Extraction of human face features from color images

IEEE, Geometric Modeling and Imaging (GMAI'06), 2006

Human face detection is concerned with finding the location and size of every human face in a given image. Face detection plays a very important role in human computer interaction field. It represents the first step in a fully automatic face re cognition, facial features detection, and expression recognition. There are many techniques used in face detection, each one has its advantages and disadvantages. The face detection system presented in this paper is a hybrid of known algorithms. First stag e of the proposed method is applying skin detection algorithm to specify all skin locations in the image. Second, extract face features like eyes, mouth and nose. At the last, a verification step is applied to ensure that the extracted features are facial features. In experiments on images having upright frontal faces with any background our system has achieved high detection rates and low false positives.

We present in this thesis paper an approach to automatic detection of human faces in color images. Human face detection has received substantial attention from researchers in biometrics, computer vision, pattern recognition, and cognitive psychology communities because of the increased attention being devoted to security, man-machine communication, content-based image retrieval, and image/video coding. Face detection is a challenging computer vision problem because of lighting conditions, a high degree of variability in size, shape, background, color, etc. To build fully automated systems, Gaussian model and efficient face detection algorithms are required. In our thesis, we propose a feature based algorithm for detecting human faces color images with multiple faces that is sufficiently generic and is also easily extensible to cope with more demanding variations of the imaging conditions. Our proposed algorithm first corrects the color bias by a lighting compensation technique. We overcame the difficulty of detecting the low-luma and high-luma skin tones by applying a nonlinear transformation to the YCbCr color space with lighting compensation technique. We first detect the skin region from color images using Gaussian Model and then finding face candidates from grouping skin region using filtering. We constructed eye, mouth boundary maps to verify each face candidate. To detect eye we used Daugman's method & detect mouth using HSB color transformation. After that we construct eye and mouth maps in order to work on the triangle relationship between them. Our face detector has been applied to several test images, and satisfactory results have been obtained.

— Face detection and tracking have wide applications, for example, law enforcement, gaming, image search, marketing, etc. The detection and tracking tasks are quite challenging. Tracking the face in an image includes gender identification and segmenting the face using the skin color. In the past, many methods have been proposed to identify a face in videos, such as feature-based detection, skin color segmentation, appearance, and eigen-based identification. These techniques use image classifiers (both weak and strong) but are not very accurate. This paper introduces a hybrid method that utilizes skin color for increasing accuracy of detection and tracking. In our method, an image is rescaled and divided based on skin color, using RGB (red-green-blue) color combinations. The divided skin tone image is combined with the edges of images before applying the morphological operations. As a final step, a bounding box appears on the detected face.

International journal of engineering research and technology, 2018

Human face in addition to revealing the identity of an individual also provides information regarding his/her emotional and psychological state. In addition to this various facial features are utilized in applications like biometrics (face, iris and lips), automatic lip-reading, speech recognition etc. For accomplishing these tasks reliable face and facial feature detection methods are required. In this work a hybrid facial feature extraction approach based on wavelets and information from HSV and YCbCr color spaces has been presented. The main emphasis of this paper is to extract lip-region for further applications. The algorithm has been tested on images captured by different resolution cameras under varying illumination condition.

2014

This paper presents a new face parts information analyzer, as a promising model for detecting faces and locating the facial features in images. The main objective is to build fully automated human facial measurements systems from images with complex backgrounds. Detection of facial features such as eye, nose, and mouth is an important step for many subsequent facial image analysis tasks. The study covers the tasks detection, landmark localization and measurement facial part that have traditionally been approached as separate problems with different techniques. Different set of techniques have been introduced recently, for example; principal component analysis, geometric modeling, auto-correlation, deformable template, neural networks, color analysis, window classifiers, view-based Eigen space methods, and elastic graph models. The study present a novel and simple model approach based on a mixture of techniques and algorithms in a shared pool based on Viola–Jones object detection fra...

International Journal of Emerging Technologies in Computational and Applied Sciences (IJETCAS)International Association of Scientific Innovation and Research (IASIR), 2013

Human face detection plays important roles in many applications such as video surveillance and face image database management. In our Paper, we have studied worked on face detection By YCbCrHs Techniques and developed algorithms for them. In face detection, we have developed an algorithm that can detect human faces from an image. We have taken skin color as a tool for detection. This technique works well for Indian faces which have a specific complexion varying under certain range. We have taken real life examples and simulated the algorithms in MATLAB successfully. Our aim, which we believe we have reached, was to develop a method of face recognition that is fast, robust, reasonably simple and accurate with a relatively simple and easy to understand algorithms and. techniques. The examples provided in this Paper are real-time and taken from our own surroundings. While the RGB, HSV and YUV (YCbCr) are standard models used in various colour imaging applications, not all of their information are necessary to classify skin colour. In This Paper presents a novel skin colour model, RGB-YCbCrHs for the detection of human faces.

International Journal of Image, Graphics and Signal Processing, 2014

Real time faces detection and face tracking is one of the challenging problems in application like computer human interaction, video surveillance, biometrics etc. In this paper we are presenting an algorithm for real time face detection and tracking using skin color segmentation and region properties. First segmentation of skin regions from an image is done by using different color models. Skin regions are separated from the image by using thresholding. Then to decide whether these regions contain human face or not we used face features. Our procedure is based on skin color segmentation and human face features (knowledgebased approach). We have used RGB, YCbCr, and HSV color models for skin color segmentation. These color models with thresholds, help to remove non skin like pixel from an image. Each segmented skin regions are tested to know whether region is human face or not, by using human face features based on knowledge of geometrical properties of human face.

Trying to detecting a face from any photo is big problem and got these days a focusing because of it importance, in face recognition system,face detection in one of the basic components. A lot of troubles are there to be solved in order to create a successful face detection algorithm. The skin of face has its properties in color domain and also a texture which may help in the algorithm for detecting faces because of its ability to find skins from photo. Here we are going to create a new algorithm for human face detection depending on skin color tone specially YCbCr color tone as an approach to slice the photo into parts. In addition, Gray level has been used to detect the area which contains a skin, after that anotherlevel used to erase the area that does not contain skin. The system which proposed applied on many photos and passed with great accuracy of detecting faces and it has a good efficient especially to separate the area that does not contain skin or face from the area which contain face and skin. It has been agreed and approved that the accuracy of the proposed system is 98% in human face detection.

Facial feature extraction is an important preprocessing step in facial expression recognition. The performance of facial feature extraction can greatly influence the accuracy of the recognition system. In this paper, an automated facial feature extraction method that is tolerant to cluttered scene, illumination change and depth is proposed. The method first utilizes skin color to swiftly detect the face in the image and then proposes a hierarchical geometrical framework to estimate the range of facial features. An adaptive gradient mask which can resist to illumination change and adapt to deformation of facial features is devised. Experimental results show that the method can extract exact locations of deformed facial features in five kinds of facial expressions.

Journal of Emerging …, 2010

Biometric technology has been frequently utilized in identifying and recognizing human components. This technology identifies human's unique and static body parts, such fingerprints, eyes, and face. One of the most biometric technologies which are widely used is facial recognition. The identification and recognition of a human face utilize the face components' processing and analysis. This technique consists of determining face components' region and their characteristics, which establishes the role of individual component in face recognition. This research develops a system that defines face components by determining the distance of face components (i.e.: the eyes, nose, mouth) and other facial components. This process conducted on a frontal single still image to acquire the components. Distances between components are determined by detecting the based skin color, cropping to normalize face region, and extracting eyes, nose, and mouth components. This research utilizes 150 Indonesian face samples and has successfully determined the face components'. From the experiment we conclude that the determination of face components and face components' distances can be used to identify a face as a subsystem of a face recognition system. Test of uniqueness to 150 samples has succeeded. The result indicated that eight face component distances give better result than the previous one, which only applied three components distance. The test of uniqueness with eigenspace showed the existence of different characteristic for every face image.