Rajveer S Shekhawat

IT security has always been a major concern for all organizations, especially after the rise in IT Integration amongst all processes, Post pandemic this has become a bigger issue than earlier. The organizations are growing also with IT Integration the negative impact of information related risk incidents are also increasing worldwide. There are several IT Risk Assessment Frameworks in use to address information security assaults, vulnerabilities, threats, and breaches, including ISO 270001/27005. COBIT, NIST SP- 800/53 etc, though following and implementation of these protocols, still organizations face challenges of IT risk, which may involve an asset or information. The biggest challenge is the data Security or privacy. Based on survey data, this study evaluates the majority of the current IT risk management frameworks and makes an effort to pinpoint any shortcomings in them. Based on the analysis done, a new IT Security framework is proposed and implemented in two organizations f...

Background: Olive (Oleo europaea L.) cultivars are widely cultivated all over the world. However, they are often attacked by pests and pathogens. This deteriorates the quality of the crop, leading to less yield of olive oil. The different infections that cause comparable disease symptoms on olive leaves can be classified using image processing techniques. Objective: The olive has established itself as a superfood and a possible source of medicine, owing to the rapid increase in the availability of data in the field of nutrigenomics. The goal of this review is to underline the importance of applying image processing techniques to detect and classify diseases early. Method: PubMed, ScienceDirect, and Google Scholar were used to conduct a systematic literature search using the keywords olive oil, pest and pathogen of olives, and metabolic profiling. Results: Infections caused by infectious diseases frequently result in significant losses and lowquality olive oil yields. Early detection...

These days’ security plays the greatest role in order to provide safe platforms and surveillance becomes the essential need to provide accurate results in case of security breach. Gait recognition is a biometric technique that does not need human intervention. Through this technique, human can be uniquely identified. Gait is defined as the way human walks (human locomotion) and this can be used as biometric identity because the manner in which every person walks can uniquely categorize person. But there are many challenges like variation in viewpoints, clothing variations, carrying conditions and so on. A novel approach using deep learning is proposed to address this challenge. To address these limits, we give an idea of having multi-view gait-based recognition system, to provide robust system that is capable of handling one camera and subject walking on different angles from 0° to 180° of view. To achieve the results 3D CNN-based model is used in order to obtain spatio-temporal fea...

Modern campuses have been built with many modern facilities and aids for comfort and to enhance teaching quality. Some of these include smart electronic boards, projectors, desktops, air conditioning apart from lights, and fans. Most campuses have large number of classroom and student population with good number of courses. This demands very complex classroom allocation prices for their optimal utilization. In spite of best effects, continuous use of classroom is difficult. This causes same classroom going vacant regularly and many times, the equipment and light/fans are left ON, causing heavy consumption of energy. A system to eliminate wastage of energy, a smart classroom energy management system (CEMS) has been considered and pilot of the same has been tested. Based on the survey of a typical modern campus academic building, the proposed system can give 10–30% of energy saved which in turn provides a savings of lakhs of rupees per month. The system can be easily adopted to other ...

Plant diseases occur in various parts of plants and have a variety of identifying symptoms; the majority of them can be visually identified and evaluated. Arnal Barbedo [1] has thoroughly reviewed a variety of techniques used for detection, quantification, and classification of plant diseases using image processing techniques. We have explored the plant diseases which are visually dominant and can be observed at the earlier stage of its life cycle. In this work, we have detected and quantified spot diseases which affected several plants. We have used the images from the PlantVillage dataset and collected five different spot disease images. We have isolated the region of interest (ROI) in the infected leaf using three different methodologies. Firstly, we tried to apply thresholding technique on the histogram values of the leaf image in the La*b* color model; we also have used the k-means-based color segmentation on the RGB color model; we have also tested Delta-E color segmentation. ...

Road crashes and resulting fatalities and injuries have evolved as a major issue across the globe. According to the Global Status Report on Road Safety 2018 [1] published by World Health Organization, the burden of road traffic injuries and deaths is borne by vulnerable road users living in low-and middle-income countries where deaths are increasing due to abrupt growth in number of motorized transport. It is estimated that road crashes alone are responsible for 3 to 4 percent of GDP loss in India. Considering the severity of the issue it is important to identify underlying factors of road crashes to reduce the damage caused to human lives and national asset. By applying various Machine Learning algorithms on road crash data, workable models can be built which can predict outcome based on past trends of road crashes.

ABSTRACT Plant disease classification using image processing techniques is a prominent and challenging area of research. We have developed a novel classification technique to classify, especially spot and blight diseased leaf images of four different plant species. In this technique, we have dealt with the infection patterns manifested on leaves. The infection patterns seem to correlate with diseases. Both these diseases cause similar patterns on leaves, and hence they are hard to distinguish. The proposed technique succeeded in handling the task to a reasonable extent. Statistical texture features derived from Grey-Level-Co-occurrence-Matrix (GLCM) are considered as features. The final feature set contains strongly correlated features. An impact level of each feature is derived from its standard deviation for the image set. The novel classification technique makes use of these impact levels. A 74% disease classification accuracy is achieved in the best-case scenario and identified an optimal threshold range that helps us classify the diseases.

The design of ASICs (Application Specific Integrated Circuits) is based on three main approaches; (i) GATE ARRAYS ICs, (ii) CELL-BASED ICs, (iii) PLD and more recently EPLD based ICs. Gate arrays contain matrices of transistors fabricated on silicon base wafers. They are customised by the placement of vendor supplied and customer specified macrocells and interconnection patterns. Cell Based ICs are single chips designed from predefined cells consisting of a few gates on upto microprocessor cores and memory arrays containing tens of thousands of gates. They require eight to twelve weeks for designing a prototype as opposed to a few days or weeks for gate arrays or a few hours for PLD's. Cell Based ICs are the premier ASICs where high integration and high performance are required. EPLD use EPROM-MOS technology and can be reprogrammed over and over again. They require less power to operate and are easier to test than standard PLD's.Erasable Progammable Logic Devices (EPLDs) currently offer the ability to com...

There is intense pressure on agricultural productivity due to the ever-growing population. Several diseases affect crop yield and thus, effective control of these can significantly improve the production of food for all. In this regard, detection of diseases at an early stage and quantification of the severity, in general, has acquired urgent attention of the researchers. In this study, a summary of prevalent techniques and methodologies used for the detection, quantification and classification of diseases is presented to understand the scope of improvement. The study pays attention to critical gaps that exist in available approaches and enhance them for the early prediction of diseases. Diseases affect almost all parts of plants, e.g. root, stem, flower, leaf; a manifestation in different ways for different parts of the plant of the same disease presents a challenge for researchers. This study extends the review work published by JGA Barbedo in 2013, as there have been significant advances and numerous new techniques introduced since then. A novel approach of classifying and categorisation of the existing techniques based on pathogen types is a significant contribution by the authors in this study.

The yearly increase in the population of India was 68.33 million in 1981 and grew to 121.01 million in 2011, respectively. It is estimated that by the year 2028 India will hold the largest population of the world. The prompt upsurge of the Indian population will force people to migrate from the rural areas to the mega cities. This enormous migration will increase the demand for more space to live in mega cities and will lead to a situation of urban sprawl. The key challenges are to achieve sustainable development and to predict the future urban sprawl. In this paper, we have proposed a novel model to predict the future urban sprawl. We have used an integrated approach of remote sensing, GIS, and MLP to propose a novel model to predict the future urban sprawl for the city Jaipur up to the year 2051.

ABSTRACT A major source of existing knowledge viz. existing mathematical models has never been used directly for the purpose of fuzzy modelling. The standard way mathematical models are utilised for fuzzy models is through generating i/o data from these using appropriate probability distribution followed by standard identification methods to elicit the fuzzy model. The novel approach presented in this paper based on the extension principle of fuzzy set theory is much simpler compared to the i/o based model identification. The approach being presented involves extending the MISO mathematical models into fuzzy space to arrive at a fuzzy system with a large number of output fuzzy sets, partition these undesirably large number of fuzzy sets into a more manageable smaller number of subjective fuzzy sets resulting in a fuzzy relational model. The paper also shows how to provide an interpretation of the system knowledge in more human-like language viz. fuzzy rules. One could stop at the first stage of the work limited to fuzzy extension only and still be able to provide the linguistic interpretation of the system but that would be restricted to only SISO systems

Plant diseases occur in various parts of plants and have a variety of identifying symptoms; the majority of them can be visually identified and evaluated. Barbedo [1] have thoroughly reviewed a variety of techniques used for detection, quantification and classification of plant diseases using image processing techniques. We have explored the plant diseases which are visually dominant and can be observed at the earlier stage of its life cycle. In this work, we have detected and quantified Neofabraea leaf spot in olive plants. The data has been collected from local olive farms and through online resources. We have isolated the Region of Interest(ROI) in the infected leaf using two different methodologies. Firstly we tried to apply thresholding technique on the Histogram values of the leaf image in the La*b* color model; we also have used the k-means based color segmentation on the RGB color model. Quantification of the disease is also performed using the ratio of the infected region b...

The unorganised, unplanned, uncontrolled, and unauthorised development has been termed as urban sprawl and often referring to as a complex pattern of transportation, social, economic, and land use development. The rapid increase in population and economic development are the primary influences of urban sprawl. Urban sprawl is the major problem of metropolitan areas all over the world especially developing countries. The impact of urban sprawl can be noted as pollution, traffic, loss of prime agricultural land, deforestation, congestion of places, and water pollution. In this paper, we have proposed a fuzzy-CA-Markov model for modelling urban sprawl. The factors which influence the urban sprawl is considered as fuzzy parameter like accessibility from local road, accessibility from main road, accessibility from major road, slop, altitude, and density. The future prediction of urban sprawl was measured through Markov's cellular automata model.

Many IoT applications demand devices operating on battery, which restricts the useful life of the developed solution as most scenarios do not allow replacement of batteries. Given a target period of survivability of IoT devices, we must design devices and sub-systems meeting the estimated energy consumption rate so as meet target lifetime. Thus, for a given scenario, to meet demanding solutions offering optimal energy utilization, we propose an approach addressing energy consumption minimization right from design/development stage through deployment and finally concluding at operational level. The key parameters of design and operation relate to data acquisition, communication and security sub-systems, where security may not always be required. The approach is encouraged by the fact that about 50–60% energy used by networked devices is spent by communication system and about 30% on security/encryption mechanisms. Thus, any attempt at optimal use of energy can lead to significant inc...

Energy efficiency is an important issue in wireless sensor networks; clustering of nodes and sensor data aggregation are popular techniques to address the issue. Sensors generate sensitive data in many applications and thus methods to secure the data so as to prevent easy access by unauthorized agents are essential. Security against false data injection or data tampering can be provided through encryption of sensor data; but it increases load on processing. Securing data, however, puts a burden on the node battery. Data aggregation aims both at data reduction so as to spend less time for encryption as well as reduce transmission load, but demanding additional processing at aggregating node. Thus, aggregation and security are contradictory solutions. The dilemma can be resolved by allowing targeted malleability of encrypted data through homomorphic encryption. The recent research works recommend schemes which use homomorphic MACs, signatures and a blend of private and public cryptogr...

There is intense pressure on agricultural productivity due to the ever-growing population. Several diseases affect crop yield and thus, effective control of these can significantly improve the production of food for all. In this regard, detection of diseases at an early stage and quantification of the severity, in general, has acquired urgent attention of the researchers. In this study, a summary of prevalent techniques and methodologies used for the detection, quantification and classification of diseases is presented to understand the scope of improvement. The study pays attention to critical gaps that exist in available approaches and enhance them for the early prediction of diseases. Diseases affect almost all parts of plants, e.g. root, stem, flower, leaf; a manifestation in different ways for different parts of the plant of the same disease presents a challenge for researchers. This study extends the review work published by JGA Barbedo in 2013, as there have been significant ...

The yearly increase in the population of India was 68.33 million in 1981 and grew to 121.01 million in 2011, respectively. It is estimated that by the year 2028 India will hold the largest population of the world. The prompt upsurge of the Indian population will force people to migrate from the rural areas to the mega cities. This enormous migration will increase the demand for more space to live in mega cities and will lead to a situation of urban sprawl. The key challenges are to achieve sustainable development and to predict the future urban sprawl. In this paper, we have proposed a novel model to predict the future urban sprawl. We have used an integrated approach of remote sensing, GIS, and MLP to propose a novel model to predict the future urban sprawl for the city Jaipur up to the year 2051.

The world's urban population has increased 2% to over 50% in the last two centuries. It is estimated that it will reach over 75% by 2030. Due to increase in urban population, developing countries like India are facing more problems in mega cities like uncontrolled, unauthorised, uncoordinated, and unplanned urban growth; which is often termed as urban sprawl. Urban sprawl is generally spread between urban and rural areas, which we cannot consider either in urban nor in rural often known as semi-urban areas. The semi-urban areas are in the phase of new development from rural to urban hence it attracts more urban sprawl. In this regard, we have proposed, a fuzzy pixel classification model for identification of potential areas of urban sprawl using fuzzy logic, GIS and remote sensing. This study will help urban planner, town planner, and government to identify potential areas of urban sprawl.

Fieldbus promises to revolutionise the industrial automation. The hierarchical system of networks of ‘factories of the future’ connecting different system units ( in several hierarchical levels) will include a network at the field level, known as Fieldbus, as well. Conventional systems use 4-20 mA signaling standard for transmission of measured process variables from factory floor to control room, with inherent disadvantages of analog transmission.The fieldbus, on the other hand, is an all digital, low cost and high performance multi drop communications protocol, enabling the remote access to the field devices such as sensors, actuators, I/O racks and local controllers. PROFIBUS (PROcess FIeld BUS) is one of the most promising among the fieldbuses and is the German national standard. It specifies necessary functions to allow data communication among fieldbus devices. Smart transmitters are the back bone of modern plants. They incorporate various features such as sensor linearisation...

Fuzzy set theory provides a formal system for modelling complex systems for which only an imprecise or approximate specification is available. A system model captures the relationship between the variables specifying the input and output domains. In classical modelling, the relationships are expressed mathematically as a function whose domain consists of the possible inputs to the system and whose range comprises the appropriate responses. As the systems being modelled become more sophisticated, it becomes increasingly difficult to construct mathematical models directly from our knowledge of the system. This is due both to the complexity of interactions within the system and incomplete knowledge of the system parameters.

The inability to construct mathematical models provided the impetus for the development of alternative approaches for system modelling. The popularity of the fuzzy models is attributable to its ability to linguistically specify relationships that are too complex or ill-understood to be directly describable by precise mathematical models. Fuzzy models have been successfully employed as expert systems in database systems, decision analysis, and process control systems etc.

Historically, fuzzy rule bases were constructed from human expertise and heuristic knowledge of the systems being modelled. Recently, learning techniques have been developed to construct rules from system observations used as training data. A training set is obtained from the actually measured data of excitations and responses pertaining to an actual operational system. In either method of constructing a rule base, it is possible to encounter input conditions for which no rules have been produced. When rules are obtained by knowledge acquisition from domain experts, there may be sets of conditions that have never been experienced or anticipated by the experts in dealing with the system under study. Similarly, when rules are learnt from training data, it is possible that infrequently occurring system configurations may not be represented in the training set. A lot of work has thus to be done to arrive at reliable and complete rule-base of practical utility. Reports of the works done by the various researchers are available in literature regarding the reliability and completeness of the rule sets, using the knowledge embedded in the existing rule set itself [Sudkamp96].

This thesis presents novel methods to develop fuzzy relational models of complex non-linear Multi-Input Multi-Output (MIMO) systems. The three most common knowledge sources for systems for model building are operator experience, input-output observations, and mathematical models built from basic principles.  Two distinct approaches have been tried in the present work.    The first approach is based on input-output data  which is either directly available from actual measurements on real systems or generated from approximate mathematical models of real systems arrived at using traditional identification/modelling methods.  The novelty of the present work as compared to work reported so far in the literature lies in the identification of the fuzzy relational model by suitably training a neural network on the basis of fuzzified input-output data, followed by the extraction of the fuzzy relations. There is however a basic problem with this approach, where one has to build a fuzzy model from scratch every time.  The second approach is a completely new look at the problem of fuzzy modelling of systems in which existing system knowledge embodied in the mathematical model is utilised. Here fuzzy relations are  generated for describing a system by direct fuzzification of an approximate mathematical model of the system. Both of these approaches to the best of our knowledge represent completely new methodologies.

In the first approach, based on i/o observations of plants, it has been demonstrated how one can make use of standard back propagation neural networks to extract the fuzzy relational models avoiding use of specialised fuzzy neural network (FNN) models and complex learning algorithms which are usually employed for this purpose. These specialised FNN models and their learning approaches lack sound proofs of convergence and stability for general cases. One disadvantage of standard NN’s viz. difficulty of extracting the embedded knowledge after training has however been removed in the present work and it is shown that the fuzzy relational model can be very easily extracted after the learning phase. It has also been shown that one can either use neural recall itself to predict the output of fuzzy relational model or use the fuzzy inference process with a choice of suitable composition operators for this purpose.  This also establishes equivalence between the two prediction methods using the fuzzy relational models.

The second approach is based on existing mathematical description of systems. It is usually the case that some approximate mathematical models of systems are easily built based on some gross simplification of complex systems. While building fuzzy models of these complex systems, this initial knowledge can be very useful to derive start-up fuzzy models which can then be further refined if need be. This approach leads to a much faster generation of a near-optimal model as compared to traditional methods based on input-output data only. The method investigated here, to make use of the mathematical description for such fuzzy modelling, is based on the concept of extension principle, which has been used in the past to fuzzify mathematical operators and structures. The key steps involved are: fuzzification of the mathematical description using the extension principle, partitioning of a large number of extended fuzzy sets of output into a smaller and manageable number of subjective fuzzy sets, and optimisation of the entire process.  A number of single-output systems, both linear and non-linear; single-input as well as multi-input, both static and dynamic types have  been used to demonstrate the effectiveness of the novel approach. 

The deployment of these novel techniques widens the area of fuzzy modelling methods already available to the practising engineer and allows him to tap additional sources of system knowledge and deployment of simpler and well-understood standard neural networks for the modelling purpose.