Tata Consultancy Services Limited

Shailesh Shankar Deshpande

Balamuralidhar Purushothaman

Jayavardhana Rama Gubbi Lakshminarasimha

Mahesh M Kshirsagar

Sonam Yashpal Sharma

Mariswamy Girish Chandra

Girish Keshav Palshikar

Vinay Vasant Kulkarni

Angshul Majumdar

Asha Rajbhoj

Karthik Seemakurthy

Shivani Nigam

Ashley Varghese

Athiappan G

Rohit Thakur

Sandeep Nk

Chaman Banolia

Apoorva Khanna

Mariaswamy Girish Chandra

Shashwat Shahi

Gargi Uday Kulkarni

Shamsuddin Nasiruddin Ladha

Piyush Yadav

Karan Sharad Owalekar

Karthikeyan Vaiapury

Manish Kausik Hari Baskar

Lakshminarasimha Jayavardhana Rama Gubbi

Assessing sustainability performance of an enterprise is a challenging task. Embodiments of present disclosure provide a method and system for SDG performance assessment of an enterprise with a conceptually simpler data model and processing pipeline. Enterprise data collected from hard and soft sensors is mapped to appropriate indicators of the SDGs. Further, a semantic network is constructed with nodes corresponding to each indicator and edges connecting nodes belonging to same SDG. Each node of the semantic network is further linked to a first layer of a neuro fuzzy network which calculates degree of impact of the indicator on Social, Economic and Environment values. Output of the first layer activates second layer of the neuro fuzzy network which determines BBV scores indicating whether the indicator is a burden, benefit, or vulnerability. The BBV scores are transformed to a colour space to generate a colour that indicates SDG performance of the enterprise.

Method and system for sustainable development goal (SDG) performance assessment of an enterprise

Embodiments herein provide a method and system for a hyperspectral artificial vision for machines. The system receives a hyperspectral signal of a target material as an input to a neural network model. The system initializes by selecting the number of primitive layers to be used. The system iteratively cycles through all training data (pixels) and updating weights for each unsuccessful material class prediction. Model with two primitives serves as baseline, after which the system adds another primitive layer and repeats the training procedure. The system keeps repeating these processes until obtains convergence. Where the system come to a halt, the system obtains the optimal number of primitives for the given materials. The generated new color pixel is used as a discriminator to aid in locating the target material. The new artificial color is a mixture of weighted chromatic primitives which are optimized for sensitivity/(Spectral Response Functions) SRFs.

System and method for generating hyperspectral artificial vision for machines

The embodiments of present disclosure herein address problem of urban metabolism with respect to water demand and carbon dioxide emissions, the discussion is based on the reported data by the urban areas. The embodiments herein provide a method and system for estimating urban metabolism based on remotely sensed data. The system is configured to develop a model for identifying correct features from image or proxy features from image and then develop/use relation between the image feature or proxy feature from the image with the urban metabolic parameter. Further, the system develops an urban growth model which predicts spatial extent of the given proxy features. The urban growth scenario for each such conditions is different. By changing conditions of the model, different growth scenarios are played out. For each scenario, at least one urban metabolic parameter is predicted by taking output of the urban growth predictor.

Method and system for estimating urban metabolism

Quality of weld images in with bad lighting condition and specific image color formats add constraints to existing automated weld inspection systems. Embodiments herein provide a method and system based on Domain Knowledge Infused Adaptive-Network-based Fuzzy Inference System (DKI-ANFIS) for weld quality inspection. The DKI-ANFIS inspects the quality of weld joint using domain driven quality inspection techniques. A segmentation algorithm is used to extract the weld joint in form of fractals followed by an unsupervised technique to extract useful geometrical features from the fractals. These geometrical features are used for quality index generation. A weld inspection model comprising the DKI-ANFIS is used for determining the quality of the weld joint. DKI-ANFIS modifies layers of ANFIS by infusing layer of domain knowledge to give better results even if there is a class imbalance in the data or the data is skewed or there is only a short corpus of data available.

Weld quality inspection with domain knowledge infused adaptive-network-based fuzzy inference system

State of art techniques performing image labeling of remotely sensed data are computation intensive, consume time and resources. A method and system for efficient retrieval of a target in an image in a collection of remotely sensed data is disclosed. Image scanning is performed efficiently, wherein only a small percentage of pixels from the entire image are scanned to identify the target. One or more samples are intelligently identified based on sample selection criteria and are scanned for detecting presence of the target based on cumulative evidence score Plurality of sampling approaches comprising active sampling, distributed sampling and hybrid sampling are disclosed that either detect and localize the target or perform image labeling indicating only presence of the target.

Efficient retrieval of a target from an image in a collection of remotely sensed data

The embodiments herein provide a method and system that analyzes the pixel vectors by transforming the pixel vector into two-dimensional spectral shape space and then perform convolution over the image of graph thus formed. Method and system disclosed converts the pixel vector into image and provides a DCNN architecture that is built for processing 2D visual representation of the pixel vectors to learn spectral and classify the pixels. Thus, DCNN learn edges, arcs, arcs segments and the other shape features of the spectrum. Thus, the method disclosed enables converting a spectral signature to a shape, and then this shape is decomposed using hierarchical features learned at different convolution layers of the disclosed DCNN at different levels.

Method and system for learning spectral features of hyperspectral data using DCNN

This disclosure relates to automatic creation of an image processing pipeline that has a dependency on image processing domain experts. Given the image processing problem has multiple viable solutions, search space of creating a solution for a given goal in a given constrained infrastructure is large and, choosing an optimal image processing solution is an effort, time and intellect intensive endeavor. The present disclosure models image processing goals in abstract task categories and externalizes domain knowledge as an object model and a set of rules defined over it. Rules codify task template selection criteria and algorithm selection criteria for implementing a task. Recommendations are provided to choose suitable algorithm (s) for carrying out image processing tasks. Effects of algorithm execution are recorded for subsequent tasks recommendations. The stored state of execution and trace information enables rerouting the execution and knowledge upgrade. Deployable code is generated after a validated pipeline is created.

Systems and methods for automatically creating an image processing pipeline

This disclosure relates generally to image processing, and more particularly to method and system for image reconstruction using deep dictionary learning (DDL). The system collects the degraded image as test image and processes the test image to extract sparse features from the test image, at different levels, using dictionaries. The extracted sparse features and data from the dictionaries are used by the system to reconstruct the HR image corresponding to the test image.

Method and system for solving inverse problems in image processing using deep dictionary learning (DDL)

The present disclosure provides a method and a system for optimizing Hidden Markov Model based land change prediction. Firstly, remotely sensed data is pre-processed and classified into a plurality of land use land cover classes (LULC). Then socio-economic driver variables data for a pre-defined interval of time are provided from a database. A Hidden Markov Model (HMM) is defined with LULC as hidden states and socio-economic driver variables data as observations and trained for generating a MINI state transition probability matrix. Again the defined MINI is trained by taking input data from scenario based temporal variables to generate another set of HMM state transition probability matrix. The generated MINI state transition probability matrix is then integrated with a spatio-temporal model to obtain an integrated model for predicting LULC changes to generate at least one prediction image.

Methods and systems for optimizing hidden Markov Model based land change prediction

A system and method for identifying real time change in a scene of an unknown environment using an unmanned vehicle. In the context of unmanned vehicle navigation, it is critical to calculate the saliency map in real time and employ them in scene understanding. This will reduce the search space and ensure that the process is quicker. A domain specific ontology is created and a saliency model is developed. The saliency model detects key domain specific regions of interest in two consecutive images. The regions of interest is used for registration and change detection. The system is detecting the change by using visual saliency as an abstract feature that is developed in the environment. Probability of change is derived using the salient maps of the two images.

System and method for detecting change using ontology based saliency

System and method of the present disclosure provide a linguistic approach to image processing. Prior art focused on extracting well-defined single objects occupying large portion of an image area. However, there was no focus on higher level semantics or distribution of object categories within the image. In contrast to imagery by handheld devices, remotely sensed data contains numerous objects because of relative large coverage and distribution over objects is critical to analyzing such large coverage. Accordingly, in the present disclosure, a generative statistical model is defined wherein an aerial image is modelled as a collection of the one or more themes and each of the one or more themes is modelled as a collection of object categories. The model automatically adapts to a scale of the aerial image and appropriately identifies the themes which may be used for applications including monitoring land use, infrastructure management and the like.

System and method for theme extraction

Disclosed is a method and system for creating labels for cluster in computing environment. The system comprises receiving module, candidate items selector, combination array generator, coverage value analyzer, candidate pair selector, unique word filter and cluster label selector. Receiving module receives input data and candidate items selector selects candidate items occurring repetitively using n-gram technique to generate list of candidate items with frequency of occurrence. Combination array generator selects candidate items to populate two-dimensional array wherein each array element represents pair of n-gram. Coverage value analyzer determines coverage value for each pair of n-gram from array. Candidate pair selector selects pairs of n-gram from two-dimensional array to process and generate list of candidate pairs. The unique word filter determines number of unique words in each candidate pair. Cluster label selector sorts list of candidate pairs using coverage value and number of unique words to select cluster label.

Growing community of inventors

Bengaluru, India

Shailesh Shankar Deshpande