Augmented grasshopper optimization algorithm by differential evolution: a power scheduling application in smart homes

With the increasing number of electricity consumers, production, distribution, and consumption problems of produced energy have appeared. This paper proposed an optimization method to reduce the peak demand using smart grid capabilities. In the proposed method, a hybrid Grasshopper Optimization Algorithm (GOA) with the self-adaptive Differential Evolution (DE) is used, called HGOA. The proposed method takes advantage of the global and local search strategies from Differential Evolution and Grasshopper Optimization Algorithm. Experimental results are applied in two scenarios; the first scenario has universal inputs and several appliances. The second scenario has an expanded number of appliances. The results showed that the proposed method (HGOA) got better power scheduling arrangements and better performance than other comparative algorithms using the classical benchmark functions. Moreover, according to the computational time, it runs in constant execution time as the population is increased. The proposed method got 0.26?% enhancement compared to the other methods. Finally, we found that the proposed HGOA always got better results than the original method in the worst cases and the best cases.

     

Synchronization of Fractional Order Neutral Type Fuzzy Cellular Neural Networks with Discrete and Distributed Delays via State Feedback Control

Neural Processing Letters - The motivation behind this paper is to explore the issue of synchronization of fractional order neutral type fuzzy cellular neural networks with state feedback control....     

Video compression using frame redundancy elimination and discrete cosine transform coefficient reduction

The need of human beings for better social media applications has increased tremendously. This increase has necessitated the need for a digital system with a larger storage capacity and more processing power. However, an increase in multimedia content size reduces the overall processing performance. This occurs because the process of storing and retrieving large files affects the execution time. Therefore, it is extremely important to reduce the multimedia content size. This reduction can be achieved by image and video compression. There are two types of image or video compression: lossy and lossless. In the latter compression, the decompressed image is an exact copy of the original image, while in the former compression, the original and the decompressed image differ from each other. Lossless compression is needed when every pixel matters. This can be found in autoimage processing applications. On the other hand, lossy compression is used in applications that are based on human visual system perception. In these applications, not every single pixel is important; rather, the overall image quality is important. Many video compression algorithms have been proposed. However, the balance between compression rate and video quality still needs further investigation. The algorithm developed in this research focuses on this balance. The proposed algorithm exhibits diversity of compression stages used for each type of information such as elimination of redundant and semi redundant frames, elimination by manipulating consecutive XORed frames, reducing the discrete cosine transform coefficients based on the wanted accuracy and compression ratio. Neural network is used to further reduce the frame size. The proposed method is a lossy compression type, but it can reach the near-lossless type in terms of image quality and compression ratio with comparable execution time.

     

Enhanced halftone-based secure and improved visual cryptography scheme for colour/binary Images

Visual Cryptography (VC) is gaining attraction during the past few years to secure the visual information in the transmission network. It enables the visual data i.e. handwritten notes, photos, printed text, etc. to encrypt in such a way that their decryption can be done through the human visual framework. Hence, no computational assistance is required for the decryption of the secret images they can be seen through naked eye. In this paper, a novel enhanced halftoning-based VC scheme is proposed that works for both binary and color images. Fake share is generated by the combination of random black and white pixels. The proposed algorithm consists of 3 stages i.e., detection, encryption, and decryption. Halftoning, Encryption, (2, 2) visual cryptography and the novel idea of fake share, make it even more secure and improved. As a result, it facilitates the original restored image to the authentic user, however, the one who enters the wrong password gets the combination of fake share with any real share. Both colored and black images can be processed with minimal capacity using the proposed scheme.

     

Recognizing arabic handwritten characters using deep learning and genetic algorithms

Automated techniques for Arabic content recognition are at a beginning period contrasted with their partners for the Latin and Chinese contents recognition. There is a bulk of handwritten Arabic archives available in libraries, data centers, historical centers, and workplaces. Digitization of these documents facilitates (1) to preserve and transfer the country’s history electronically, (2) to save the physical storage space, (3) to proper handling of the documents, and (4) to enhance the retrieval of information through the Internet and other mediums. Arabic handwritten character recognition (AHCR) systems face several challenges including the unlimited variations in human handwriting and the leakage of large and public databases. In the current study, the segmentation and recognition phases are addressed. The text segmentation challenges and a set of solutions for each challenge are presented. The convolutional neural network (CNN), deep learning approach, is used in the recognition phase. The usage of CNN leads to significant improvements across different machine learning classification algorithms. It facilitates the automatic feature extraction of images. 14 different native CNN architectures are proposed after a set of try-and-error trials. They are trained and tested on the HMBD database that contains 54,115 of the handwritten Arabic characters. Experiments are performed on the native CNN architectures and the best-reported testing accuracy is 91.96%. A transfer learning (TF) and genetic algorithm (GA) approach named “HMB-AHCR-DLGA” is suggested to optimize the training parameters and hyperparameters in the recognition phase. The pre-trained CNN models (VGG16, VGG19, and MobileNetV2) are used in the later approach. Five optimization experiments are performed and the best combinations are reported. The highest reported testing accuracy is 92.88%.

     

Role of situational method engineering to improve visual information systems in agile distributed environment

Software product prone to continuous evolution due to increase in the use of technology. Therefore, more stakeholders are involved in software evolution increases the cost and complexity. This required optimization of resources and cost to handle evolution with Global Software Development (GSD) to utilize time zones efficiently. The significance challenge of GSD reports: time zone difference, geographical location, communication delays, knowledge sharing, control among stakeholders and development team. Because of these challenges, the requirements for development in GSD environment are also challenge as compared to on site development. Different requirement engineering methods have been used to improve the requirements analysis to deal with ambiguities and inconsistency in large set of requirements. The customization and tailoring of requirements according to changing project’s situations required to improve project development with reusing existing agile methods during requirement engineering phase. Moreover, complex information systems where heterogeneity is inevitable that implies the involvement of divergent stakeholders and necessitate a comprehensive framework to capture multidimensional viewpoints and fulfill aforementioned issues. Therefore, a situational multi-dimensional agile requirement engineering method has been proposed to support team and stakeholders’ viewpoints. The schema of the proposed method is based on challenges recognized by performing Literature Review. Then proposed method has been evaluated via experimental approach and statistical analysis conducted to validated reliability of data collected. This result is significant approved both practically and statistically that the proposed approach ease to use, implement, trained and increased productivity and performance. Hence, the experimental study for the evaluation of the proposed approach results concluded that, proposed approach is the important multimedia tool for supporting organization and distributed development team for information sharing, collaboration, product development.

     

Color image DNA encryption using mRNA properties and non-adjacent coupled map lattices

This paper proposes a novel algorithm for encrypting color images. The innovation in this study is the use of messenger ribonucleic acid (mRNA) encoding to import into Deoxyribonucleic acid (DNA) encoding. For permutation of the plain image bits, we use Arnold’s Cat Map at the bit-level. Then, using Non-Adjacent Coupled Map Lattices (NCML), we apply diffusion operations to the permuted color channels. We also provide the upgrade of the diffusion phase with DNA encoding. In the proposed algorithm, the choices are random depending on the secret key, which is implemented using a simple logistic map. Hashing the string entered by the user, the secret key, parameters, and initial values are generated by the Double MD5 method. The results of tests and security analysis showed that the results of encryption with this scheme are effective, and the key space is large enough to withstand common attacks.

     

Automatic detection of coronavirus disease (COVID-19) using X-ray images and deep convolutional neural networks

The 2019 novel coronavirus disease (COVID-19), with a starting point in China, has spread rapidly among people living in other countries and is approaching approximately 101,917,147 cases worldwide according to the statistics of World Health Organization. There are a limited number of COVID-19 test kits available in hospitals due to the increasing cases daily. Therefore, it is necessary to implement an automatic detection system as a quick alternative diagnosis option to prevent COVID-19 spreading among people. In this study, five pre-trained convolutional neural network-based models (ResNet50, ResNet101, ResNet152, InceptionV3 and Inception-ResNetV2) have been proposed for the detection of coronavirus pneumonia-infected patient using chest X-ray radiographs. We have implemented three different binary classifications with four classes (COVID-19, normal (healthy), viral pneumonia and bacterial pneumonia) by using five-fold cross-validation. Considering the performance results obtained, it has been seen that the pre-trained ResNet50 model provides the highest classification performance (96.1% accuracy for Dataset-1, 99.5% accuracy for Dataset-2 and 99.7% accuracy for Dataset-3) among other four used models.

     

Simple Robust PID Tuning for Magnetic Levitation Systems Using Model-free Control and $${{\cal H}_\infty}$$ ℋ ∞ Control Strategies

International Journal of Control, Automation and Systems - This paper proposes two control techniques to provide robust tracking for magnetic levitation systems (MLS): model-free control (MFC) and...     

Process Mining Approach of a New Water Quality Index for Long-Term Assessment under Uncertainty Using Consensus-Based Fuzzy Decision Support System

One of the biggest challenges in water quality monitoring is how to optimize big Data gathered from a wide range of resources. This paper presented a new software-based pathway of process mining approach for extending a flexible WQI (Water Quality Index) that would deal with uncertainties derived from missing data occurrence in short- and long-term assessments. The methodology is based on integration of four multi-criteria group decision-making models coupled with fuzzy simulation including AHP (Analytical Hierarchy Process), fuzzy OWA (Ordered Weighting Average), TOPSIS (Technique for Order Preference by Similarity to Ideal Solution), and fuzzy TOPSIS that were used for data mining and group consensus evaluation.. Examining the methodology on groundwater resources being supplied for drinking in Shiraz, Iran showed high integrity, accuracy, and proximity-to-real interpretation of water quality. This was the first study where decision-making risks such as Decision Makers’ risk-prone or risk-aversion attitudes (optimistic degree), DMs’ power, and consensus degree of each water quality parameter have been considered in WQI research. The proposed index offered a flexible choice in defining the intended project duration, stakeholders’ judgments, types of water use and water resource, standards, as well as type and number of water quality parameters. Thus, beside sustaining the unity in structure, this methodology could be suggested as a potentially WQI for other regions. The presented methodology would help more efficient monitoring of water resources for drinking purpose with respect to water quality.