Yashwant Joshi

Marathi numeral recognition using statistical distribution features

Optical character recognition (OCR) is very old topic of research still needs much improvement. OCR system deals with the machines, so the new methodologies are to be found out to increase the overall performance of the character recognition system. This paper deals with the recognition of Devanagari (Marathi) numerals. After preprocessing the image and converting it to binary the statistical features for individual numeral can be calculated. Based on these features the numerals are classified into appropriate groups. In classification method using histogram feature matching, the numbers like three and six can be misclassified as they are having similar histogram. In the proposed system overall performance of classification can be improved if more number of features are compared. The system used in this paper is better as compared to simple histogram matching criteria.

Handwritten Numeral Identification System Using Pixel Level Distribution Features

In this paper, pixel level features of the character are used for Devanagari numeral Recognition. The pixel distribution features for each numeral can be calculated after preprocessing the document image and converting it to binary. Based on these features the numerals are classified into appropriate groups. Histogram feature matching method gives erroneous results for the numbers like one and nine as they are having nearly similar histogram. In the proposed approach pixel distribution features are extracted in four directions. The overall performance of classification can be improved if more number of features is compared. The proposed approach gives improved results as compared to simple histogram matching criteria.

Tropical littoral ambient noise probability density function model based on sea surface temperature

Ambient noise variability in tropical shallow water presents a critical challenge for sonar designers and operators due to site-specific sea surface fluctuations. Sea surface temperature (SST) is a direct measure of energy balance defining the local climate of the region and hence ambient noise characteristics. In this work, an ambient noise probability density function (pdf) model for a spectral band of 3-10 kHz has been designed based on the statistical distribution of SST and validated using real field data. This will enable early ambient noise prediction compared to existing wind speed based models to facilitate structured mitigation strategies for improving sonar performance.

Study of Variation in Ambient Noise with Fluctuations of Surface Parameters for the Indian Ocean Region

Ambient noise variability is a critical challenge encountered by multiple stakeholders, including sonar designers and operators. Among the sources of ambient noise in the ocean, wind related noise has significant impact on sonar performance. The tropical waters in the Indian Ocean Region (IOR), present random fluctuations in the surface parameters, namely the wind speed, surface temperature, wave height, etc. resulting in variations in the ambient noise characteristics. The site-specific surface fluctuations in the tropical regions restrict the possibility of generalized algorithm design to mitigate the ambient noise impact. The work attempts to study the variations in the ambient noise levels corresponding to the fluctuations in the surface parameters. The site-specific behavior of the tropical IOR is demonstrated using surface data available from moored buoy at three distinct locations of the IOR. The analysis methodology can be used to characterize, predict and improve sonar performance, particularly in severe conditions of the tropical IOR.

Electroencephalography-Based Emotion Recognition Using Gray-Level Co-occurrence Matrix Features

Emotions are very essential for our day-to-day activities such as communication, decision-making and learning. Electroencephalography (EEG) is a non-invasive method to record electrical activity of the brain. To make Human–Machine Interaction (HMI) more natural, human emotion recognition is important. Over the past decade, various signal processing methods are used for analysing EEG-based emotion recognition (ER). This paper proposes a novel technique for ER using Gray-Level Co-occurrence Matrix (GLCM)-based features. The features are validated on benchmark DEAP database upto four emotions and classified using K-nearest neighbor (K-NN) classifier.

Validation of Webster ambient noise model for real data in tropical littoral water

Ambient noise has a profound impact on the performance of underwater systems, specifically in the tropical littoral waters. The site specific nature presents its unique challenge and efforts at mitigation have not been effective. Synthetic ambient noise generation has emerged as an effective Modeling and Simulation (M&S) tool for Signal to Noise Ratio (SNR) enhancement. Webster model of random number generation with implicit ambient noise power spectral density structure has been reported to simulate synthetic noise. The power spectral estimation of ambient noise in the tropical littoral waters has always been sub-optimal due to random fluctuations. This work presents the frequency sampling method of FIR filter design for the accurate generation of colored spectra for improved tropical littoral ambient noise generation using the Webster model. This effort focuses on the efficiency of Webster model with expected Power Spectral Density (PSD), kurtosis and the Probability Distribution Function (PDF). The synthetic data generated using the proposed model has been validated using real data recorded by fixed sensors off the west coast of India with a mix of dominance of shipping and wind noise. The recording site has a unique combination of a port being close by, to incorporate dominance of shipping noise in the calm months and the wind noise dominating the spectrum during rough weather condition.

Joint Compression and Classification for Textures in the Wavelet and Ridgelet Domain

Image Compression is a widely addressed research area. Many compression standards are in place. There are many methods for image classification. But the joint compression and classification is a new research area wherein the classification is attempted in the compressed domain. The joint compression and classification (JCC) is explored in wavelet domain by some researchers. But it is not yet explored in Ridgelet domain. This paper discusses the performance of JCC for Wavelet and Ridgelet domain for Texture images. The experimentation is done with objective analysis and subjective analysis. Objective analysis is performed using the Compression metrics-RMSE, PSNR and classification metric- CCR. Subjective analysis is performed using Human Visual Perception. It is found that the Ridgelet Transform gives less Mean Squared Error (MSE) and is better for Joint Compression and Classification of Texture images. Extensive experimentation has been carried out to arrive at the conclusion.

Statistical Characterization of an Underwater Channel in a Tropical Shallow Freshwater Lake System

Underwater acoustics has made significant strides over the last century, which finds applications over a wide range from basic bathymetry study to high-end research extensions. The acoustic propagation in underwater is typically governed by physical properties of the underwater channel, such as temperature, pressure, and salinity. The seasonal fluctuations in the physical properties of the tropical region manifest as thermal stratification. The random thermal stratification has a significant impact on the Sound Speed Profile (SSP), thereby distorting the received echoes from the surface and the bottom. The site-specific behavior in the tropical region makes it an interesting research problem to investigate the correlation of the surface parameters like temperature with the surface and bottom reflection due to variations in the SSP. In this work, we attempt to present underwater channel characteristics of the tropical freshwater lake system at Khadakwasla (18.43° N, 73.76° E), located in the municipal limits of Pune city in India. The temperature gradient along the water column is computed using the one-dimensional Freshwater Lake Model (FLake) to derive the SSP using Medwin relation. The statistical analysis of the sound speed fluctuations resulted due to seasonal variation in the water temperature is presented using the Kolmogorov–Smirnov (KS) Goodness-of-Fit test is used to find a close Probability Density Function (pdf) match for the surface and the bottom path impulse response. The results indicate a good match of the surface and bottom path impulse response with Weibull distribution with a high confidence level. Such characterization can facilitate the design of adaptive algorithms to minimize the underwater channel impact based on a precise estimate of the channel impulse response.

Nonuniform Frequency Sampling Approach to FIR Filter Design

This paper investigates the new approach to FIR filter design based on nonuniform frequency sampling. This method generates the nonuniform samples in passband and stopband separately using Gaussian function. For the generated nonuniform sample, the desired frequency response values are generated using ideal filter characteristics. Then, taking its nonuniform IDFT gives the required filter coefficients. The proposed method is compared with existing methods like uniform frequency sampling and optimal filter design method and results show that the investigated approach has a better advantage over uniform frequency sampling and Parks–McClellan method with regard to the frequency response of designed filter.

Analyzing Effect of Meditation Using Higher Order Crossings and Functional Connectivity

People are experiencing difficulties in adapting to the rapid changes in work and social fabric due to the evolution of advanced technologies in everyday life. Health and well-being of an individual in the existing world is important for proper living. Meditation improves the adaptability of an individual to live a healthy and social life. To verify this, an experiment is designed with the simple meditation practice called Focused Attention for 8 weeks. The brain activity is recorded of 11 subjects using EMOTIV EPOC+ EEG device before (pre-meditation) and after (post-meditation) meditation. Features called Higher Order Crossings and Functional Connectivity are used to analyze the effect of meditation. The results indicated a decrease in HOC values for frontal, parietal, and occipital lobes and increase in HOC of temporal lobe. The interhemispheric connectivity increased after meditation practice.