Prof. Ram Meghe Institute of Technology & Research
Recent publications
This article discusses machine learning’s function in digital forensics to better understand where machine learning stands in today’s cybersecurity arena when it comes to gathering digital evidence. Hear began with discussing the history and development of digital forensics. Following that, the work proposes a short literature study to demonstrate the areas of digital forensics where machine learning techniques have been applied to date. The purpose of this article is to raise awareness about the use of machine learning in digital forensics. I am attempting to examine several machine learning applications in various fields to see how they might use machine learning in other sectors based on their current uses. The concepts described here will pave the way for developing more sophisticated and effective digital forensics tools.
Sn2+ doped white emitting alkaline-earth chalcogenide CaS and SrS phosphors were synthesized by carbo-thermal reduction method. In this method special requirement such as H2S gas flow is not required as a source of sulphur and is comparatively easy method to prepare sulfides. The crystalline phase, morphology, and photoluminescence properties were characterized by x-ray diffraction (XRD), scanning electron microscope (SEM), and fluorescence spectrophotometer, respectively. Phosphors exhibit broad band excitation which has excellent spread over nUV as well as blue region of visible light i.e. 440–480 nm. Emission is in the form of characteristic broad band of Sn2+ covering nUV and almost entire regions of visible spectrum i.e. 400–650 nm. The XRD pattern of prepared phosphor well matches with International Center for Diffraction Data. Synthesized phosphor particles are of different sizes, with smooth surfaces, from less than 1 micron to few microns. The chromaticity coordinates of synthesized phosphors have been calculated from its corresponding emission spectra monitored at their excitation wavelengths. They observed to falls in white region of CIE diagram. These points are close to standard white point D65, corresponding to daylight with correlated color temperature (CCT) 6500 °K, indicating better color purity of the synthesized phosphor and they are promising material for a color converter using blue LED as the primary light (pumping) source in phosphor converted white LED (pc wLED) for solid state lighting.
Grading and classification of fruits is based on observations and through experiences. The system exerts image- processing techniques for classification and grading the quality of fruits. Two-dimensional fruit images are classified on shape and color-based analysis methods. However, different fruit images have different or same color and shape values. Hence, using color or shape analysis methods are still not that much effective enough to identify and distinguish fruits images. Therefore, computer vision and image processing techniques have been found increasingly useful in the food industry, especially for applications in quality detection. Research in this area indicates the feasibility of using computer vision systems to improve product quality, the use of computer vision for the inspection of food has increased during recent years. This proposed work presents food quality detection system. The system design considers some feature that includes fruit colors and size, which increases accuracy for detection of roots pixels. Histogram of oriented gradients is used for background removal, for color classification, support vector machine is used.
Toxic heavy metals and metalloids, like lead, mercury, arsenic, and selenium, are perpetually free into the surroundings atmosphere. There is a vital need to develop low-priced, effective, and supportable technique for removal or detoxification. Plant primarily based approaches, like phytore mediation, are unit comparatively cheap since they are performed in place and are solar-driven. Now this review, Specific advances in plant-based approaches for the remediation of contaminated water and soil. Phytoremediation is an alternate technology to remove of heavy metals in polluted soil. Wild plants were chosen for arsenic removal experiment. Removal of arsenic by conventional method is very costly; this paper focuses the review on method of phyto remediation to remove arsenic from soil. This method is being aesthetically pleasing and is on average tenfold cheaper than other physical, chemical or thermal remediation methods. This paper attempted to provide a brief review on recent progresses in research and practical applications of phytoremediation for soil and water resources.
The present work investigated the optimization of the production method of biodiesel from waste cooking oil (WCO) using hydrodynamic cavitation process. To achieve the maximum WCO biodiesel yield with a minimum number of experimental trials, one of the most popular optimization technique L9 Taguchi has been used for the validation of hydrodynamic cavitation (HC) process. In this HC experimental process, 260 mL of methanol were mixed with 0.5% of wt. KOH into a 500 mL conical flask. The mixture was later added to 1000 mL of warmed WCO, preheated to 35 °C in a storage tank of 30 L capacity. The HC reactor then ran for 30 min. The products were allowed to settle for 30 min and then the final sample of biodiesel were collected. The same process was repeated for all nine readings. The controllable features that were selected for the investigation, were the catalyst concentration (5 ≤ B ≤ 20 g), reaction temperature (35 ≤ C ≤ 55 °C), reaction time (30 ≤ D ≤ 50 min) and molar ratio (1:6 ≤ A ≤ 1:12). After the optimized setting, it was observed that at the molar ratio of 12:1, and other parameters such as catalyst concentration (5 g), reaction temperature (55 °C), and reaction time (40 min), 97% WCO biodiesel yield was observed. During the investigation, it was found that the reaction temperature has significant contribution on the WCO biodiesel yield. The value of coefficient of determination was found to be unity which shows the efficacy of the present investigation. The investigation has provided low cost feedstock for the WCO biodiesel production with maximum yield.
VRML (Virtual Reality Modeling Language) is a nonproprietary language used for scientific simulation and visualization. VRML alone do not have the capability of holding numerical values but with the help of slider design, the numeric values can be dynamically varied. The VRML Graphic User Interface (GUI) presented in this paper demonstrates dynamic simulation of Seed Grader Machine Parameters and its effect on final output. This paper describes how VRML capabilities can be utilized for simulating Seed Grader Machine Parameters in virtual Environment. The concept can be applied for variety of applications involving similar requirements.
The catchment boundary of Indla Ghatkhed watershed covers an area about 14..62 sq km. The erosion is a natural geomorphic process occurring continually over the earth’s surface and it largely depends on topography, vegetation, soil and climatic variables and, therefore, exhibits pronounced spatial variability due to catchments heterogeneity and climatic variation. This problem can be circumvented by discrediting the catchments into approximately homogeneous sub-areas using Geographic Information System (GIS). Soil erosion assessment modeling was carried out based on the Revised Universal Soil Loss Equation (RUSLE). A set of factors are involved in RUSLE equation are A = Average annual soil loss (mt/ha/year), R = Rainfall erosivity factor (mt/ha/year), k = Soil erodibility factor, LS = Slope length factor, C = Crop cover management factor, P = Supporting conservation practice factor. These factors extracted from different surface features by analysis and brought in to raster format. The output depicts the amount of sediment rate from a particular grid in spatial domain and the pixel value of the outlet grid indicates the sediment yield at the outlet of the watershed.
The software defined networking (SDN) is a networking technique to disintegrate control plane and data (forwarding) plane in the network, allows use of open protocols to control network switches and routers using software controls and abstract infrastructure as per application and network services requirements. This makes it possible to implement network management and control through software. Word software means that network devices are programmable but software is not controlling everything centrally. This approach helps to achieve network and infrastructure manageability, control plane modularity, cost-effectiveness, generalized data planes and adaptable network, making virtualization easy to implement, as per need of an hour. The SDN yields a network infrastructure which will be programmable. This inherently qualifies for control to be shifted to software and network flow with network devices to be operated by software programs. A network engineer must now be able to re-program instead of re-build the network infrastructure manually. The following sections explain terminologies, working, and operations related to SDN. This paper helps readers to gain understanding and overview of SDN.
Natural disasters can be very destructive to mankind so predicting it can be a very powerful tool since it supports the modification of the loss of damage to mankind and nature. Predicting events required continues processing of large amount of data with respect to time. With the help of good quality datasets, deep learning (DL) mechanism can become capable of predicting the existence of several natural disasters. It can be the difference between life and death for thousands of people. The concept of DL is not new. This paper discusses basic introduction of DL along with its role in disaster prediction system. It also encompasses different ideas to develop the system which may helpful to give any future direction to new researchers to start their research.
This paper is devoted to some new applications of Laplace-Weierstrass transform (i.e., for solving two dimensional diffusion equations). Solution of Cauchy’s linear differential equation is also given. Some results are also given which are required for solving Cauchy’s linear differential equation.
WSN is used in various applications such as hospital, environment monitoring, and experiments. It is also used in (military) [1, 2] battlefield for target tracking. If it is deployed in hostile environment in military application, then nodes can be captured and data can be updated. Wireless sensor network is equipped with low battery [3], storage, and computational power [4]. Hence, the main problem is the security of resource-constrained WSN. Various schemes are available to address the issues of WSN security. Proposed schemes are based on public and private encryption technique. Few schemes are based on biometric concept. While designing the security scheme researchers need to consider resource-constraint nature of wireless sensor network. Schemes are available but all are not applicable to WSN. This paper discusses various aspects of WSN to understand how schemes can be designed for WSN. Paper starts with application of WSN, summarizes security threats and analyzes various key management and security schemes.
Numerous techniques have been evolved for the detection of violence in human beings. Prior detection of human action can help to prevent and control suspicious and criminal activities. The offline video processing system has been used for post-action analysis. We address the violence detection trouble of humans in real-time visual surveillance such as punching, fighting. The present research work proposes a novel framework that processes real-time video data received from fixed cameras installed area of interest under surveillance. To determine the security level, we developed a new algorithm based on the decision-making classifier to recognize the violent situation in real time. In the view of human violence detection, the proposed work is simple and unique. The transition effects observed during violence detection are deliberated in detail. It has wide applications in the area of visual indexing, biometrics, telehealth, and human–computer interaction.
It is found that the malaria disease is a prime and major cause to the human health. The pernicious trappings of malaria stooge to the physical body cannot be making light of it. In this research work, a fuzzy-based expert system for the total handling of malaria disease had granted for providing judgment support platform to the specialist and healthcare researchers in the same endemic province. The proposed and implemented system consists of major components which include the cognitive content, the fuzzification, the inference engine and de-fuzzification for decision making. The fuzzy inference engine developed during this work is that the root sum square. This method is the depiction of inference that was designed and developed to infer the info from the fuzzy-based rules used in this algorithm. Triangular fuzzy membership function was accustomed that shows the degree of attendance of every input specification, and therefore, the de-fuzzification technique employed during this research is that the centre of gravity. This fuzzy-based expert system had been developed to help and to support clinical perception, diagnosis and therefore the expert’s proficiency. For validation and empharical analysis, the data of thirty patients with malaria defection was used. The results that were calculated are within the range of that was predefined and predicted by the territory proficient.
Indian farmers are behind as compared to other countries just not because of economic condition, but it has many reasons like they are lacking in the latest technologies, unaware about soil analysis, plant diseases, water table, quality of seeds and most important is a traditional way of farming. Indian farmers are not aware of modern way of farming. Various machine learning techniques are developed to improve farming techniques. The farmers can improve fruits quality and crop production with the help of machine learning. In this paper, we review agriculture problems that solved by using machine learning and also provide common steps that used to identify the objects from image dataset. In a nutshell, smart farming is the need of today’s farmer.
The Proxy Mobile IPv6 (PMIPv6) is a protocol which manages mobility. It uses signaling and home agent’s activity of MIPv6 through a proxy mobility agent in a localized network. In any type of wireless communication, handover delay should be as less as possible. Tremendous work has been done to minimize the handover delay in PMIPv6, and many solutions have been proposed by the many researchers. Almost all of the solutions have one thing in common that the authentication information of the MN should be sent to new MAG in advance, and to send information to new MAG, it is mandatory to anticipate the new MAG. This paper proposes an algorithm to predict new MAG and also provides the results achieved by simulating the proposed algorithm on NS-2.29.
Today with the enormously increasing and the advancing technologies the need of the automation has been reached to a great extent and will further increase in the future. Therefore, in this view our research work deals with the design, testing and implementation of cost-effective CNC based PCB Milling Machine which can be deployed in small scale industries and most targeted areas like academicinstitutions(engineering) toencourage self-designing of electronic circuits among students. Technically the control mechanism is sole work of AT Mega 328P Arduino microcontroller with auxiliary drivers which help in precise step control of stepper motors whose rotational movement appears as a controlled linear motion of x, y& z-axis each deployed with this motor. The input signal in form of design, picture is interpreted in form of G-codes through dedicated firmware which are installed in the Arduino Programming IDE to help generate the actuating signals proportional to the interpolating situations. This paper discusses about the additional feature of human safety system using IR sensors which are employed around the periphery of the working area.
Considering the network security aspect, one of the best way of preventing network infrastructure against anomalous activities is to monitor its traffic for suspicious activities. The reliable resource to accomplish this task is past network flow data, which can be analyzed to detect congestions, attacks or anomalies to ensure effective QoS of network infrastructure. Network traffic prediction involves analysis of past network flow data by capturing-storing data, preprocessing data, analyzing it based on various parameters & forming behavior patterns for various nodes in network. Once the patterns are observed for different nodes in network, their future communication can be predicted. Upon prediction of anomalous behavior, the preventive action will be initiated without wasting much of a time. Thus reducing the MTTR (mean time to respond) is the outline of our paper. The importance of network traffic data, traffic prediction methods and literatures available on topic are studied in this paper.
Friedmann–Robertson–Walker (FRW) space–time with bulk viscosity in the context of f(R) gravity is considered. The field equations are solved for the Power and Exponential volumetric expansion. Two types of functional relationship i.e. f(R) = R + bRm and f(R)=R-λ4R\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$ \,f(R) = R - \frac{{\lambda^{4} }}{R} $$\end{document} are investigated. The Phantom, Chaplygin gas and Tachyon fields are discussed. It is observed that the universe is open and inflationary.
We have investigated the spatially homogeneous and isotropic Friedmann–Robertson–Walker (FRW) universe filled with barotropic fluid and dark energy in the framework of the Brans–Dicke theory of gravitation. Here we have discussed three models: (i) law of variation for Hubble’s parameter, which leads to a constant value of deceleration parameter, (ii) hybrid expansion law model, and (iii) special form of deceleration parameter model. We have found that among all these derived models, the most suitable standard cosmological model according to the recent cosmological observations is the model with special form of deceleration parameter.
Sewing machine operators suffer from musculoskeletal problems imposed due to constrained restricted body postures. This study was conducted to investigate the effects of three design parameters (fore/aft sewing distance, sewing desk inclination and sewing desk height) of sewing workstation on postural variables and subjective experience and to develop guidelines for sewing workstation design. At a prototype of adjustable sewing workstation, ten professional sewing machine operators performed sewing task in nine different workstation arrangements. Sewing machine operators working posture and perceptions were recorded. It was shown that trunk, neck and arm postures were influenced by fore/aft sewing distance, sewing desk inclination and sewing desk height. The determinant factor for sewing machine operators’ perception on the trunk and neck found to be fore/aft sewing distance, sewing desk inclination. The sewing desk height influences the arm posture significantly. Based on the results, the following guidelines were developed: (1) Fore/aft sewing distance should be adjusted to 140 mm towards the sewing operator; (2) a 10° sewing desk inclination towards sewing should be used at sewing workstations. (3) Sewing desk height should be adjusted between 762 mm and 787 mm from the ground.
Institution pages aggregate content on ResearchGate related to an institution. The members listed on this page have self-identified as being affiliated with this institution. Publications listed on this page were identified by our algorithms as relating to this institution. This page was not created or approved by the institution. If you represent an institution and have questions about these pages or wish to report inaccurate content, you can contact us here.
121 members
Dr. Nitin W. Ingole
  • Department of Civil Engineering
Tushar R. Deshmukh
  • Department of Mechanical Engineering
Nitin B Ingale
  • Department of Applied Physics
Chandrashekhar Deshmukh
  • Department of Electronics and Telecommunication Engineering
Amrāvati, India