International Institute of Information Technology, Hyderabad

About: International Institute of Information Technology, Hyderabad is a education organization based out in Hyderabad, India. It is known for research contribution in the topics: Authentication & Internet security. The organization has 2048 authors who have published 3677 publications receiving 45319 citations. The organization is also known as: IIIT Hyderabad & International Institute of Information Technology (IIIT).

RespireNet: A Deep Neural Network for Accurately Detecting Abnormal Lung Sounds in Limited Data Setting

Abstract: Auscultation of respiratory sounds is the primary tool for screening and diagnosing lung diseases. Automated analysis, coupled with digital stethoscopes, can play a crucial role in enabling tele-screening of fatal lung diseases. Deep neural networks (DNNs) have shown a lot of promise for such problems, and are an obvious choice. However, DNNs are extremely data hungry, and the largest respiratory dataset ICBHI has only 6898 breathing cycles, which is still small for training a satisfactory DNN model. In this work, RespireNet, we propose a simple CNN-based model, along with a suite of novel techniques -- device specific fine-tuning, concatenation-based augmentation, blank region clipping, and smart padding -- enabling us to efficiently use the small-sized dataset. We perform extensive evaluation on the ICBHI dataset, and improve upon the state-of-the-art results for 4-class classification by 2.2%

Private blockchain-envisioned security framework for AI-enabled IoT-based drone-aided healthcare services

Abstract: Internet of Drones (IoD) architecture is designed to support a co-ordinated access for the airspace using the unmanned aerial vehicles (UAVs) known as drones. Recently, IoD communication environment is extremely useful for various applications in our daily activities. Artificial intelligence (AI)-enabled Internet of Things (IoT)-based drone-aided healthcare service is a specialized environment which can be used for different types of tasks, for instance, blood and urine samples collections, medicine delivery and for the delivery of other medical needs including the current pandemic of COVID-19. Due to wireless nature of communication among the deployed drones and their ground station server, several attacks (for example, replay, man-in-the-middle, impersonation and privileged-insider attacks) can be easily mounted by malicious attackers. To protect such attacks, the deployment of effective authentication, access control and key management schemes are extremely important in the IoD environment. Furthermore, combining the blockchain mechanism with deployed authentication make it more robust against various types of attacks. To mitigate such issues, we propose a private-blockchain based framework for secure communication in an IoT-enabled drone-aided healthcare environment. The blockchain-based simulation of the proposed framework has been carried out to measure its impact on various performance parameters.

Efficient search in document image collections

Abstract: This paper presents an efficient indexing and retrieval scheme for searching in document image databases. In many non-European languages, optical character recognizers are not very accurate. Word spotting - word image matching - may instead be used to retrieve word images in response to a word image query. The approaches used for word spotting so far, dynamic time warping and/or nearest neighbor search, tend to be slow. Here indexing is done using locality sensitive hashing (LSH) - a technique which computes multiple hashes - using word image features computed at word level. Efficiency and scalability is achieved by content-sensitive hashing implemented through approximate nearest neighbor computation. We demonstrate that the technique achieves high precision and recall (in the 90% range), using a large image corpus consisting of seven Kalidasa's (a well known Indian poet of antiquity) books in the Telugu language. The accuracy is comparable to using dynamic time warping and nearest neighbor search while the speed is orders of magnitude better - 20000 word images can be searched in milliseconds.

New improved 1-bit full adder cells

Abstract: The 1-bit full adder is a very important component in the design of application specific integrated circuits. In this paper, authors propose three new 1-bit full adders having a delay of 2-transistor (2T) using existing XOR and XNOR gates. The power consumption, delay and area of these new full adders are compared with existing ones and the results appear to be promising. The combination of low power, low transistor count and lesser delay makes the new full adders a viable option for efficient design.

Early warning system using tilt sensors in Chibo, Kalimpong, Darjeeling Himalayas, India

Abstract: The Indian Himalayan regions have been significantly affected by the increase in the frequency of landslide occurrence. Thirty percentage of the worldwide landslide incidents occur in the Himalayan region with 42% of India’s landslide region belonging to Darjeeling–Sikkim Himalayas (NDMA Report, 2011). Several studies have been carried out worldwide on early warning systems considering rainfall history. Although rainfall is one of the criteria which can successfully predict the probability of landslides on a regional scale, it is indeed difficult to understand the risk associated with slope failure. The reason is the spatial variation of rainfall intensity, local soil conditions, geology, hydrology and topography. An early warning and monitoring system is one of the most effective techniques to minimise the disasters influenced by slope instability as it is less expensive and easier than slope reinforcement. A reliable and robust system fitted with microelectromechanical systems tilt sensor and volumetric water content sensors were installed. The sensor monitors the tilting angle of the instrument which was installed at shallow depths, and the variation of tilting angle corresponds to lateral displacement at slope surface. The primary objective was to monitor the tilting angles at various slope sites and to formulate a dependable warning system with a low probability of false alarms in Chibo Pashyor region in the Indian state of West Bengal. Such system would assist in the installation of several low-cost sensors over an active slope as it was difficult to determine the particular section of slope which could fail in the next heavy rainfall. The change in tilting rate would help in establishing thresholds determined empirically. Such system would help in developing an early warning system and also help in calibrating thresholds calculated using empirical techniques.