Face recognition based on genetic algorithm

This paper pertains to the recognition of human face within the color images. The proposed scheme used the clue of skin color to detect faces within the color images, and Mallat wavelet transform was applied to get the LL2 frequency subband image of the original faces, then, features were extracted from the raw data by using principal component analysis (PCA). These discriminative and low dimensional features achieved were fed to a support vector machines (SVMs) with polynomial kernel for the training and fuzzy theory was applied on multi-class classification to deal with the unclassifiable cases. Experiments were conducted using the indoors photographs and the results demonstrated that the proposed methodology was efficient for the frontal faces detection and recognition.

In the era of smart digital transformation, automatic face recognition is the way of identification and verification of a person in many applications of security and authentication. If the person's 2D still image or 3D video frame is taken under controlled lighting and frontal face poses, then today automatic face recognition is a solved problem with more than 98 % accuracy. Under ideal condition of images, automatic face recognition outperforms manual face recognition rate. But the problem of uncontrolled illumination, occlusion, tilting of faces, different expressions of faces, use of accessories and hair color, growth of facial hair, aging effect of a person and low-resolution images makes automatic face recognition underperform; it gets defeated by human. Unless this problem is solved on real time, automatic face recognition system cannot be trusted for crucial security applications like e-passport, fraud detection, counter terrorism and mug-shot verification etc. Extensive research is underway to improve this technology. Traditionally Harr cascade classifier methods, histogram of oriented gradients (HoG), principal component analysis (PCA), Eigen-faces, and support vector machine (SVM) classifier are used in face recognition. Now modern deep learning innovation has superseded those traditional machine learning techniques with more computing power in GPUs and TPUs to tackle all variations in input image quality and ever-growing face database. The goal of this paper is to report the journey of face recognition system from traditional to modern techniques to increase precision and recall of the system.

Face recognition of persons has received so much attention in the recent years due to its many applications in different fields such as security applications, video surveillance, biometric systems, identifying the criminals etc. This paper develops a system that can recognize the human face in the input image after it has been detected. The system is trained with set of faces and non faces, and when the input picture is given, the face is detected using Viola Jones Algorithm. In face recognition, the features are extracted from the training dataset using Principal Component Analysis (PCA) and then the system is trained to recognize the face using Support Vector Machine (SVM) classification. When the input image is given for face recognition, features are extracted from the input picture using PCA and multiclass classification is done by SVM.

The paper deals with evaluation of automatic training samples selection method based on self-organizing map (SOM) in face recognition systems. In earlier paper [1] we presented an approach for automatic training samples selection using various clustering algorithms with good results on the CMU PIE face database. We showed that with the use of SOM we can achieve a good training samples selection. In this paper we further evaluate this approach with the use of face recognition systems based on principal component analysis (PCA) and support vector machines (SVM). We compare the results with random (uncontrolled and controlled) training samples selection and we evaluate the recognition accuracy of each method.

This paper presents the modeling of face recognition using feature extraction based on Principal Component Analysis (PCA) and Support Vector Machine (SVM) as a classifier. Three PCA techniques were compared, they are 1DPCA, 2DPCA and Bi-2DPCA. Meanwhile, three type of SVM kernel functions-linear, polynomial, and radial basis function (RBF) were used. The experiment used the ORL Face Database AT&T Laboratory, which contain 400 images with 10 images per each person. The leave one out method is used for validating each pair of extraction and classifier method. The highest accuracy is obtained by a combination of linear kernel and Bi-2DPCA85%, with 94.25%, and also the fastest computation time, is 15.34 seconds. Index Terms” Face Recognition, Principle Component Analysis, Kernel, Support Vector Machine, Leave-one Out Cross Validation

Face recognition is a significant biometric credential in the field of security authentication. It additionally assumes a noteworthy job in image processing and it is applicable in various systems like verifying the identity of the person and in security purpose. Recognizing the face with varying background, poses and illumination are the complexity involved in this face recognition. Many algorithms exist for face recognition, of which, Discrete Wavelet Transform (DWT) with Principal Component Analysis (PCA) works better for recognition of faces. An algorithm using 3 Level-DWT and modified PCA is proposed for feature extraction. The PCA and reconstruction of images using Inverse PCA, help not only for dimensionality reduction, but also to find the least principal components (PC) of an image from which the significant features of a face image can be extracted. The significant features thus extracted are used for classifying genetic and non-genetic faces. Using extracted features from 3 level DWT and PCA, Support vector machine (SVM) is utilized to classify the faces genetically. The proposed extracted features does not intend to certain features like ears, nose and eyes of the face, but corresponds to identify the faces which are genetically similar. Based on the statistical measure analysis, the proposed algorithm 3 Level dwt with modified PCA works well in extracting the features for identifying the faces which are genetically closer. This face recognition application system can be effectively used to treat a patient in other location with complete security. There is no chance for data stealing, since the concerned doctors and patient only will take part in the system. The identification of genetic faces will turn out to be an achievement in the field of health care monitoring systems.

Machine learning and pattern recognition recently become a hot topic in computing world. This is due to the fast-growing of resources as well as techniques that make it easier to solve machine learning and pattern recognition problems. Problems that require machine learning solutions may be very simple for humans but actually can be very complex for machines to solve them. Face recognition is amongst those problems. Almost all human can easily recognize others without require specific knowledge to do it, different from machines which require its. This paper discussed face recognition task using machine learning strategies which involved Kernel Principal Component Analysis (KPCA) and Support Vector Machine (SVM) to identify person. KPCA extracted features from 2D image input and produced the important features of an image input. The extracted face features are recognized by SVM by classifying human face according to their stored identity in a database. SVM, which was basically a binary classifier worked by using one-against-one strategy to compare the face feature vector of a single test image to the stored face image in a face image database. Experiment results on grayscale images with size 92x112 pixels gave 96.25% of accuracy rate. Hence, KPCA and SVM for face recognition is a robust machine learning method.

Aiming at Shenzhen component index in the stock market change trend prediction problem, a new method based on Principal component analysis and Support vector machine was proposed. Principal component analysis was used to calculate the influence factors of the share price of Shenzhen component index and the closing price of Shenzhen component index, and then the influence factors were reduced from high dimension to low dimension. From 1991 to 2018, More than 6,700 groups of data related to the Shenzhen component index were selected to test the method. The average relative error of the Shenzhen component index closing price calculated is 7.35%, the experimental results showed that. The convergence speed and accuracy of the experiment are improved by the principal component analysis method. As a result, based on principal component analysis and support vector machine, it can be able to make use of the data of known Shenzhen component index to speculate on the trend of closing prices. And it has a certain application prospect in the field of stock forecasting research.

Breast cancer risk assessment and early diagnosis using Principal Component Analysis and support vector machine techniques

Predicting profitability of listed construction companies based on principal component analysis and support vector machine—Evidence from China

The Sixth International Symposium on Neural Networks (ISNN 2009)

Nowadays, face detection and recognition have gained importance in security and information access. In this paper, an efficient method of face detection based on principal components analysis (PCA) and support vector machine (SVM) is proposed. It firsly filter the face potential area using statistical feature which is generated by analyzing local histogram distribution. And then, SVM classifier is used to detect face feature in the test image, SVM has great performance in classification task. PCA is used to reduce dimension of sample data. After PCA transform, the feature vectors, which are used for training SVM classifier, are generated. Our tests in this paper are based on CMU face database. The experimental results demonstrate that the proposed method is encouraging with a successful detection rate.

In this paper we propose an efficient algorithm based on principal component analysis (PCA) and fuzzy support vector machine (SVM) for the diagnosis of breast cancer tumor. First, PCA algorithm is implemented to project high-dimensional breast tumor data into much lower dimensional space, then the processed data are classified by a fuzzy SVM classifier. Experimental and analytical results show that in the diagnosis of breast cancer tumor the proposed method can greatly speed up the training and testing of the classifier, get high testing correct rate and pick out untypical cases to be reexamined by experienced doctors, superior to the traditional rigid margin SVM classifier.

Gaining knowledge from health data may contribute to novel approaches or improved patient care plans. To intelligently acquire knowledge from the stroke data, machine learning (ML) technique is utilised and healthcare systems are used. Globally, security has become a major concern, and biometric technologies such as face detection and recognition technologies have been developed to solve security issues and security threats in the world. It has become possible because of developments in technology in the fields of automatic face analysis, machine learning, and pattern recognition. This chapter proposes a prototype to classify brain stroke and face recognition using ML techniques in the medical image. Face recognition and classification of stroke through ML technique like Decision Tree (DT), Artificial Neural Network (ANN), Support Vector Machine (SVM), and Deep Learning with Convolution Neural Network (CNN) based on texture features and statistical features. A CNN model is experimentally used along with a collection of hyper-parameters for deep face recognition and stroke classification. Promising experimental results, with a total accuracy of 98.86% and 98.5%, are obtained, which demonstrates the effectiveness of deep face identification and stroke classification, respectively, in biometric systems.

The purpose of triage is to prevent the delay of treatment for patients in real emergencies due to excessive numbers of patients in the hospital. This study uses the data of patients of consistent triage to develop the triage prediction model. By integrating Principal Component Analysis (PCA) and Support Vector Machine (SVM), the anomaly detection (overestimate and underestimate) prediction accuracy rate can be 100%, which is better than the accuracy rate of SVM (about 89.2%) or Back- propagation Neural Networks (BPNN) (96.71%); afterwards, this study uses Support Vector Regression (SVR) to adopt Genetic Algorithm (GA) to determine three SVR parameters to predict triage. After using the scroll data predictive values, we calculate the Absolute Percentage Error (APE) of each scroll data. The resulting SVR's Mean Absolute Percentage Error (MAPE) is 3.78%, and BPNN's MAPE is 5.99%; therefore, the proposed triage prediction model of this study can effectively predict anomaly detection and triage.