Patent publication number: United States Patent 8,793,118 (application #13/330,712) Published on: 29th July, 2014 (Filed on 20th December, 2011)

Research Domain: Center for Pattern Recognition

The invention is a system that permits visually and/ or hearing impaired persons to communicate with each other and with people who have normal sight and hearing. It accepts input in one of the multiple modes: audio (speech), video (gestures) or text and relays the output in any one or more of the three modes. If the input and output modes are different, the system processes the input to convert it from one format to another using gesture, speech or text recognition systems (as applicable) that are based on machine learning algorithms. The system permits the user to define their own constructs and dictionaries, thus its use extends beyond standard Indian and International languages. Further, it adapts to the user’s proficiency level, adjusting the input and output rates to the speed at which the user processes the input/ output. Since a novice user may become proficient with time, the system is designed to continuously adapt to the changing proficiency of the user.

Provisional application filed

Research Domain: Intelligent Systems

This project is a step to bridge the gap between autistic children and the world and to give them a sense of acceptance in the society. It is a wearable device in the form of a wrist band with an array of sensors to capture the physiological parameters of subjects. These parameters are analysed to recognize different emotions like: happy, sad and involvement. The project is based on the fact that,” every emotional change has a physiological basis to it”. This project is carried out with the help of medical experts in the department of Child and Adolescent Psychiatry, NIMHANS, Bangalore.

Patent application number 3410/CHE/2014, Patent Filed on 09-07-2014

Research Domain: Signal Processing

The present subject matter relates to the use of set of binary Support Vector Machines (SVM) classifiers to represent and implement combinational circuit design. The combinational functionality is achieved by using a single binary SVM classifier for each bit of the output of the digital combinational function and each of these SVM classifiers would have the multiple inputs of the blocks. The parameters that characterize each of the SVM classifiers for that output bit is loaded into the memory of the SVM classifier apriori to the implementation for use. The functional aspect of the combined functional SVM classifier set is thus a function of the parameters loaded. Each functional combinational logic is to be achieved would only be characterized by the parameter set and the number of SVM classifiers used. A set of functional switches on the memory selection and output selection suitably synchronized over time can enable one binary SVM classifier to have multiple digital combinational functions.

Patent application number 2378/CHE/2014, Patent Filed on 12-05-2014

Research Domain: Networks and Security

This project aims at providing the end users with an innovative way to share resources across cross platform devices using internet as a medium of transfer and gesture based user interface. The main objective is to provide convenience to users to transfer contents by using one simple gesture rather than using additional hardware devices, emails or other such approaches. The project intends to build an application on different platforms which can share documents, media files and links over the internet using hand gestures. The main aim is to provide easy interface for user through gestures to share previously mentioned resources. Thus enables user to share resources without any external hardwares like pen drives and hard disks. Also usage of emails, chat features or other social sites can be avoided to share links and files.

Inconel 718 Titanium composites coating for plunger in bio-fuel industries to enhance life of plunger.

Patent application number 3410/CHE/2014, Patent Filed on 09-07-2014

Research Domain: Advanced Composites Research Center