face recognition andgrowth prediction system for the recovery of lost children: predict the current appearance of children,then use face recognition technology to find the lost children.At the same time,the lost children can predict the appearance of his childhood as a child and use this to find their family members,so the success rate of the search will be improved,and the duration of the search can be well extended.Secondly,searching for information will not be spread widely on social networks,which can also reduce online fraud.The main contents of this article are as follows:1.In-depth understanding of the status quo of lost children in China in recent years,investigation of the current domestic and foreign lost children search platforms,analysis of the advantages and disadvantages of their recovery methods,and propose the use of face recognition and face prediction to find lost children,which is of great help to the long-term lost children search.2.Complete face recognition and face prediction,face recognition using MTCNN model and FaceNet model,complete the recognition of one-to-one and one-to-many photos,at the same time combined with platform usage scenarios,choose to use the Asian face FaceNet model exercise training data sets,compared with the official training model,a new training model on the identification accuracy of asians have improved.Conditional antagonistic automatic encoder(CAAE)is adopted to realize face aging and de-aging,at the same time,the movie star photos were collected and tested independently,and the experiment verified the reliability of the face prediction model.3.Make a comprehensive demand analysis and system design for the system based on the investigation of the existing missing children search platform,then encapsulate the face recognition and face prediction functions into a forum built by micro-framework flask based on Python,and complete the implementation and test of the system.In the process of paying attention to the growth of children,lost children were found in time is the basic topic.Through the research in this paper,a child face recognition and growth prediction system is finally realized,which can be used for the recovery of lost children,avoiding the loss of a large number of human,material and financial resources,and better build a standard,reasonable and effective platform forthe recovery of lost children.We will protect the safety of children,ensure the basic happiness of families,and safeguard social harmony.
Keywords/Search Tags:lost children found, Face recognition, Face prediction, BBS
PDF Full Text Request
Related items