Genetic and Ant Colony Algorithms for Face Recognition and Detection Systems
Dr C.Sunil Kumar, C.N. Ravi, J Dinesh · IOSR Journal of Computer Engineering · 2014
Face detection is one of the challenging problems in the digital image processing.Digital images have an enormous information and characteristics measures.But until today, a complete capable mechanism to extract these characteristics in an automatic way is so far unknown.Referring to facial images, its detection in an image is a problem that requires a thorough investigation due to its high complexity.Face detection is an important application of visual object detection and it is one of the main components of face analysis and understanding with face localization and face recognition.Here the investigation aspects of genetic Algorithms (GA's) in face recognition are characterized as one of search technique.GA is efficient technique in reducing computational time for a huge stack space.Face recognition from a very huge stack space is a time consuming job hence GA based approach is used to recognize the unidentified image within a short duration of time.This work analyzes the work done by distinguished authors previously and throws light on what next is to be done.Though they don't give exact and accurate results but are very efficient in time bound recognition for very huge databases.For promptness on a random pickup base it gives fastest result.GA is used when user has no time or less time for giving results without going for check related to each database containing facial images.Feature extraction along with GA will prove better for quicker face recognition.The basic plan of face detection is to determine if there is any face in an image and then locate a position of face in an image.Human face detected in an image can represent the presence of a human in a place.Obviously, face detection is the first step towards creating an automated system, which may involve other face processing.A novel face detection system is presented in this work.Early face-detection algorithms focused on the detection of frontal human faces, whereas newer algorithms attempt to solve the more general and difficult problem of multi-view face detection.That is, the detection of faces that is either rotated along the axis from the face to the observer or rotated along the vertical or left-right axis or both.The newer algorithms take into account variations in the image by factors such as face appearance, lighting, and pose.The goal of face detection is to detect human faces in still images or videos, in diverse situations.A global overview focuses on a detector which processes images very quickly while achieving high detection rates.In certain applications face detection may require advanced classification methods that would precisely identify the faces even if the face visibility measure is less.This work proposes two methods, first is an edge detection technique.Where it establishes a pheromone matrix that represents the edge information at each pixel based on the routes formed by the Ant Colony Genetic Algorithm (ACOG) dispatched on the image [1, 2].Second one is face detection based on usage of GA for advance classification of cases and objects of the input image.This work is based on preliminary segmentation of images into regions that contain non face objects and face objects.This idea may greatly accelerate the efficiency of face detection.