Tata Consultancy Services Limited

Arijit Chowdhury

Tapas Chakravarty

Arpan Pal

Balamuralidhar Purushothaman

Avik Ghose

Smriti Rani

Tanushree Banerjee

Anwesha Khasnobish

Arijit Sinharay

Nasimuddin Ahmed

Shalini Mukhopadhyay

Debatri Chatterjee

Rajat Kumar Das

Dibyanshu Jaiswal

Varsha Sharma

Taniya Das

Soumya Chakravarty

Soma Bandyopadhyay

Arijit Mukherjee

Chirabrata Bhaumik

Rohan Banerjee

Achanna Anil Kumar

Anirban Dutta Choudhury

Sakyajit Bhattacharya

Sounak Dey

Rahul Dasharath Gavas

Arun George

Shreyasi Datta

Vivek Chandel

Dibyendu Roy

Kayapanda Muthana Mandana

Raj Kumar Rakshit

Andrew Gigie

Ramu Reddy Vempada

Arindam Ray

Rahul Sinha

Dighanchal Banerjee

Sonia Sharma

Tapas Chakravar

Ghose Avik

This disclosure relates generally to action recognition and more particularly to system and method for real-time radar-based action recognition. The classical machine learning techniques used for learning and inferring human actions from radar images are compute intensive, and require volumes of training data, making them unsuitable for deployment on network edge. The disclosed system utilizes neuromorphic computing and Spiking Neural Networks (SNN) to learn human actions from radar data captured by radar sensor(s). In an embodiment, the disclosed system includes a SNN model having a data pre-processing layer, Convolutional SNN layers and a Classifier layer. The preprocessing layer receives radar data including doppler frequencies reflected from the target and determines a binarized matrix. The CSNN layers extracts features (spatial and temporal) associated with the target's actions based on the binarized matrix. The classifier layer identifies a type of the action performed by the target based on the features.

System and method for real-time radar-based action recognition using spiking neural network(SNN)

This disclosure relates generally to material quality inspection. Conventional approaches available for material quality inspection are unable to address concerns of complexity and cost involved. The technical problem of occluded object detection and material quality inspection for intrinsic defects identification is addressed in the present disclosure. The present disclosure provides a system and method for non-intrusive material quality inspection using three-dimensional monostatic radar based imaging, where the object under inspection undergoes a circular translation motion on a rotating platform. A modified delay-and-sum (m-DAS) algorithm is built by incorporating virtual antenna array to obtain a 3D image reconstruction of the object. From 3D reconstructed images, radial displacement as well as the angular locations of the object is identified which are further used for quality inspection of the material comprised in the object.

Systems and methods for non-intrusive material quality inspection using three- dimensional monostatic radar based imaging

An e-commerce business model has witnessed several cases where packages with faulty goods, returned by buyers, without procured object, rather replaced by different device. This disclosure relates a method to detect whether object under test a desired object. A plurality of back-scattered signals is received from the object under test occluded by packaging with continuous motion on conveyer based on first antenna-radar combination. The plurality of back-scattered signals is processed by applying four-tap difference filter to obtain motion-filtered data matrix. A low pass filter is applied on the motion-filtered data matrix to obtain enveloped motion-filtered data matrix. A sliding constant false alarm rate is applied on the enveloped motion-filtered data matrix to determine detection threshold value. A check is performed to detect whether the object under test is the desired object based on whether intensity of the plurality back-scattered signals exceeds the detection threshold value.

Method and system for detection of a desired object occluded by packaging

This disclosure relates generally to a method and system for online handwritten signature verification providing a simpler low cost system. The method comprises extracting signature data for the subject from a sensor array for the predefined time window at regular predefined time instants. Further, differentiating the matrix row wise and column wise to generate a row difference matrix and a column difference matrix. Further, determining an idle signature time fraction for the extracted signature data of the subject being monitored from the column difference matrix. Further, determining a plurality of signature parameters based on the row difference matrix and the column difference matrix. Further, analyzing the idle signature time fraction and the plurality of signature parameters of the subject being monitored based on a Support Vector Machine (SVM) classifier, wherein the SVM classifier performs online classification of the extracted signature data into one of a matching signature class and a non-matching signature class.

Method and system for handwritten signature verification

This disclosure relates to systems and methods for shapelet decomposition based recognition using radar. State-of-the-art solutions involve use of standard machine learning classification techniques for gesture recognition which suffer with problem of dependency on collected data. The present disclosure overcome the limitations faced by the state-of-the-art solutions by obtaining a plurality of time domain signal using a radar system comprising three vertically arranged radars and one or more sensors, identifying one or more gesture windows to obtain one or more shapelets corresponding to one or gestures which are further decomposed into a plurality of sub shapelets. Further, at least one of (i) a positive or (i) a negative time delay is applied to each of the plurality of sub shapelets to obtain a plurality of composite shapelets which are further mapped with a plurality of trained shapelets to recognize gestures comprised in one or more activities performed by a subject.

Systems and methods for shapelet decomposition based gesture recognition using radar

Temperature measurement is an important part of many potential applications in process industry. Conventional temperature measurement methods require manual intervention for process monitoring and fail to provide accurate and precise measurement of temperature of an enclosed mixed fluid chamber. The present disclosure provides artificial intelligence based temperature measurement in mixed fluid chamber. A plurality of inputs pertaining to the mixed fluid chamber are received to build a fluid based model. The fluid based model is used to generate one or more fluid parameters. The one or more fluid parameters are used along with a ground truth temperature data and the received plurality of inputs for training an artificial intelligence (AI) based model. However, the AI based model is trained with and without knowledge of fluid flow. The trained AI based model is further used to accurately estimate temperature of the mixed fluid chamber for a plurality of test input data.

Artificial intelligence based temperature measurement in mixed fluid chamber

Recognizing mental states from physiological signal is a concern not only for medical diagnostics, but also for cognitive science, behavioral studies as well as brain machine interfaces. Embodiments of the present disclosure utilize respiration signals to decipher mental states wherein non-linear baseline drifts in signal is implemented to extract the respiratory features in most effective way. Presence of class imbalance, is effectively rectified using Synthetic Minority Oversampling Technique (SMOTE) to resolve class imbalance problem, which not only increased the classification accuracy, but also reduced classifier bias towards the majority class, which in turn exceedingly enhanced the classifier sensitivity.

Stress level monitoring of users using a respiratory signal and alerting thereof

This disclosure relates generally to a unified platform for domain adaptable human behaviour inference. The platform provides a unified, low level inference and high level inference of domain adaptable human behaviour inference. The low level inferences include cross-sectional analysis techniques to infer location, activity, physiology. Further the high inference that provide useful and actionable for longitudinal tracking, prediction and anomaly detection is performed based on several longitudinal analysis techniques that include welch analysis, cross-spectrum analysis, Feature of interest (FOI) identification and time-series clustering, autocorrelation-based distance estimation and exponential smoothing, seasonal and non-seasonal models identification, ARIMA modelling, Hidden Markov models, Long short term memory (LSTM) along with low level inference, human meta-data and application domain knowledge. Further the unified human behaviour inference can be obtained across multiple domains that include health, retail and transportation.

Unified platform for domain adaptable human behaviour inference

The disclosure herein generally relates to the field of determination of cardiopulmonary signals for multi-persons, and, more particularly, to determination of cardiopulmonary signals for multi-persons using in-body signals obtained by ultra-wide band (UWB) radar. The disclosed method determines of cardiopulmonary signals for multi-persons using in-body signals, wherein a UWB radar signals/waves reflected from inside a human body is utilized for efficient determination of cardiopulmonary signals. The disclosed method and system utilize the UWB radar signals to identify a number of persons along with several details about the persons that include a girth of the each identified person and the orientation of the identified person towards the one or more UWB radar. Further a chest wall distance, a breathing rate, a heart wall distance and a heart rate are determined for all the identified persons based on the identified girth and the identified orientation along with the UWB radar signals.

Determination of cardiopulmonary signals for multi-persons using in-body signals obtained by UWB radar

Reckless behavior of drivers like, speeding, sudden acceleration and swerving through lanes can cause fatality and financial loss. Conventional methods mainly focus on driving style classification. The conventional methods mainly focus on driver classification and are not able to provide trip classification of a driver. Hence there is a challenge in trip classification of the driver based on acceleration data. The present disclosure for trip classification addresses the problem of end to end trip classification based on the acceleration data. Here, a journey is segmented into a plurality of sub-journey segments and each sub-journey segment is associated with a plurality of driving events. An event score is calculated for each sub-journey and a normalization is performed on the event score. Further, the journey is classified into at least one of good, average or bad based on the normalized data by utilizing a fuzzy based classification.

Method and system for trip classification

This disclosure relates generally to assessment of cognitive workload using breathing pattern of a person, where cognitive workload is the amount of mental effort required while doing a task. The method and system provides assessment of cognitive workload based on breathing pattern extracted from photoplethysmograph (PPG) signal, which is collected from the person using a wearable device. The PPG signal collected using the wearable device are processed in multiple stages that include breathing signal extraction to extract breathing pattern. The extracted breathing pattern is used for assessment of cognitive workload using a generated personalized training model, wherein the personalized training model is generated and dynamically updated for each person based on selection of a sub-set of breathing pattern features using feature selection and classification techniques that include maximal information coefficient (MIC) techniques. Finally based on personalized training model, the extracted breathing pattern is classified as high cognitive workload or low cognitive workload.

Method and system for assessment of cognitive workload using breathing pattern of a person

While performing heart rate estimation of a user, if the user is in motion, a signal is measured and is likely to have noise data, which in turn affects accuracy of estimated heart rate value. Method and system for heart rate estimation when the user is in motion is disclosed. The system estimates value of a noise signal present in a measured PPG signal by performing a Principal Component Analysis (PCA) of an accelerometer signal collected along with the PPG signal. The system further estimates value of a true cardiac signal for a time window, based on value of the true cardiac signal in a pre-defined number of previous time windows. The system then estimates frequency spectrum of a clean PPG signal based on the estimated noise signal and the true cardiac signal. The system further performs heart rate estimation based on the clean PPG signal.

Method and system for heart rate estimation of a user

This disclosure relates generally to detection of mild cognitive impairments in subjects. The method and system proposed provides a continuous/seamless monitoring platform for MCI detection in subjects by continuously monitoring routine activities of subjects (Activities of Daily Living (ADL)) in a smart environment using plurality of passive, unobtrusive, binary, unobtrusive non-intrusive sensors embedded in living infrastructure. The proposed method and system detects symptoms of MCI at the onset of the disease, while also addressing issue of sensor failures that causes gaps in the data. The collected sensor data is pre-processed in several stages which includes which includes pre-processing of sensor data, behavior deviation detection, and abnormality detection and so on. Further, the disclosure also proposes an autoencoder based technique, to reduce the dimension of the data to find personalized deviations in behavior of a subject which is used to detect if a subject could be a potential case of MCI.

System and method for early detection of mild cognitive impairment in subjects

This disclosure relates generally to tracking motion of target in indoor environment. The method includes estimating an initial position of the target in a mesh grid form based on radar data captured from radar devices installed in the indoor environment. For a subsequent target movement, a subsequent position of the target is estimated in the mesh grid form based on the initial position and a resultant velocity vector of the target. A number of outlier grid-points is computed with a threshold number, and based on comparison the outlier grid-points are either replaced with interpolated grid-points or the subsequent position of the target is repaired based on a probable position of the target obtained from at least one of a linear regression based analysis of prior positions of the target, prior knowledge of the target velocity and sampling interval, and a trilateration based technique.

System and method for tracking motion of target in indoor environment

A system and method for estimating blood pressure (BP) using photoplethysmogram (PPG) has been explained. The PPG is captured from a PPG sensor (). For preparing training model, a pulse oximeter is used for capturing PPG. For testing, a smartphone camera is used for capturing the PPG signal. A plurality of features are extracted from the preprocessed PPG signal. A BP distribution is then generated using the plurality of features and the training model. The BP distribution is part of a set of BP distributions generated from different subjects. Finally, a post-processing methodology have been used to reject inconsistent data out of the set of BP distributions and BP value is estimated only for the remaining BP distributions and a statistical average is provided as the blood pressure estimate.

Growing community of inventors

Kolkata, India

Arijit Chowdhury