Janusz Bobulski

Two-Dimensional Hidden Markov Models for Pattern Recognition

Hidden Markov models are well-known methods for image processing. They are used in many areas where 1D data are processed. In the case of 2D data, there appear some problems with application HMM. There are some solutions, but they convert input observation from 2D to 1D, or create parallel pseudo 2D HMM, which is set of 1D HMMs in fact. This paper describes authentic 2D HMM with two-dimensional input data, and its application for pattern recognition in image processing.

2DHMM-Based Face Recognition Method

So far many methods of recognizing the face arose, each has the merits and demerits. Among these methods are methods based on Hidden Markov models, and their advantage is the high efficiency. However, the traditional HMM uses one-dimensional data, which is not a good solution for image processing, because the images are two-dimensional. Transforming the image in a one-dimensional feature vector, we remove some of the information that can be used for identification. The article presents the full ergodic 2D-HMM and applied for face identification.

Hidden Markov Models For Two-Dimensional Data

Hidden Markov models are well-known methods for image processing. They are used in many areas where 1D data are processed. In the case of 2D data, there appear some problems with application HMM. There are some solutions, but they convert input observation from 2D to 1D, or create parallel pseudo 2D HMM, which is set of 1D HMMs in fact. This paper describes authentic 2D HMM with two-dimensional input data, and its application for pattern recognition in image processing.

Pseudo-entropy Similarity for Human Biometrics

With complex multimedia data, we see the emergence of biometric identification/authentication systems in which the fundamental operation is the similarity assessment of natural information-carrying objects. We have developed a similarity measure JeK, based on new notion of pseudo-entropy. The measure exhibits several features that match experimental findings in multimedia perception. We show how the measure can be used for identification/authentication of complex biometric objects, such as faces and its emotions, voices, and so on. We address the use of the pseudo-entropy measure JeK to deal with relations among the varied properties of 1D, 2D and 3D biometric objects.

Comparison of the Effectiveness of 1D and 2D Hmm in the Pattern Recognition

Abstract Hidden Markov Model (HMM) is a well established technique for image recognition and has also been successfully applied in other domains such as speech recognition, signature verification and gesture recognition. HMM is widely used mechanism for pattern recognition based on 1D data. For images one dimension is not satisfactory, because the conversion of one-dimensional data into a twodimensional lose some information. This paper presents a solution to the problem of 2D data by developing the 2D HMM structure and the necessary algorithms.

Face Recognition with 3D Face Asymmetry

Using of 3D images for the identification was in a field of the interest of many researchers which developed a few methods offering good results. However, there are few techniques exploiting the 3D asymmetry amongst these methods. We propose fast algorithm for rough extraction face asymmetry that is used to 3D face recognition with hidden Markov models. This paper presents conception of fast method for determine 3D face asymmetry. The research results indicate that face recognition with 3D face asymmetry may be used in biometrics systems.

Face Recognition Method with Two-Dimensional HMM

This paper presents an automatic face recognition system, which bases on two-dimensional hidden Markov models. The traditional HMM uses one-dimensional data vectors, which is a drawback in the case of 2D image processing, as part of the information is lost during conversion. The paper presents the full ergodic 2D-HMM and uses it to identify faces. The experimental results demonstrate that the system, basing on two-dimensional hidden Markov models, is able to achieving an average recognition rate of 94 %.

Two-Dimensional Hidden Markov Models in Road Signs Recognition

This paper presents an automatic road sign recognition system. The system bases on twodimensional hidden Markov models. First, a study of the existing road sign recognition research is presented. In this study, the issues associated with automatic road sign recognition are described, the existing methods developed to tackle the road sign recognition problem are reviewed, and a comparison of the features of these methods is given. Second, the created road sign recognition system is described. The system is able to recognize the road signs, which was detected earlier. The system makes use of two dimensional discrete wavelet transform for features extraction of road signs. In recognition process system bases on two dimensional hidden Markov models. The experimental results demonstrate that the system is able to achieving an average recognition rate of 83% using the two-dimensional hidden Markov models and the wavelet transform.

Single camera based location estimation with dissimilarity measurement

In this paper we present a method that uses a single camera as a passive sensor for location estimation. The implementation is focused on the image processing techniques and dissimilarity measurement. Experimental results have indicated that good state estimates could be obtained, considering very little use of visual information. Consequently, popular cameras can be perceived as a simple and interesting solution for visual odometry and present very promising possibility in autonomous navigation.

PET Waste Classification Method and Plastic Waste DataBase - WaDaBa.

The main purpose of this work was creation of a plastic waste database of images of objects constituting the typical contents of municipal waste. This group of waste, by using methods of Computer Vision can be automatically selected on the sorting lines businesses for waste disposal. Digital images of items that will be received for processing should reflect the specific conditions of places where real objects have to be found. Thus, each thing is placed in this database should be presented in the course of several collections of images, taking into account different lighting conditions and different arrangement relative to the image recorder, and the different degree of deformation of these objects as a result of previous processes. Images created in the collection will be divided into groups based on the type of material from which individual objects were made. An second main aim of the article is to present the method of plastic waste selection based on histogram analysis. The method has to be fast so that it can be used in a waste sorting plant.