An overview of the self-organizing map algorithm, on which the papers in this issue are based, is presented in this article.

The Self-Organizing Map

Emotion recognition in human-computer interaction

This book provides an overview of the theory and practice of continuous and discrete wavelet transforms. Divided into seven chapters, the first three chapters of the book are introductory, describing the various forms of the wavelet transform and their computation, while the remaining chapters are devoted to applications in fluids, engineering, medicine and miscellaneous areas. Each chapter is well introduced, with suitable examples to demonstrate key concepts. Illustrations are included where appropriate, thus adding a visual dimension to the text. A noteworthy feature is the inclusion, at the end of each chapter, of a list of further resources from the academic literature which the interested reader can consult. The first chapter is purely an introduction to the text. The treatment of wavelet transforms begins in the second chapter, with the definition of what a wavelet is. The chapter continues by defining the continuous wavelet transform and its inverse and a description of how it may be used to interrogate signals. The continuous wavelet transform is then compared to the short-time Fourier transform. Energy and power spectra with respect to scale are also discussed and linked to their frequency counterparts. Towards the end of the chapter, the two-dimensional continuous wavelet transform is introduced. Examples of how the continuous wavelet transform is computed using the Mexican hat and Morlet wavelets are provided throughout. The third chapter introduces the discrete wavelet transform, with its distinction from the discretized continuous wavelet transform having been made clear at the end of the second chapter. In the first half of the chapter, the logarithmic discretization of the wavelet function is described, leading to a discussion of dyadic grid scaling, frames, orthogonal and orthonormal bases, scaling functions and multiresolution representation. The fast wavelet transform is introduced and its computation is illustrated with an example using the Haar wavelet. The second half of the chapter groups together miscellaneous points about the discrete wavelet transform, including coefficient manipulation for signal denoising and smoothing, a description of Daubechies' wavelets, the properties of translation invariance and biorthogonality, the two-dimensional discrete wavelet transforms and wavelet packets. The fourth chapter is dedicated to wavelet transform methods in the author's own specialty, fluid mechanics. Beginning with a definition of wavelet-based statistical measures for turbulence, the text proceeds to describe wavelet thresholding in the analysis of fluid flows. The remainder of the chapter describes wavelet analysis of engineering flows, in particular jets, wakes, turbulence and coherent structures, and geophysical flows, including atmospheric and oceanic processes. The fifth chapter describes the application of wavelet methods in various branches of engineering, including machining, materials, dynamics and information engineering. Unlike previous chapters, this (and subsequent) chapters are styled more as literature reviews that describe the findings of other authors. The areas addressed in this chapter include: the monitoring of machining processes, the monitoring of rotating machinery, dynamical systems, chaotic systems, non-destructive testing, surface characterization and data compression. The sixth chapter continues in this vein with the attention now turned to wavelets in the analysis of medical signals. Most of the chapter is devoted to the analysis of one-dimensional signals (electrocardiogram, neural waveforms, acoustic signals etc.), although there is a small section on the analysis of two-dimensional medical images. The seventh and final chapter of the book focuses on the application of wavelets in three seemingly unrelated application areas: fractals, finance and geophysics. The treatment on wavelet methods in fractals focuses on stochastic fractals with a short section on multifractals. The treatment on finance touches on the use of wavelets by other authors in studying stock prices, commodity behaviour, market dynamics and foreign exchange rates. The treatment on geophysics covers what was omitted from the fourth chapter, namely, seismology, well logging, topographic feature analysis and the analysis of climatic data. The text concludes with an assortment of other application areas which could only be mentioned in passing. Unlike most other publications in the subject, this book does not treat wavelet transforms in a mathematically rigorous manner but rather aims to explain the mechanics of the wavelet transform in a way that is easy to understand. Consequently, it serves as an excellent overview of the subject rather than as a reference text. Keeping the mathematics to a minimum and omitting cumbersome and detailed proofs from the text, the book is best-suited to those who are new to wavelets or who want an intuitive understanding of the subject. Such an audience may include graduate students in engineering and professionals and researchers in engineering and the applied sciences.

The Illustrated Wavelet Transform Handbook: Introductory Theory and Applications in Science, Engineering, Medicine and Finance

advanced digital signal processing and noise reduction is available in our digital library an online access to it is set as public so you can download it instantly. Our books collection spans in multiple locations, allowing you to get the most less latency time to download any of our books like this one. Kindly say, the advanced digital signal processing and noise reduction is universally compatible with any devices to read.

Advanced Digital Signal Processing And Noise Reduction

Deep neural networks for automatic detection of osteoporotic vertebral fractures on CT scans

Mel Frequency Cepstral Coefficients (MFCCs) are the most widely used features in the majority of the speaker and speech recognition applications. Since 1980s, remarkable efforts have been undertaken for the development of these features. Issues such as use suitable spectral estimation methods, design of effective filter banks, and the number of chosen features all play an important role in the performance and robustness of the speech recognition systems. This paper provides an overview of MFCC's enhancement techniques that are applied in speech recognition systems. The details such as accuracy, types of environments, the nature of data, and the number of features are investigated and summarized in the table combined with the corresponding key references. Benefits and drawbacks of these MFCC's enhancement techniques have been discussed. This study will hopefully contribute to raising initiatives towards the enhancement of MFCC in terms of robustness features, high accuracy, and less complexity.

Mel frequency cepstral coefficients (Mfcc) feature extraction enhancement in the application of speech recognition: A comparison study

How to Construct Open Ended Questions

Emotions play an important role in human interactions. They can be integrated into the computer system to make human---computer interaction (HCI) more effective. Affective computing is an innovative computational modeling and detecting user's emotions to optimize system responses in HCI. However, there is a trade-off between recognition accuracy and real-time performance in some of the methods such as processing the facial expressions, human voice and body gestures. Other methods lack efficiency and usability in real-world applications such as natural language processing and electroencephalography signals. To accomplish a reliable, usable and high-performance system, this paper proposes an intelligent hybrid approach to recognize users' emotions by using easily accessible and low computational cost input devices including keyboard, mouse (touch pad: single touch) and touch screen display (single touch). Using the proposed approach, the system is developed and trained in a supervised mode by artificial neural network and support vector machine (SVM) techniques. The result shows an increase in accuracy of 6 % (93.20 %) by SVM in comparison with the currently existing methods. It is a significant contribution to show new directions of future research in emotion recognition, user modeling and emotional intelligence.

Hybrid affective computing--keyboard, mouse and touch screen: from review to experiment

This paper discusses a fuzzy model for multi-level human emotions recognition by computer systems through keyboard keystrokes, mouse and touchscreen interactions. This model can also be used to detect the other possible emotions at the time of recognition. Accuracy measurements of human emotions by the fuzzy model are discussed through two methods; the first is accuracy analysis and the second is false positive rate analysis. This fuzzy model detects more emotions, but on the other hand, for some of emotions, a lower accuracy was obtained with the comparison with the non-fuzzy human emotions detection methods. This system was trained and tested by Support Vector Machine (SVM) to recognize the users' emotions. Overall, this model represents a closer similarity between human brain detection of emotions and computer systems.

Fuzzy Model in Human Emotions Recognition.

To date, most of the human emotion recognition systems are intended to sense the emotions and their dominance individually. This paper discusses a fuzzy model for multilevel affective computing based on the dominance dimensional model of emotions. This model can detect any other possible emotions simultaneously at the time of recognition. One hundred and thirty volunteers from various countries with different cultural backgrounds were selected to record their emotional states. These volunteers have been selected from various races and different geographical locations. Twenty-seven different emotions with their strengths in a scale of 5 were questioned through a survey. Recorded emotions were analyzed with the other possible emotions and their levels of dominance to build the fuzzy model. Then this model was integrated into a fuzzy emotion recognition system using three input devices of mouse, keyboard and the touch screen display. Support vector machine classifier detected the other possible emotions of the users along with the directly sensed emotion. The binary system (non-fuzzy) sensed emotions with an incredible accuracy of 93 %. However, it only could sense limited emotions. By integrating this model, the system was able to detect more possible emotions at a time with slightly lower recognition accuracy of 86 %. The recorded false positive rates of this model for four emotions were measured at 16.7 %. The resulted accuracy and its false positive rate are among the top three accurate human emotion recognition (affective computing) systems.

Hafizah Husain

Papers

Mel frequency cepstral coefficients (Mfcc) feature extraction enhancement in the application of speech recognition: A comparison study

How to Construct Open Ended Questions

Hybrid affective computing--keyboard, mouse and touch screen: from review to experiment

Fuzzy Model in Human Emotions Recognition.

Fuzzy model of dominance emotions in affective computing