Performance improvement of Deep Learning Models using image augmentation techniques

The major barrier while using deep learning models is lack of large number of images in the training dataset. In fact, there is a need of thousands of images in each image categories based on the complexity of problem. Prior studies have shown that picture augmentation techniques can be used to enhance the number of images in a training dataset artificially. These techniques can aid in improving the overall learning process and performance of a deep learning model. Hence, to address this problem we have proposed three algorithms. Firstly, two image acquisition algorithms have been proposed to systematically obtain real field images for testing and images from public datasets for training a model. Secondly, an algorithm is proposed to describe the procedure how the augmentations can be applied to enhance the datasets. During this study, we have investigated 52 augmentations that can allow enhancing the size of input dataset by improving the quantity of images. To perform the classification process of four maize crop diseases, a new convolutional neural network model is developed and several experiments have been performed to prove its effectiveness. Firstly, two tests were carried out using the original dataset from Kaggle public repository and the augmented dataset. When compared with the original dataset, the model improved by 5.14% with the augmented dataset. Secondly, three experiments carried out to evaluate the performance of proposed augmentation method. Experimental results demonstrated that the proposed approach outperforms the existing three approaches by 27.38%, 3.14%, and 1.34% during the classification process. The proposed IPA augmentation method has been compared with six existing methods: Full Stage Data Augmentation Framework, LeafGAN, Novel Augmentation method based on GAN, Wasserstein Generative Adversarial Network (WGAN), Activation Reconstruction-GAN, and Step-by-Step Data Augmentation Method and experimental results show that performance is better than existing methods by 28.31%, 19.76%, 20.18%, 13.75%, 2.42%, and 12.68% respectively.

     

Rheology of nylon 6 containing metal halides

Addition of metal halides to nylons has been shown to be advantageous in a number of ways. The decrease in melting temperature, increase in glass transition temperature and melt viscosity by such additions have allowed more convenient processing of low molecular weight polymers and thermally unstable polymers. Rheological data depicting the variation of melt viscosity with shear rate at temperatures relevant to processing are necessary in optimizing and trouble-shooting plastics processing operations. In the present paper, a method has been proposed to estimate, complete flow curves or rheograms of nylon—metal halide systems with the use of a master curve knowing the melt flow index and glass transition temperature of the system. The validity of the approach has been verified for the nylon 6—lithium chloride system and shown to hold good for any nylon—metal halide combination.     

Sequence determination in MMA-EMA copolymers by 13C NMR

The ¹³C nuclear magnetic resonance spectra of MMA-EMA copolymers have been recorded and analyzed. The carbonyl carbon resonance region contains six resonating peaks due to its sensitivity to triad concentrations. From these signals, triad concentrations were determined and compared with calculated values based on Bernoullian and first-order Markovian statistics. Molecular weights were determined by viscosity measurement method.     

Frustrated-total-internal-reflection fiber-optic pressure sensor

In this work the design and construction of a simple fiber-optic pressure sensor based on the principle of frustrated-total-internal-reflection in a fiber cut beyond the critical angle are described. The characteristic of the device is obtained and compared with the predictions of a simple theory. Modifications to alter its sensitivity by the use of different wavelengths and fiber cut angle are discussed.     

A Rate \text{ M }_{\mathrm{T}} Full Diversity STF Block Coded 4\times 4 MIMO-OFDM System with Reduced Complexity

Multiple input multiple output (MIMO) communication systems with orthogonal frequency division multiplexing (OFDM) has a great role to play for 4G broadband wireless communications. In this paper, a space time frequency (STF) code is presented with reduced decoder complexity and to achieve code rate $\text{ M }_\mathrm{T}$ with full diversity of $\text{ M }_{\mathrm{T}} \text{ M }_{\mathrm{R}} \text{ N }_{\mathrm{b}}$ L i.e., product of number of transmit antennas ( $\text{ M }_\mathrm{T}$ ), receive antennas $(\text{ M }_{\mathrm{R}})$ , fading blocks $(\text{ N }_{\mathrm{b}})$ and channel taps (L). The maximum achievable diversity with high rate of STF block coded MIMO-OFDM is analyzed and verified by simulation results. The decoder complexity is resolved by employing several approaches like maximum likelihood (ML), sphere decoder (SD) and array processing. The performance of STF code is compared with existing layered algebraic STF code in terms of decoder complexity and bit error rate (BER). Further, the closed form expressions for BER performance of STFBC MIMO-OFDM systems are derived and evaluated for frequency selective block fading channels with MPSK constellations.     

KERNEL: Enabler to build smart surrogates for online optimization and knowledge discovery

KERNEL – A novel parameter-free surrogate building algorithm using Adaptive Neuro Fuzzy Inference System (ANFIS) is presented to provide an intelligent and robust technology to optimally estimate the configuration of ANFIS along with Sobol-based fast sample size determination (SSD) methodology. The proposed algorithm is capable of fine-tuning the existing knowledge base about the physics of the process in terms of human experience. It also enables knowledge discovery through a multi-objective optimization problem (MOOP) solved by non-dominated sorting genetic algorithm, NSGA-II, thus presenting machine-invented physics of the process. Experimentally validated polymerization reaction network model is considered and ANFIS surrogates are built using KERNEL. Surrogate-based optimization was found to be nine times faster than conventional optimization using the time expensive model, thus enabling its online implementation. Comparison of ANFIS with Kriging is also included.     

Microbicidal activity of TiO2 nanoparticles synthesised by sol–gel method

In this study, the authors investigated antimicrobial activity of TiO₂ nanoparticles (NPs) synthesised by sol–gel method. As synthesised TiO₂ NPs were characterised by X‐ray diffraction, scanning electron microscopy and ultraviolet‐visible absorption spectroscopy. The antimicrobial activity of calcined TiO₂ nanoparticle samples was examined in day light on Gram positive bacteria (Staphylococcus aureus, Streptococcus pneumonia and Bacillus subtilis), Gram negative bacteria (Proteus vulgaris, Pseudomonas aeruginosa and Escherichia coli) and fungal test pathogen Candida albicans. The synthesised TiO₂ NPs were found to be effective in visible light against Streptococcus pneumonia, Staphylococcus aureus, Proteus vulgaris, Pseudomonas aeruginosa and Candida albicans.Inspec keywords: titanium compounds, microorganisms, nanomedicine, biomedical materials, nanofabrication, sol‐gel processing, ultraviolet spectra, visible spectra, X‐ray diffraction, scanning electron microscopy, nanoparticles, antibacterial activityOther keywords: microbicidal activity, titanium dioxide nanoparticle, sol‐gel method, antimicrobial activity, X‐ray diffraction, scanning electron microscopy, ultraviolet‐visible absorption spectroscopy, Gram positive bacteria, Staphylococcus aureus, Streptococcus pneumonia, Bacillus subtilis, TiO₂ , Candida albicans, fungal test pathogen, Escherichia coli, Pseudomonas aeruginosa, Proteus vulgaris, Gram negative bacteria     

Inhomogeneous Electronic Structure of FeTe1−x Se x Revealed by X-ray Absorption Near Edge Structure

Ab initio Fe K-edge X-ray absorption near edge structure (XANES) calculations on FeTe_1?x Se _x , based on its crystallographic structure, are compared with experimental data. The calculated XANES spectra are inconsistent with the observed increase of spectral weight in the preedge region of the experimental spectra on the ternary system in which Te is substituted by Se. However, the weighted average of the XANES of FeTe and FeSe binaries reproduce well the observed trend in the experimental spectra of the ternary systems. This suggests that the FeTe_1?x Se _x is characterized by an inhomogeneous local electronic structure, resulting from a random-alloy like local atomic structure. The calculated unoccupied Fe d-density of states at the Fermi level increases monotonically with Te content unlike the observed change in T _c with Te content. T _c increases with increasing Te content up to x=0.5; however, the superconductivity is suppressed in the FeTe due to magnetic order, which is not considered in these calculations.     

Conservation v. generation: The significance of Demand-Side Management (DSM), its tools and techniques

A balance between electricity supply and demand is needed at all times to ensure a stable and reliable market. This can be done in two ways, either by Supply-side Management by adding supply when demand is high or by Demand-Side Management (DSM) by curtailing the system demand when supply availability is less. Satish Saini explores the benefits of DSM as a solution to supply shortage and alternative to new plant build.