Design And Implementation Of Image Management System Based On Face Clustering

The project is carried out to satisfy requirements of the internship company for rapid construction of training data based on face clustering and the scientific management of existing face image data.This system is designed for the management of daily training data of employees within the company,the construction of training sets and other aims.It is necessary to meet the needs of employees for system functions.The system should have the following three aspects of function: Firstly,supporting the uploading and downloading of image files.Secondly,processing both video files and image files.Thirdly,managing already processed images.Today,the improvement of computer vision is the hot topic and frontier of research in the field of artificial intelligence.Under the existing conditions and environment of the company,the paper aims to analyze the design and implementation of the system by referring to the original technology about image management system and some research focusing on face clustering.First of all,the paper describes the source and practical significance of the project,and reveal the urgency of applying the face image management system to the actual needs of the company.Then,the whole development process of the image management system based on face clustering is introduced from the aspects of related technology,feasibility analysis,requirements analysis,system design,implementation and testing by utilizing methods of software engineering.By adopting the B/S architecture and using the SSH as the framework for background development,the system is combined with human face detection and features extraction to uniformly encodes the extracted features into a file.Next,the clustering algorithm is employed to group and store different pictures of the same person together,thus the user can construct a new data set according to the information of the existing data set that has been constructed as needed.Through independent development and testing of each functional module as well as the final system integration,the test is completed.At last,the system can be applied normally.