An Efficient Handwritten Digit Recognition based on Convolutional Neural Networks with Orthogonal Learning Strategies

The predictions of characters/text/digits from the handwritten images have made the research community spotlight towards recognition. There are enormous applications and ambiguity that made prediction possible with Deep Learning (DL) approaches. Primarily, there are four necessary steps to be carried out with handwriting prediction. First, consideration of a dataset that is more appropriate for DL validation an inefficient manner. Here, Special Database 1 and Special Database 2 are used, which are combined and modified by the National Institute of Standards and Technology (NIST). Next is pre-processing of input handwritten digit recognition data by data normalization, extraction of efficient features which provides better prediction accuracy. The proposed idea uses pixel values as features with the analysis of hyper-parameters to enhance near-human performance. With SVM, non-linear and linear models are built to extract the appropriate features for further processing. The features are separate and placed over the Bag of Features (BoF), which is used by the next processing stage. Finally, a novel Convolutional Neural Network (CNN) is by built modifying the network structure with Orthogonal Learning Particle Swarm Optimization (CNN-OLPSO). This modification is adopted for evolutionarily optimizing the number of hyper-parameters. This proposed optimizer predicts the optimal values from the fitness computation and shows better efficiency when compared to various other conventional approaches. The novelty which relies on CNN adoption is to endeavor a suitable path towards digitalization and preserve the handwritten structure and help automatic feature extraction using CNN by offering better computation accuracy. The optimization approach helps to avoid over-fitting and under-fitting issues. Here, metrics like accuracy, elapsed time, recall, precision, and [Formula: see text]-measure are evaluated. The results of CNN-OLPSO give better accuracy, reduced error rate and better execution time (s) compared to other existing methods. Thus, the proposed model shows better tradeoff in the recognition rate of handwritten digits.

Orthogonal Learning-Based Gray Wolf Optimizer For Identifying The Uncertain Parameters of Various Photovoltaic Models

Determining the optimal parameters for the photovoltaic system (PV) model is essential during the design, evolution, development, estimation, and PV systems analysis. Therefore, it is crucial for the proper advancement of the best parameters of the PV models based on modern computational techniques. Thus, this work suggests a new Orthogonal-Learning-Based Gray Wolf Optimizer (OLBGWO) through a local exploration for estimating the unknown variables of PV cell models. The exploitation and exploration capability of the basic Gray Wolf Optimizer (GWO) is improved by the orthogonal-learning-based (OLB) approach, and this arrangement promotes a highly reliable equilibrium between the exploitation and exploration levels of the algorithm. In OLBGWO, the OLB strategy is used to find the best solution for the poor populations and directs the population to review the potential search area during the iterative process. Also, an exponential decay function is employed to decrease the value of vector a in GWO. The developed algorithm is directly applied to the parameter identification problem of the PV system. The proposed OLBGWO algorithm estimates the unknown parameters of the single-diode model (SDM), double-diode model (DDM), and PV module model. The performance of the OLBGWO is compared with other competitive algorithms to prove its superiority. The simulation results prove that the OLBGWO algorithm can achieve high solution accuracy with high convergence speed.

A novel krill herd algorithm with orthogonality and its application to data clustering

Krill herd algorithm (KHA) is an emerging nature-inspired approach that has been successfully applied to optimization. However, KHA may get stuck into local optima owing to its poor exploitation. In this paper, the orthogonal learning (OL) mechanism is incorporated to enhance the performance of KHA for the first time, then an improved method named orthogonal krill herd algorithm (OKHA) is obtained. Compared with the existing hybridizations of KHA, OKHA could discover more useful information from historical data and construct a more promising solution. The proposed algorithm is applied to solve CEC2017 numerical problems, and its robustness is verified based on the simulation results. Moreover, OKHA is applied to tackle data clustering problems selected from the UCI Machine Learning Repository. The experimental results illustrate that OKHA is superior to or at least competitive with other representative clustering techniques.