A Systematic Implementation of Deepfake Multimedia Video Generation and Detection using Deep Learning

Debnath Bhattacharyya,Eali Stephen Neal Joshua,N.Thirupathi Rao 한국컴퓨터게임학회 2021 한국컴퓨터게임학회논문지 Vol.34 No.2

In the following years, technology has progressed in so many ways that it has provided the cyber society with a resource that only computers can excel at, such as the art of counterfeit of media, which was before unavailable. Deepfakes are a term used to describe this kind of deception. The majority of well-documented Deep Fakes are produced using Generative Adversarial Network (GAN) Models, which are essentially two distinct Machine Learning Models that perform the roles of attack and defence. These models create and identify deepfakes until they reach a point where the morphing no longer detects the deepfakes anymore. Using this algorithm/model, it is possible to discover and create new media that has a similar demographic to the training set, resulting in the development of the ideal Deep Fake media. Because the alterations are carried out utilising advanced characteristics, they cannot be seen with the human eye. However, it is completely feasible to develop an algorithm that can automatically identify this kind of tampering carried out via the internet. This not only enables us to broaden the scope of our search beyond a single media item, but also beyond a large library of mixed media. The more it learns, the better it becomes as artificial intelligence takes over in full force with automation. In order to create better deep fakes, new models are being developed all the time, making it more difficult to distinguish between genuine and morphing material.

On extremal rough I-convergence limit point of triple sequence spaces defined by a metric function

N. Subramanian,A. Esi,S. Debnath 한국전산응용수학회 2020 Journal of applied mathematics & informatics Vol.38 No.3

We introduce and study some basic properties of rough I-convergent of triple sequence spaces and also study the set of all rough I-limits of a triple sequence spaces.

Latest Technologies used in Industries for Better Production: An Extensive Review

N.Thirupathi Rao,Debnath Bhattacharyya 사단법인 미래융합기술연구학회 2020 아시아태평양융합연구교류논문지 Vol.6 No.6

Technology had grown very fast in recent years in terms of all types of techniques used in various industries. Various types of trends can be observed in almost all sorts of technologies being used in various fields of the market. The utilization of new technology and its trends had become very interesting and making the functioning of the works smoother and more comfortable also reduced the utilization of more man power and reduced the problems faced with manpower. The utilization of such technology trends is giving proper market analysis and also better business processing applications. The trends of the market can be observed or understood within minutes by the utilization of such advanced technologies. In the current article, an attempt has been made to provide the details about the latest technologies and their trend of utilizing in industries for manufacturing and their future scope of development and their utilization in the current day applications.

Present Scenario and Future Prospects of Phytase in Aquafeed - Review -

Debnath, Dipesh,Sahu, N.P.,Pal, A.K.,Baruah, Kartik,Yengkokpam, Sona,Mukherjee, S.C. Asian Australasian Association of Animal Productio 2005 Animal Bioscience Vol.18 No.12

Aquaculture pollution is a major concern among the entrepreneurs, farmers and researchers. Excess discharge of phosphorus and nitrogen into the water bodies is the principal pollutant responsible for this. Plant-based feed ingredients due to its high phytic acid content enhances both nitrogen and phosphorus discharge thereby increasing the pollution level. Dietary phytase treatment is probably the best answer to address this problem. This review explains the nature and properties of phytate, its interactions with other nutrients and the application of phytase in aquafeed to reduce the pollution. This review also covers the different biotechnological aspects for lowering the phytic acid level in the common aquafeed ingredients, as an alternate approach to controlling the pollution level. Some of future research needs have also been highlighted to attract the attention of more researchers to this area.

Illumination Invariant Face Recognition Model using Tetrolet Transform and Truncated Gaussian Mixture Model

B. N. Jagadesh,Nazma Sultana Shaik,Venkata Naresh Mandhala,Debnath Bhattacharyya 보안공학연구지원센터 2016 International Journal of Software Engineering and Vol.10 No.10

Face Recognition is the most focused topic in the field of computer vision and pattern recognition. It has become a major thrust area for research in the last two decades due to the security aspects and demand for video. Face is focus of attention in the social intercourse and it plays a vital role in identification and recognition of individual emotions. Several facial recognition algorithms were developed and discussed in the literature but very little work is focused on facial recognition based on illumination. The appearance of the face image is usually affected by illumination conditions that hinder the facial recognition process. Hence in this paper, we propose and develop a new facial recognition algorithm based on Adaptive Haar Wavelet Transform called Tetrolet Transfrom. In Tetrolet transform, the determined orthonormal basis functions are adapted to geometrical features of the image follow a Truncated Gaussian Mixture Model (TGMM). The truncation on the feature vector has a significant influence in improving the recognition rate of the system.

ON EXTREMAL ROUGH I-CONVERGENCE LIMIT POINT OF TRIPLE SEQUENCE SPACES DEFINED BY A METRIC FUNCTION

SUBRAMANIAN, N.,ESI, A.,DEBNATH, S. The Korean Society for Computational and Applied M 2020 Journal of applied mathematics & informatics Vol.38 No.3

We introduce and study some basic properties of rough I-convergent of triple sequence spaces and also study the set of all rough I-limits of a triple sequence spaces.

Bayesian Probabilistic Approach for Model Updating and Damage Detection for a Large Truss Bridge

Samim Mustafa,N. Debnath,Anjan Dutta 한국강구조학회 2015 International Journal of Steel Structures Vol.15 No.2

A Bayesian probabilistic methodology is presented for structural model updating using incomplete measured modal data which also takes into account different types of errors such as modelling errors due to the approximation of actual complex structure, uncertainties introduced by variation in material and geometric properties, measurement errors due to the noises in the signal and the data processing. The present work uses Linear Optimization Problems (LOP) to compute the probability that continually updated the model parameters. A real life rail-cum-roadway long steel truss bridge (Saraighat bridge) is considered in the present study, where identified modal data are available from measured acceleration responses due to ambient vibration. The main contributions of this paper are: (1) the introduction of sufficient number of model parameters at the element property level in order to capture any variations in the sectional properties; (2) the development of an accurate baseline model by utilizing limited sensor data; (3) the implementation of a probabilistic damage detection approach that utilizes updated model parameters from the undamaged state and possibly damaged state of the structure.

Voice Playback Detection based on Long-window Scale-factors

U. Nanaji,N. Thirupathi Rao,Ch. Raj Kumar,Debnath Bhattacharyya,Hye-jin Kim 보안공학연구지원센터 2016 International Journal of Security and Its Applicat Vol.10 No.12

With the popularity of high fidelity and portable recording device, it becomes very easy for the attacker to spoof speaker verification system by voice playback. In this paper, it found that there exists obvious difference in the scale-factors, a parameter of MP3 codec, when the original and playback voices are compressed by MP3. So, a detection algorithm for the playback voice is proposed. The experimental results show that the detection accuracy of the proposed algorithm can reach to 99.51% for 4 different types of eavesdropping devices. Meanwhile, the algorithm is integrated into the speaker verification system based on GMM-UBM. The equal error rate (EER) of the system has dropped 32% and the system's ability to resist playback attack is improved.

Gelatinized Carbohydrates in the Diet of Catla catla Fingerlings: Effect of Levels and Sources on Nutrient Utilization, Body Composition and Tissue Enzyme Activities

Yengkokpam, Sona,Sahu, N.P.,Pal, A.K.,Mukherjee, S.C.,Debnath, Dipesh Asian Australasian Association of Animal Productio 2007 Animal Bioscience Vol.20 No.1

A Feeding trial was conducted to study the effects of three different sources and two levels of dietary gelatinized carbohydrate (GC) on nutrient utilization, growth, tissue composition and tissue enzyme activities of fingerlings of Catla catla (15.1-15.3 g). Six isocaloric (17.1-17.5 kJ/g) semi-purified diets were prepared either with rice, corn or tapioca at 40 or 50% GC each. The crude protein (CP) level used in the diet was 35% and 25% for 40% and 50% GC level, respectively to study the protein sparing effect of GC. The degree of gelatinization was higher for corn and tapioca than rice under similar cooking conditions. After a 60-d feeding trial, dry matter, carbohydrate, protein and lipid digestibility were higher in tapioca fed groups at both the levels of GC. However, the highest specific growth rate (SGR) and protein efficiency ratio (PER) were observed in the corn fed groups at 50% GC level indicating better utilization of nutrients from gelatinized corn. Feed conversion ratio (FCR) was almost similar in corn and tapioca fed groups between two levels of GC but in rice fed groups, FCR was lower in 40% GC than 50% GC level. The results indicated higher protein-sparing effect in corn and tapioca fed groups than rice fed groups. The order of gelatinized carbohydrate utilization in Catla catla fingerlings at 50% GC level was corn>tapioca>rice. At 40% GC level, corn and tapioca were comparable and more efficiently utilized than rice. In the corn fed groups, 50% GC was comparable with 40% GC level, whereas in rice and tapioca fed groups the 40% GC was better in terms of nutrient utilization. Liver glycogen content and hepatosomatic index were significantly (p<0.05) higher in those groups fed high GC (50%) irrespective of carbohydrate sources. Higher intestinal amylase and glucose-6-phosphate dehydrogenase activities were observed in higher GC fed groups than the lower GC groups. No mortality was found in any groups at any levels of GC.

Ciphertext-Policy Attribute-Based Encryption for Access Control of Data in Cloud

Vijaya Lakshmi Paruchuri,N Lakshmipathi Anantha,Vara Lakshmi Konagala,Debnath Bhattacharyya 보안공학연구지원센터 2016 International Journal of Software Engineering and Vol.10 No.8

In Distributed systems, the users with a certain set of attributes can only be able to access the data. At present this process can be done through a trusted server where we will store the data and there will be certain constraints on the access of the data. In this case there will be a possibility to compromise the data and so the confidentiality of the data is lost. An Attribute-Based Encryption (ABE) is an encryption scheme, where users with some attributes can decrypt ciphertexts associated with these attributes. Now this is our turn to develop a system with a more complex policy of access of the encrypted data and which can be called as Ciphertext Policy Attribute-based Encryption (CP-ABE). By using this method the information can't be traded off even through the trusted server where the information is put away. These methods are also secure against the collusion attack. In this method attributes are generally assigned in the form of access trees. The attributes are placed at the leaf nodes of this access tree. In older Attribute-based encryption strategies encrypted data is described by the attributes and policies are given to the user’s keys, while in our system users credentials are described by the attributes and there will be a policy where it tells us about who should access or decrypt the data. So, this type of access method is very much closer to the Role-based attribute-based encryption.