Showing papers by "National Institute of Technology, Karnataka published in 2018"

Tomato Leaf Disease Detection Using Convolutional Neural Networks

Abstract: The tomato crop is an important staple in the Indian market with high commercial value and is produced in large quantities. Diseases are detrimental to the plant's health which in turn affects its growth. To ensure minimal losses to the cultivated crop, it is crucial to supervise its growth. There are numerous types of tomato diseases that target the crop's leaf at an alarming rate. This paper adopts a slight variation of the convolutional neural network model called LeNet to detect and identify diseases in tomato leaves. The main aim of the proposed work is to find a solution to the problem of tomato leaf disease detection using the simplest approach while making use of minimal computing resources to achieve results comparable to state of the art techniques. Neural network models employ automatic feature extraction to aid in the classification of the input image into respective disease classes. This proposed system has achieved an average accuracy of 94–95 % indicating the feasibility of the neural network approach even under unfavourable conditions.

Preserving Semantic Relations for Zero-Shot Learning

Abstract: Zero-shot learning has gained popularity due to its potential to scale recognition models without requiring additional training data. This is usually achieved by associating categories with their semantic information like attributes. However, we believe that the potential offered by this paradigm is not yet fully exploited. In this work, we propose to utilize the structure of the space spanned by the attributes using a set of relations. We devise objective functions to preserve these relations in the embedding space, thereby inducing semanticity to the embedding space. Through extensive experimental evaluation on five benchmark datasets, we demonstrate that inducing semanticity to the embedding space is beneficial for zero-shot learning. The proposed approach outperforms the state-of-the-art on the standard zero-shot setting as well as the more realistic generalized zero-shot setting. We also demonstrate how the proposed approach can be useful for making approximate semantic inferences about an image belonging to a category for which attribute information is not available.

A Switched-Capacitor-Based Multilevel Inverter Topology With Reduced Components

Abstract: This letter presents an improved sensorless nine-level inverter topology with reduced number of components. It is formed by cascading a three-level T-type neutral clamped point inverter with a floating capacitor (FC) fed two-level converter unit. Additionally, two line-frequency switches are appended across the dc-link. A simple logic-form equations-based pulse width modulator is designed which is in-charge of maintaining the FC voltage at its reference value without any aid of voltage and current sensor. Thus, the complexity in control of the proposed topology is very minimal. The working principle of the proposed inverter and formulation of logic-form equations is deliberated in detail. Furthermore, experimental results obtained from the developed prototype are presented to validate feasibility and operability of the proposed topology. Finally, a comprehensive comparison with some of the recently reported inverter topologies proving the merits of the proposed topology is included.

Green Synthesis of Nanomaterials

Abstract: The nanotechnology industry is increasingly promoting nano as a “green” technology that will improve the environmental performance of existing industries, reduce consumption of resources and energy, and allow achievement of environmentally benign economic expansion. Eco-friendly solutions are gaining popularity in the contemporary world. A green method for nanoparticle synthesis should be assessed considering three aspects: the solvent, the capping agent, and the reducing agent compared to physical and chemical methods. Particularly, the plant extracts mediated process is a good and advantageous method for the development of metal nanoparticles compared to using microorganisms, in which the cell maintenance time is limited. Synthesis and characterization of nanoparticles are important steps to be adopted to apply nanoparticles in field applications, and these steps include preparation of leaf extract, Phytochemical screening, and preparation of precursor. Various factors affecting the synthesis of nanoparticles, such as pH, temperature, and time, will be discussed. The degradation of any organic compounds by the green approach (plant extracts) is mainly due to the presence of polyphenols in the biodegradable material. Although the synthesis of nanoparticles has been trending higher, their application in the area of waste water treatment has been limited until recently. Hence, the application of nanoparticles to waste water treatment will be discussed with a view toward paving the way for an alternate source of water.

Removal of metal ions and humic acids through polyetherimide membrane with grafted bentonite clay

Abstract: Functional surfaces and polymers with branched structures have a major impact on physicochemical properties and performance of membrane materials. With the aim of greener approach for enhancement of permeation, fouling resistance and detrimental heavy metal ion rejection capacity of polyetherimide membrane, novel grafting of poly (4-styrenesulfonate) brushes on low cost, natural bentonite was carried out via distillation-precipitation polymerisation method and employed as a performance modifier. It has been demonstrated that, modified bentonite clay exhibited significant improvement in the hydrophilicity, porosity, and water uptake capacity with 3 wt. % of additive dosage. SEM and AFM analysis showed the increase in macrovoides and surface roughness with increased additive concentration. Moreover, the inclusion of modified bentonite displayed an increase in permeation rate and high anti-irreversible fouling properties with reversible fouling ratio of 75.6%. The humic acid rejection study revealed that, PEM-3 membrane having rejection efficiency up to 87.6% and foulants can be easily removed by simple hydraulic cleaning. Further, nanocomposite membranes can be significantly employed for the removal of hazardous heavy metal ions with a rejection rate of 80% and its tentative mechanism was discussed. Conspicuously, bentonite clay-bearing poly (4-styrenesulfonate) brushes are having a synergistic effect on physicochemical properties of nanocomposite membrane to enhance the performance in real field applications.

The Effects of Credible Online Reviews on Brand Equity Dimensions and Its Consequence on Consumer Behavior

Abstract: Consumers are using social media platform to gain and share knowledge on brands. In the virtual environment, consumers are exposed to various online reviews on brands that leave an impression of br...

Adsorption and photocatalytic properties of NiO nanoparticles synthesized via a thermal decomposition process

Abstract: NiO nanoparticles (NPs) were synthesized at different annealing temperatures via a thermal decomposition process and characterized using X-ray diffraction, scanning electron microscopy, and UV-vis spectroscopy. The NiO NPs prepared at higher annealing temperature (400 °C) were shown excellent adsorption and photocatalytic activity toward textile dyes reactive black 5 (RB-5) and methylene blue (MB). About 87.2% of RB-5 in 60 min and 70.2% of MB in 5 h was removed using NiO NPs synthesized at 400 °C. The photocatalytic degradation of MB was found to increase with an increase in the annealing temperature of the catalyst. Moreover, the kinetic study revealed that the adsorption and photocatalytic activity of NiO NPs followed the second and first-order kinetics, respectively. The enhanced performance of NiO NPs toward dye removal might be related to its optical and structural properties.

Additive Manufacturing of Syntactic Foams: Part 1: Development, Properties, and Recycling Potential of Filaments

Abstract: This work focuses on developing filaments of high-density polyethylene (HDPE) and their hollow particle-filled syntactic foams for commercial three-dimensional (3D) printers based on fused filament fabrication technology. Hollow fly-ash cenospheres were blended by 40 wt.% in a HDPE matrix to produce syntactic foam (HDPE40) filaments. Further, the recycling potential was studied by pelletizing the filaments again to extrude twice (2×) and three times (3×). The filaments were tensile tested at 10−4 s−1, 10−3 s−1, and 10−2 s−1 strain rates. HDPE40 filaments show an increasing trend in modulus and strength with the strain rate. Higher density and modulus were noticed for 2× filaments compared to 1× filaments because of the crushing of some cenospheres in the extrusion cycle. However, 2× and 3× filament densities are nearly the same, showing potential for recycling them. The filaments show better properties than the same materials processed by conventional injection molding. Micro-CT scans show a uniform dispersion of cenospheres in all filaments.

Groundwater quality assessment of urban Bengaluru using multivariate statistical techniques

Abstract: Groundwater quality deterioration due to anthropogenic activities has become a subject of prime concern. The objective of the study was to assess the spatial and temporal variations in groundwater quality and to identify the sources in the western half of the Bengaluru city using multivariate statistical techniques. Water quality index rating was calculated for pre and post monsoon seasons to quantify overall water quality for human consumption. The post-monsoon samples show signs of poor quality in drinking purpose compared to pre-monsoon. Cluster analysis (CA), principal component analysis (PCA) and discriminant analysis (DA) were applied to the groundwater quality data measured on 14 parameters from 67 sites distributed across the city. Hierarchical cluster analysis (CA) grouped the 67 sampling stations into two groups, cluster 1 having high pollution and cluster 2 having lesser pollution. Discriminant analysis (DA) was applied to delineate the most meaningful parameters accounting for temporal and spatial variations in groundwater quality of the study area. Temporal DA identified pH as the most important parameter, which discriminates between water quality in the pre-monsoon and post-monsoon seasons and accounts for 72% seasonal assignation of cases. Spatial DA identified Mg, Cl and NO3 as the three most important parameters discriminating between two clusters and accounting for 89% spatial assignation of cases. Principal component analysis was applied to the dataset obtained from the two clusters, which evolved three factors in each cluster, explaining 85.4 and 84% of the total variance, respectively. Varifactors obtained from principal component analysis showed that groundwater quality variation is mainly explained by dissolution of minerals from rock water interactions in the aquifer, effect of anthropogenic activities and ion exchange processes in water.

Quasi-static compressive response of compression molded glass microballoon/HDPE syntactic foam

Abstract: Quasi-static compressive behavior of different density glass microballoon (GMB) reinforced high density polyethylene (HDPE) syntactic foams are investigated in the present work. Reducing the weight of thermoplastic components has been always a high priority in transportation, aerospace, consumer products and underwater vehicle structures. Despite continued interest in developing lightweight thermoplastic syntactic foams, they have not been studied extensively for quasi-static response with focus on wall thickness and volume fraction variations. Compression molded GMB/HDPE sheets are subjected to 0.001, 0.01 and 0.1 s−1 strain rates. Compressive modulus of foams is higher compared to neat HDPE. Increasing strain rates and decreasing filler content increases yield strength for all the foams investigated compared to neat HDPE. Yield strain and energy absorption of GMB/HDPE foams increases with an increasing strain rate and wall thickness. Specific modulus and strength of GMB/HDPE foams are superior and are comparable to neat HDPE. GMB/HDPE foam achieved high stiffness to weight ratio making them suitable for wide variety of applications. Theoretical model based on differential scheme predicts a good estimate of elastic modulus for all the type of GMB/HDPE foams. Finally, property map is exhibited to present comparative studies with existing literature.

Compressive behavior of cenosphere/epoxy syntactic foams in arctic conditions

Abstract: In this paper, the effects of arctic condition on the compressive response of ceno-sphere/epoxy syntactic foams are investigated. Understanding the behavior of such foams under extreme conditions is critical for exploring their suitability for constructing lightweight platforms used in arctic explorations, which are exposed to subzero temperatures for extended periods of time potentially degrading their mechanical properties. In the research study presented here, samples of cenosphere/epoxy syntactic foams were conditioned under arctic environment at a temperature of −60 °C for a period of 57 days. Compression tests were then conducted at room temperature as well as in-situ −60 °C on the conditioned samples and compared against unconditioned samples tested at room temperature. Combinations of surface modification and cenosphere volume fractions were considered. For the case of unconditioned samples, compressive strength decreased with increasing cenosphere volume fraction for both surface modified and unmodified cenospheres. For the arctic conditioned samples, cenospheres/epoxy foams did not present visible signs of degradation prior to testing, but manifested a reduction in compressive modulus in a range of 47–57% and 47–65% for untreated and treated cenospheres/epoxy syntactic foams as compared to their unconditioned counterparts. On the other hand, the compressive strength increased in a range between 32–68% for untreated and 59–80% for treated cenosphere foams in arctic environment, which can be attributed to the matrix hardening introduced by frigid in-situ environment. Also, under in-situ arctic compressive loading, the post peak response for all foam types have shifted from a progressive failure to a brittle type behavior.

Additive Manufacturing of Syntactic Foams: Part 2: Specimen Printing and Mechanical Property Characterization

Abstract: High-density polyethylene (HDPE) and its fly ash cenosphere-filled syntactic foam filaments have been recently developed. These filaments are used for three-dimensional (3D) printing using a commercial printer. The developed syntactic foam filament (HDPE40) contains 40 wt.% cenospheres in the HDPE matrix. Printing parameters for HDPE and HDPE40 were optimized for use in widely available commercial printers, and specimens were three-dimensionally (3D) printed for tensile testing at strain rate of 10−3 s−1. Process optimization resulted in smooth operation of the 3D printer without nozzle clogging or cenosphere fracture during the printing process. Characterization results revealed that the tensile modulus values of 3D-printed HDPE and HDPE40 specimens were higher than those of injection-molded specimens, while the tensile strength was comparable, but the fracture strain and density were lower.

A 20-Pulse Asymmetric Multiphase Staggering Autoconfigured Transformer For Power Quality Improvement

Abstract: This letter presents a 20-pulse multiphase staggering autoconfigured transformer (MSAT) for power quality improvement in medium and high power applications. The proposed autoconfigured transformer is a combination of delta and zig-zag transformers. Contrary to the prior art multipulse autoconnected transformers, magnetic rating of the proposed configuration is only 40% of the load rating in addition to improved power quality indices. Moreover, the creation of a neutral point ascribed to inbuilt zig-zag windings supplements as one of the remarkable advantages of the proposed MSAT. Therefore, with this suggested configuration, the performance of zig-zag transformer is accomplished within the autoconfigured transformer. A detailed comparison of the proposed MSAT configuration with that of other multipulse autoconnected transformers in terms of magnetic rating and power quality indices are presented. Further, the feasibility and operability of the proposed MSAT are verified and validated through experimental tests on the laboratory prototype. The test results so obtained complies with the IEEE-519 and IEC 61000-3-2 standards on power quality requirements.