Ihor Paliy

Approach to Face Recognition Using Neural Networks

The paper describes the approach to automatic face recognition for access control application area using wavelet transform and neural networks ensemble. Wavelet transform implements the compression of the face images and thus accelerates the classifiers work, while neural networks ensemble with proposed decision rule provides low recognition error. Proposed ensembles decision rule carries out a high level of unknown peoples access rejection, which is the most significant requirement for the access control systems, and gives a good balance between known and unknown peoples recognition errors.

Homography-based face orientation determination from a fixed monocular camera

This paper presents a framework for determining the orientation of human faces with a fixed monocular camera which can be used for the purposes of the gaze tracking afterwards. We use homography relation between two views/frames to handle with the lack of depth information. In order to compensate for the lack of depth information in the relationships between the 2D images in the image plane and the 3D Euclidean space, we present a complete vision-based approach to pose estimation. The homography relates corresponding points captured at two different locations of the face and determines the relationships between the two locations using pixel information and intrinsic parameters of the camera. In order to determine the mapping between the two images, it is assumed that in each frame in the video sequence, we are able to locate, extract and labeled four feature points of the face located at virtual plane attached to the face. Face detection and facial feature extraction are executed with Viola-Jones method. The verification stage for face detection use combined cascade of neural network classifiers uses the convolutional neural network.

Recognition of facial images with subspace projection and dissimilarity representation

In this work, we present a framework for face recognition, combining face detection algorithm, dimensionality reduction method and a dissimilarity-based classifier. The face detection algorithm is intended to detect and extract faces in complex scenes, prior to face recognition. The Spectral Regression method, in sparse setting, is used for dimensionality reduction. The classification problem is solved by the Proximity Index ”Shape Coefficient” with SVM decision rules and Prototype Selection based classification. The results with real world experiments encourage us to propose this framework as good alternative to other face recognition methods.

Micro nucleus detection in human lymphocytes using convolutional neural network

The application of the convolution neural network for detection of the micro nucleuses in the human lymphocyte images acquired by the image flow cytometer is considered in this paper. The existing method of detection, called IMAQ Match Pattern, is described and its limitations concerning zoom factors are analyzed. The training algorithm of the convolution neural network and the detection procedure were described. The performance of both detection methods, convolution neural network and IMAQ Match Pattern, were researched. Our results show that the convolution neural network overcomes the IMAQ Match Pattern in terms of improvement of detection rate and decreasing the numbers of false alarms.

Fast and robust face detection and tracking framework

Face detection and tracking framework is described in the paper. Face detection is based on combined cascade of neural network-classifiers. Tracking is performed using Kalman filter. The framework was experimentally researched on a test video sequence and adjusted to obtain high processing speed.

Structural statistic method identifying facial images by contour characteristic points

This study discusses structural static method of human face identification for computer recognition systems based on building an identification vector as a set of characteristic points representing significant facial features with reduced data. It allowed providing an accurate classification invariant to the scale, rotation and shear at low computing complexity.

Human age-group classification of facial images with subspace projection and support vector machines

In this paper a system is presented for classification of facial images based on facial age estimation. The presented age-group classification system can be used for multimedia forensics where the investigation can be facilitated by automated analysis of large image datasets. Hence, two main features of such systems are usually desired: increased speed of computation and reliability to real-life images. In an attempt to meet these criteria we propose a combination of subspace projection algorihtm and a classifier for age group classification. Additionally a face detection algorithm is integrated to ensure detection of faces in complex scenes. The proposed system is implemented as software application and was tested on a large image dataset with real-life capturing conditions.

Detection of micro nucleus in human lymphocytes altered by Gaussian noise using convolution neural network

The application of convolution neural network for the detection of Micro Nucleuses (MNs) in human lymphocyte images acquired by an image flow cytometer is considered in this paper. The existing method of detection, IMAQ Match Pattern, is described. The training algorithm of the convolution neural network (CNN) and the detection procedure are presented. The performance of both detection methods are explored on the set of human lymphocyte images at the different intensities of Gaussian noise alteration. Our results show that the IMAQ Match Pattern method provides low detection rates of the MNs at the presence even of the small intensity of Gaussian noise alteration. Instead the CNN provides much higher detection rates at the different intensities of Gaussian noise alteration.

Improved Method of Face Detection Using Color Images

The paper describes the improved method of human face detection on the static color images using skin color segmentation and artificial neural networks ensemble. The method can be later implemented as the first part of human identification system.