Face Recognition From Video Using Enhanced Social Collie Optimization-based Deep Convolutional Neural Network

The face recognition system is a process which is part of computer vision as a key feature of video surveillance systems. In a face recognition system stands on two main pillar which are object recognition and authentication of object or person who present in a frame with the help of distinctive picture generation like i.e. Cameras. In this field of Computer Vision many researchers work on detection and verification of faces from frames in the subsequent part of video. It is a complex and complicated process to recognize accurate faces by the use of artificial vision algorithms. In machine learning main important concept is Convolutional neural networks (CNNs) in real time system which has video process input for face recognition systems. For such complex problem effective approaches to work on by solution as CNNs have excelled capability to handle, with significant results. Which also including process based voice recognition, face localization, face recognition and picture frame selection and categorization. Recently all the top performing techniques which work on the preprocessed Labelled Face dataset for the Real Time system for convolutional neural networks (CNNs).So here in this research use of deep convolutional neural network (optimized Deep CNN) to getting generated precise outcome of face images from the input video by the Hybrid weighted texture pattern descriptor (HWTP). The help of deep CNN techniques with improved tuned featured extraction and introduce novel ideas about enhanced social collie optimization (ESCO), which result in better solutions with different combinations of strategies. The finest parameters are used to recognise the face of an individual is determined. The proposed model by deep convolutional neural network achieves 91.30% accuracy, 95.07% precision, 95.078% recall and 95.078% f-measures.

1. Introduction

The descriptor performs more effective than the real video, without the feature loss. Hence, contrasted to conventional technique, it is defined as the more extensive and elaborated video local information scales. The effective video description operator can be greatly aided by the accurate representation of the spatial as well as the time-based reliable information from videos, representing significant research value in perspective and analysis of video. As a result, numerous works have been discussed in this section along with remarkable advancements from diverse angles and a variety of problems. The fast and accurate face recognition through video data is highly demanding and crucial thing as increasing and demanding utilization of image video by CCTV and smart devices such as digital cameras or smartphone world-wide [1]. However, such complex issues are comparatively exciting and challenging, just because of different factors having dependency on quality of input. But right now just because of high computer vision development in some recent years, it has obtained good milestones. In reality, such a complex task is fragmented into smaller parts which are able to handle properly to achieve simpler and handy solutions. The main approach is to identify the face image frame from input video, then first task to relevant face detection in a picture and second face recognition. Meanwhile, some other activities were carried out, like localization of faces, validating faces or extraction of additional features and characteristics from them. There are multiple algorithms and methods to handle such complex tasks, like Eigen faces or Active Shape models, used rigorously and continuously just because of their outstanding result time. It has been found that one of the most popular and promising methods of Deep Learning (DL), principally the Convolutional Neural Networks. Now Convolutional Neural Networks yield most accurately results with continuous improvement. After looking at all the current situation and state of result, we selected to focus on our study approach to concentrate on more factors which help to improve the outcome quality in terms of time and accuracy. Additionally, it collected face images from high definition devices like cameras, effectively using deep learning techniques to concentrate on face localization, identification and recognition purposes, to improve performance of the overall system.