A respiratory disorder that attacks COVID-19 patients requires intensive supervision of medical practitioners during the isolation period. non-contact monitoring device will be a suitable solution for reducing spread risk virus while patient. This study uses Frequency-Modulated Continuous Wave (FMCW) radar and Machine Learning (ML) to obtain information analyze signals, respectively. Multiple subjects in room can detected simultaneously by calculating Angle Arrival (AoA) received signal...

This paper presents a (language-independent) method of locating rectangular text regions in natural scene images. The consists two steps that can be applied succession or independently: the frequency edge pixels across vertical and horizontal scan lines, fundamental Fourier domain. feature images is highly intuitive, this focus research. detection rectangles using Hough transform also addressed. Texts are meaningful to many viewers usually appear colours high contrast background. Hence it...

Online recognition of cursive words is a difficult task owing to variable shape and ambiguous letter boundaries. The approach proposed based on hidden Markov modeling letters inter-letter patterns called ligatures occurring in script. For each the we create one HMM that models temporal spatial variability handwriting. By networking two kinds HMMs, can design network model for all or composite characters. incorporates knowledge sources grammatical structural constraints so it better capture...

In this paper, we describe a dynamic Bayesian network or DBN based approach to both two-hand gestures and one-hand gestures. Unlike wired glove-based approaches, the success of camera-based methods depends greatly on image processing feature extraction results. So proposed method DBN-based inference is preceded by fail-safe steps motion tracking. Then new gesture recognition model for set framework which makes it easy represent relationship among features incorporate information model. an...

To pursue a healthy lifestyle, people are increasingly concerned about their food ingredients. Recently, it has become common practice to use an online recipe select the ingredients that match individual’s meal plan and diet preference. The information from recipes can be extracted used develop various food-related applications. Named entity recognition (NER) is often extract such information. However, problem in building NER system lies massive amount of data needed train classifier,...

The idea of combining the network HMMs and dynamic programming-based search is highly relevant to online handwriting recognition. word model HMM can be systematically constructed by concatenating letter ligature HMM's while sharing common ones. Character recognition in such a defined as task best aligning given input sequence path network. One distinguishing feature approach that segmentation obtained simultaneously with but no extra computation required.

In this paper, we propose a new gesture recognition model for set of both one-hand and two-hand gestures based on the dynamic Bayesian network framework which makes it easy to represent relationship among features incorporate information model. Unlike coupled HMM, proposed has room common hidden variables are believed be shared between two variables. an experiment with ten isolated gestures, obtained rate upwards 99.59% leave-one-out cross validation. The is have strong potential successful...

A statistical approach to recognizing on-line cursive Hangul character is proposed. Viewing a handwritten syllable as an alternating sequence of letters and ligatures, all legal characters are modeled with finite state network that concatenation letter ligature HMMs. Given input the network, recognition, which corresponds finding most likely path, performed using dynamic programming technique. Experiments have shown boundary detection well handwriting variability resolution achieved good...

Efficient Market Hypothesis (EMH), states that at any point in time a liquid market security prices fully reflect all available information. This paper presents study of proving the hypothesis through daily Twitter sentiments using hybrid approach lexicon-based and naive Bayes classifier. In this research we analyze currency exchange rate movement Indonesia Rupiah vs US dollar as way testing Hypothesis. order to find correlation between prediction from data actual trends collect every day...

This paper presents a method for computer-assisted slide presentation using vision-based gesture recognition. The proposed consists of sequence steps, first detecting hand in the scene projector beam, then estimating smooth trajectory or pointing finger Kalman Filter, and finally interfacing to an application system. Additional navigation control includes moving back forth pages presentation. is help speakers effective with natural improved interaction computer. In particular, believed be...

In this paper we consider a hidden Markov mesh random field (HMMRF) for character recognition. The model consists of "hidden" (MMRF) and an overlying probabilistic observation function the MMRF. Just like 1-D HMM, layer is characterized by initial transition probability distributions, defined distribution functions vector-quantized (VQ) observations. HMMRF-based method two phases: decoding training. training algorithms are developed using dynamic programming maximum likelihood estimation...

Age, gender, and race classification pose challenges across various domains, including computer vision, social sciences, marketing. Despite advancements in deep machine learning, convolutional neural networks (CNNs) remain crucial for addressing these tasks. This paper introduces an innovative approach utilizing CNNs with residual blocks to enhance accuracy efficiency age, classification. Incorporating connections enables the model capture both low-level high-level features, improving while...

It is well known that the stochastic approach using HMM and dynamic programming-based search particularly suited to analysis of time series signals including on-line handwriting. The starting point this research a network HMMs which models whole set characters. Then it followed by assertion for script can be applied not only character recognition but also handwriting synthesis even pen-trajectory recovery in off-line images. solutions these problems are based on single principle DP-based...

In this paper, we propose a novel method for analyzing human interactions based on the walking trajectories of subjects. Our principal assumption is that an interaction episode composed meaningful smaller unit interactions, which call `sub-interactions.' The whole represented by ordered concatenation or network sub-interaction models. From experiments, could confirm effectiveness and robustness proposed internal work comparing performance with other previous approaches.

The collective health clinic data of people in a society is surmised to have variety characteristic states and certain dynamics governing state changes over time. Given such collection samples, we propose way estimating set dynamic context using the tool hidden Markov model (HMM). We also present method predicting future based on or duration statistics derived from parameters clusters. In proposed design number HMMs, each for sequences with particular disease history interest. They are used...

Mastering a musical instrument for an unskilled beginning learner is not easy task. It requires playing every note correctly and maintaining the tempo accurately. Any music comes in two forms, score it rendition into audio music. The proposed method of assisting players both aspects employs popular pattern recognition methods audio-visual analysis; they are support vector machine (SVM) hidden Markov model (HMM) performance tracking. With proper synchronization results, learning assistant...