Irfan Siddavatam

Work place: K.J. Somaiya College of Engineering, Mumbai, India



Research Interests: Artificial Intelligence, Hardware Security, Network Security


Irfan Siddavatam received his PhD in Electronics Engineering from Veermata Jijabai Technological Institute, Mumbai in 2018. He is Associate Professor in Department of Information Technology at K.J.Somaiya College of Engineering, where he has been since 2001. His research interest includes Information and Cybersecurity, Artificial Intelligence, Machine Learning, Blockchain.

Author Articles
A Novel Approach for Video Inpainting Using Autoencoders

By Irfan Siddavatam Ashwini Dalvi Dipti Pawade Akshay Bhatt Jyeshtha Vartak Arnav Gupta

DOI:, Pub. Date: 8 Dec. 2021

Inpainting is a task undertaken to fill in damaged or missing parts of an image or video frame, with believable content. The aim of this operation is to realistically complete images or frames of videos for a variety of applications such as conservation and restoration of art, editing images and videos for aesthetic purposes, but might cause malpractices such as evidence tampering. From the image and video editing perspective, inpainting is used mainly in the context of generating content to fill the gaps left after removing a particular object from the image or the video. Video Inpainting, an extension of Image Inpainting, is a much more challenging task due to the constraint added by the time dimension. Several techniques do exist that achieve the task of removing an object from a given video, but they are still in a nascent stage. The major objective of this paper is to study the available approaches of inpainting and propose a solution to the limitations of existing inpainting techniques. After studying existing inpainting techniques, we realized that most of them make use of a ground truth frame to generate plausible results. A 'ground truth' frame is an image without the target object or in other words, an image that provides maximum information about the background, which is then used to fill spaces after object removal. In this paper, we propose an approach where there is no requirement of a 'ground truth' frame, provided that the video has enough contexts available about the background that is to be recreated. We would be using frames from the video in hand, to gather context for the background. As the position of the target object to be removed will vary from one frame to the next, each subsequent frame will reveal the region that was initially behind the object, and provide more information about the background as a whole. Later, we have also discussed the potential limitations of our approach and some workarounds for the same, while showing the direction for further research.

[...] Read more.
Industrial Control Systems Honeypot: A Formal Analysis of Conpot

By Sheetal Gokhale Ashwini Dalvi Irfan Siddavatam

DOI:, Pub. Date: 8 Dec. 2020

Technologies used in ICS and Smart Grid are overlapping. The most discussed attacks on ICSs are Stuxnet and Black energy malware. The anatomy of these attacks not only pointed out that the security of ICS is of prime concern but also demanded to execute a proactive approach in practicing ICS security. Honeypot is used to implement defensive measures for security. The Honeynet group released Honeypot for ICS labelled as Conpot in 2013. Though the Conpot is low interactive Honeypot, it emulates processes of different cyber-physical systems, typically Smart Grid. In the literature, the effectiveness of Honeypot operations was studied by challenging limitations of the existing setup or proposing new variants. Similar approaches are followed for Conpot evaluation. However, none of the work addressed a formal verification method to verify the engagement of Honeypot, and this makes the presented work unique. For proposed work, Coloured Petri Net (CPN) tool is used for formal verification of Conpot. The variants of Conpot are modelled, including initial state model, deadlock state model and livelock model. Further evaluation of these models based on state space analysis results confirmed that Conpot could lure an attacker by engaging him in an infinite loop and thereby limiting the scope of the attacker from exploring and damaging the real-time systems or services. However, in the deadlock state, the attacker’s activity in the conpot will be restricted and will be unable to proceed further as the conpot model incorporates deadlock loop.

[...] Read more.
Cuisine Detection Using the Convolutional Neural Network

By Dipti Pawade Ashwini Dalvi Irfan Siddavatam Myron Carvalho Prajwal Kotian Hima George

DOI:, Pub. Date: 8 Jun. 2020

In today’s fast world, everyone wants the information in one click. The same rule applies when you have some food items in front of you. In social events, few cuisines are known to us while some are not. Also, in a few cases, we know the cuisine name, but we are not aware of its nutritional value. This motivated us to develop a system that can identify the cuisine name from the image and gives the nutrient value for the same. Here Convolutional Neural Network (CNN) is used to predict the cuisine name present in an image and then further its nutritional value is calculated based on the information present in a database. User needs to click the image of the cuisine; the application will identify the cuisine name and its nutrition value for standard serving amount considering the cuisine is prepared using the standard recipe.

[...] Read more.
Multi Genre Music Classification and Conversion System

By Irfan Siddavatam Ashwini Dalvi Dipen Gupta Zaid Farooqui Mihir Chouhan

DOI:, Pub. Date: 8 Feb. 2020

Artificial Intelligence (AI) has a huge scope in automating, stream- lining, and increasing productivity of Music Industry. Here, we look upon AI based techniques for classifying a piece of music into multiple genres and then later converting it into another user-specified genre. Plenty of work has been done in classification, but using traditional machine learning models which are limited in term of accuracy and rely heavily on features to train the model. The novelty of this work lies in its attempt to covert genre of music from one type to another. This paper focuses on classification achieved by using a model trained via Convolutional Neural Networks. Conversion of music genre, a relatively less worked upon field has been discussed in this paper along with details of implementation. For Conversion, we initially convert the input file to spectrogram. A database of all genre is maintained at all times and a random file from user selected genre is also converted to spectrogram. Later, these spectrograms are processed and converted back to signals. Finally the user can listen to the converted audio file. Validation of the conversion was performed via a survey with the help of end users. Thus, a novel idea of doing Music Genre Conversion was put forth and was validated with positive outcomes.

[...] Read more.
Text Analyzer for Competitive Examination

By Ashwini Dalvi Irfan Siddavatam Sagar Ailani Smith Dedhia Shyamal Makwana

DOI:, Pub. Date: 8 Nov. 2019

Competitive examination provide a platform to the user for gauging their verbal and literature skills. The tools available currently only provide some simple feature regarding text processing such as spelling correction and providing different synonyms of the selected words. A complete assessment is not done for the user’s abilities and relevant details related to the context are not taken entirely into consideration. The following paper proposes a way to implement Natural Language Processing on text to provide feedback to the user for their competitive examinations. The assessment of the text will be done according to the parameter such as grammar, vocabulary; relevance to the context.
Some applications for web and mobile platform are available to offer assessment of English language essay but limited academic research available to validate research work in this domain. This work is effort to address requirement of text analyzer for English language evaluation methods incorporating natural language processing.

[...] Read more.
Other Articles