In this paper, we present a new approach for facial face recognition. The method is based on the Fourier transform of Gabor filters and the method of regularized linear discriminate analysis applied to facial features previously localized. The process of facial face recognition is based on two phases: location and recognition. The first phase determines the characteristic using the local properties of the face by the variation of gray level along the axis of the characteristic and the geometric model, and the second phase generates the feature vector by the convolution of the Fourier transform of 40 Gabor filters and face, followed by application of the method of regularized linear discriminate analysis on the vectors of characteristics. Experimental results obtained on sample of images from the XM2VTSDB database [1] have shown that the proposed algorithm gives satisfactory results in a precise manner.

In this paper, we present image processing and pattern recognition techniques to extract human faces and facial features from color images. First, we segment a color image into skin and non-skin regions by a Gaussian skin-color model. Then, we apply mathematical morphology and region filling techniques for noise removal and hole filling. We determine whether a skin region is a face candidate by its size and shape. Principle component analysis (PCA) is used to verify face candidates. We create an ellipse model to locate eyes and mouths areas roughly, and apply the support vector machine (SVM) to classify them. Finally, we develop knowledge rules to verify eyes. Experimental results show that our algorithm achieves the accuracy rate of 96.7% in face detection and 90.0% in facial feature extraction.

Facial expression provides an important behavioral measure for studies of emotion, cognitive processes, and social interaction. Facial expression recognition has recently become a promising research area. Its applications include human-computer interfaces, human emotion analysis, and medical care and cure. In this paper, we investigate various feature representation and expression classification schemes to recognize seven different facial expressions, such as happy, neutral, angry, disgust, sad, fear and surprise, in the JAFFE database. Experimental results show that the method of combining 2D-LDA (Linear Discriminant Analysis) and SVM (Support Vector Machine) outperforms others. The recognition rate of this method is 95.71% by using leave-one-out strategy and 94.13% by using cross-validation strategy. It takes only 0.0357 second to process one image of size 256 × 256.

We propose that a machine assessment system of neonatal expressions of pain be developed to assist clinicians in diagnosing pain. The facial expressions of 26 neonates (age 18–72 h) were photographed experiencing the acute pain of a heel lance and three nonpain stressors. Four algorithms were evaluated on out-of-sample observations: PCA, LDA, SVMs and NNSOA. NNSOA provided the best classification rate of pain versus nonpain (90.20%), followed by SVM with linear kernel (82.35%). We believe these results indicate a high potential for developing a decision support system for diagnosing neonatal pain from images of neonatal facial displays.

In this paper, an intelligent sensor network is developed for object detection, classification and recognition. We utilize wireless sensors as the first layer to detect coordinates of moving objects in a secured area. Cameras are activated to capture image features for object classification and recognition. In order to reduce processing time, a hierarchical image extraction approach is developed. Global object features such as size and motion are acquired for classifying objects into a number of classes. If the moving object is considered suspicious, the cameras will be requested to capture detailed images for object recognition. Experimental results show that our system can achieve a high face recognition rate of 95.4% for the testing images captured by the surveillance system.

This paper reports experiments that explore performance differences in two previous studies that investigated SVM classification of neonatal pain expressions using the Infant COPE database. This database contains 204 photographs of 26 neonates (age 18-36 hours) experiencing the pain of heel lancing and three nonpain stressors. In our first study, we reported experiments where representative expressions of all subjects were included in the training and testing sets, an experimental protocol suitable for intensive care situations. A second study used an experimental protocol more suitable for short-term stays: the SVMs were trained on one sample and then evaluated on an unknown sample. Whereas SVM with polynomial kernel of degree 3 obtained the best classification score (88.00%) using the first evaluation protocol, SVM with a linear kernel obtained the best classification score (82.35%) using the second protocol. However, experiments reported here indicate no significant difference in performance between linear and nonlinear kernels.

Facial expression provides a crucial behavioral measure for studies of human emotion, cognitive processes, and social interaction. In this paper, we focus on recognizing facial action units (AUs), which represent the subtle change of facial expressions. We adopt ICA (independent component analysis) as the feature extraction and representation method and SVM (support vector machine) as the pattern classifier. By comparing with three existing systems, such as Tian, Donato, and Bazzo, our proposed system can achieve the highest recognition rates. Furthermore, the proposed system is fast since it takes only 1.8 ms for classifying a test image.

Facial expression provides an important behavioral measure for studies of emotion, cognitive processes, and social interaction. Facial expression representation and recognition have become a promising research area during recent years. Its applications include human-computer interfaces, human emotion analysis, and medical care and cure.In this dissertation, the fundamental techniques will be first reviewed, and the developments of the novel algorithms and theorems will be presented later. The objective of the proposed algorithm is to provide a reliable, fast, and integrated procedure to recognize either seven prototypical, emotion-specified expressions (e.g., happy, neutral, angry, disgust, fear, sad, and surprise in JAFFE database) or the action units in Cohn-Kanade AU-coded facial expression image database.A new application area developed by the Infant COPE project is the recognition of neonatal facial expressions of pain (e.g., air puff, cry, friction, pain, and rest in Infant COPE database). It has been reported in medical literature that health care professionals have difficulty in distinguishing newborn's facial expressions of pain from facial reactions of other stimuli. Since pain is a major indicator of medical problems and the quality of patient care depends on the quality of pain management, it is vital that the methods to be developed should accurately distinguish an infant's signal of pain from a host of minor distress signal. The evaluation protocol used in the Infant COPE project considers two conditions: person-dependent and person-independent. The person-dependent means that some data of a subject are used for training and other data of the subject for testing. The person-independent means that the data of all subjects except one are used for training and this left-out one subject is used for testing. In this dissertation, both evaluation protocols are experimented. The Infant COPE research of neonatal pain classification is a first attempt at applying the state-of-the-art face recognition technologies to actual medical problems. The objective of Infant COPE project is to bypass these observational problems by developing a machine classification system to diagnose neonatal facial expressions of pain. Since assessment of pain by machine is based on pixel states, a machine classification system of pain will remain objective and will exploit the full spectrum of information available in a neonate's facial expressions. Furthermore, it will be capable of monitoring neonate's facial expressions when he/she is left unattended. Experimental results using the Infant COPE database and evaluation protocols indicate that the application of face classification techniques in pain assessment and management is a promising area of investigation.One of the challenging problems for building an automatic facial expression recognition system is how to automatically locate the principal facial parts since most existing algorithms capture the necessary face parts by cropping images manually. In this dissertation, two systems are developed to detect facial features, especially for eyes. The purpose is to develop a fast and reliable system to detect facial features automatically and correctly. By combining the proposed facial feature detection, the facial expression and neonatal pain recognition systems can be robust and efficient.

Objective: It has been reported in medical literature that health care professionals have difficulty distinguishing a newborn's facial expressions of pain from facial reactions to other stimuli. Although a number of pain instruments have been developed to assist health professionals, studies demonstrate that health professionals are not entirely impartial in their assessment of pain and fail to capitalize on all the information exhibited in a newborn's facial displays. This study tackles these problems by applying three different state-of-the-art face classification techniques to the task of distinguishing a newborn's facial expressions of pain. Methods: The facial expressions of 26 neonates between the ages of 18h and 3 days old were photographed experiencing the pain of a heel lance and a variety of stressors, including transport from one crib to another (a disturbance that can provoke crying that is not in response to pain), an air stimulus on the nose, and friction on the external lateral surface of the heel. Three face classification techniques, principal component analysis (PCA), linear discriminant analysis (LDA), and support vector machine (SVM), were used to classify the faces. Results: In our experiments, the best recognition rates of pain versus nonpain (88.00%), pain versus rest (94.62%), pain versus cry (80.00%), pain versus air puff (83.33%), and pain versus friction (93.00%) were obtained from an SVM with a polynomial kernel of degree 3. The SVM outperformed two commonly used methods in face classification: PCA and LDA, each using the L1 distance metric. Conclusion: The results of this study indicate that the application of face classification techniques in pain assessment and management is a promising area of investigation. on.