Volume 7, Issue 4

Owing to the recent development of AI technology, various studies on computer-aided diagnosis systems for CT image interpretation are being conducted. In particular, studies on the detection of lung cancer which is leading the death rate are being conducted in image processing and artificial intelligence fields. In this study, to improve the anatomical interpretation ability of CT images, the lung, soft tissue, and bone were set as regions of interest and configured in each channel. The purpose of this study is to select a detector with optimal performance by improving the quality of CT images to detect lung cancer tumors. Considering the dataset construction phase, pixel arrays with Hounsfield units applied to the regions of interest (lung, soft tissue, and bone region) were configured as three-channeled, and a histogram processing the technique was applied to create a dataset with an enhanced contrast. Regarding the deep learning phase, the one-stage detector (RetinaNet) performs deep learning on the dataset created in the previous phase, and the detector with the best performance is used in the CAD system. In the evaluation stage, the original dataset without any processing was used as the reference dataset, and a two-stage detector (Faster R-CNN) was used as the reference detector. Because of the performance evaluation of the developed detector, a sensitivity, precision, and F1-score rates of 94.90%, 96.70%, and 95.56%, respectively, were achieved. The experiment reveals that an image with improved anatomical interpretation ability improves the detection performance of deep learning and human vision.

Read more…

This paper is an extended version of the journal presented at ICECCME2021. In ICECCME2021, the authors presented that we have developed a solitary wave track circuit (SW-TC), and it is energy-saving compared to existing track circuits. Furthermore, we also explained that it can realize advanced train control at a low cost, equivalent to digital automatic train control. After that, we have conducted research to improve preventive maintenance, which is a problem of existing track circuits, by using SW-TC. In this extended paper, we explain that we can further expand the functions of SW-TC, added new functions such as insulation deterioration diagnosis of the track circuit. With these new functions, the SW-TC can improve reliability, availability, maintainability, and safety. Especially, because of the effect of the insulation deterioration diagnosis function, so railway operators can significantly reduce the time required to identify the cause, when a track circuit failure occurs.

Read more…

This paper presents the design and implementation of a low-cost basket based on an ARM system on chip architecture using Raspberry Pi single board computer. The inspiration of this research is how to support the traditional low-income retail store in Thailand driving the local micro-business deal with the economic impacts of survival business from the global retailers. The concept of a smart basket system is to use the open-source software in order to save the budget with the free update system. For the product design, the kansei engineering and form follows function theory are applied. The low-cost basket consists of hardware design based on the system on chip architecture and software design using the proposed smart basket algorithm and user interface. Experimental results show that the proposed smart basket implementation can be convenient for lifestyle shopping experience in the local mini mart. This basket will replace the traditional one, which will help consumers maintain the social distancing and will support the local low-income merchant while running the local business during the COVID-19 pandemic.

Read more…

This paper presents the design and implementation of a low-cost basket based on an ARM system on chip architecture using Raspberry Pi single board computer. The inspiration of this research is how to support the traditional low-income retail store in Thailand driving the local micro-business deal with the economic impacts of survival business from the global retailers. The concept of a smart basket system is to use the open-source software in order to save the budget with the free update system. For the product design, the kansei engineering and form follows function theory are applied. The low-cost basket consists of hardware design based on the system on chip architecture and software design using the proposed smart basket algorithm and user interface. Experimental results show that the proposed smart basket implementation can be convenient for lifestyle shopping experience in the local mini mart. This basket will replace the traditional one, which will help consumers maintain the social distancing and will support the local low-income merchant while running the local business during the COVID-19 pandemic.

Read more…

In this paper, we investigate the utilization of three effective detection methods to identify potential threats of insects in date palm trees. The detection techniques presented here are the application of infrared radiation, microwave antennas and metamaterials based sensors. Experimental trials using IR radiation took place in a local farm. Moreover, the second sensing system is based on microwave antennas that are designed and numerically simulated at the 2.45 GHz-band. Lastly, the third detection method focuses on the design and development of low-power microwave sensor based on metamaterials concept. Based on the processed and analyzed results, the aforementioned sensing techniques are able to predict existence of red palm weevils within date palms.

Read more…

Hydrocarbon exploration and production has been successful in the central region of Doba basin, Chad, north-central Africa. In order to optimize the hydrocarbon production in this area, the combination of seismic and well log datas have been processed and analyzed to better characterize, image and capture the reservoirs. The 3D seismic and well log datas were used to obtain the horizon grids, fault polygons, and petrophysical parameters. The results show continuous and divergent horizons which are associated to differential subsidence and thickening of series on the inclined bedrock. The reservoir is an anticlinal structure where the hydrocarbons are trapped, in particular with reservoir levels interposed between the two stratigraphic sequences. Four mayors fault that cross the reservoir have been identified. The calculation of the average percentage of the encountered facies enabled to highlight the high percentage of sands compared to clays and clayey sands. Porosities are uniform in clays and sands in the two out of three wells, and higher in coarse sands. The permeabilities are average in sands and clays, but decrease in the fine sands. The 3-D static reservoir model integrated with structural and petrophysical parameters gives a better understanding of spatial distribution of the discrete and continuous reservoir properties. This work contributes to a future prediction of the reservoir performance, characteristics and production behavior in Doba basin.

Read more…

Mobility is a fundamental human desire. All societies aspire to safe and efficient mobility at low ecological and economic costs. ADAS systems (Advanced Driver Assistance Systems) are safety systems designed to eliminate human error in driving vehicles of all types. ADAS systems such as Radars use advanced technologies to assist the driver while driving and thus improve their performance. Radar uses a combination of sensor technologies to perceive the world around the vehicle and then provide information to the driver or take safety action when necessary. Conventional radars based on the emission of electromagnetic and ultrasonic waves have been consumed in the face of the challenges of the constraints of modern autonomous driving, and have not been generalized on all roads. For this reason, we studied the design and construction of a computer vision radar to reproduce human behavior, with a road line lane detection approach based on the histogram of the grayscale image that gives good estimates in real-time, and make a comparison of this method with other computer vision methods performed in the literature: Hough, RANSAC, and Radon.

Read more…

Various interruptions to the daily lives of researchers have necessitated the usage of simulations in projects which may not have initially relied on anything other than physical inquiry and experiments. The programs and algorithms introduced in this paper, which is an extended version of research initially published in ARAIG And Minecraft: A COVID-19 Workaround, create an optimized search space and egress path to the initial starting point of a user’s route using a modification (“mod”) of the digital game Minecraft. We initially utilize two approaches for creating a search space with which to find edges in the resulting graph of the user’s movement: a naive approach with the time complexity of O(n2) and an octree approach, with the time complexity of O(nlogn). We introduce a basic A* algorithm to search through the resulting graph for the most efficient egress path. We then integrate our mod with the visualization tool for the “As Real As It Gets” (ARAIG) haptic suit, which provides a visual representation of the physical feedback the user would receive if he were to wear it. We finish this paper by asking a group of four users to test this program and their feedback is collected.

Read more…

This article presents a roadmap for distribution grid µPMU hardware and software design consideration and implantation to ensure high performance within limited computational time of sampling frequency 512 samples/cycle. A proposed 12 channels, multi-voltage level µPMU hardware and rules of voltage and current transducer, analog filter, analog-to-digital converter, sampling rate definition, and PCB design and selection are presented. From the software view, software minimization procedures are implemented to reduce the estimation time of the proposed µPMU to 18 µsec under high sampling frequency operation. Additionally, error estimation and compensation are used to ensure robust performance, while the computational burden of the error compensation stage is reduced by Taylor series linearization. The proposed µPMU is designed to provide traditional phasor, frequency and harmonics measurements besides a point-on-wave under dynamic operation mode. The proposed device is tested under IEEE Std C37. 118.1 and 118.2 and showed accurate phasor estimation up to 0.03% for the magnitude and angle accuracy up to 0.0036o, while the frequency is estimated with maximum variation of 0.032% under dynamic operation.

Read more…

Applying black-box ML models in high-stakes fields like criminology, healthcare and real-time operating systems might create issues because of poor interpretability and complexity. Also, model building methods that include interpretability is now one of the growing research topics due to the absence of interpretability metrics that are both model-agnostic and quantitative. This paper introduces model selection methods with trade off between interpretability and accuracy of a model. Our results show 97% improvement in interpretability with 2.5% drop in accuracy in AutoMPG dataset using MLP model (65% improvement in interpretability with 1.5% drop in accuracy in MNIST dataset).

Read more…

Estimation of energy demand has important implications for economic and social stability leading to a more secure energy future. One-year-ahead energy demand estimation for Turkey has been proposed in this paper, using the metaheuristics method with GDP, the total population, and the quantities of imports and exports, as inputs variables. The records obtained from historical data were bifurcated into training and test datasets, where the training dataset is used by the algorithm in the process of generating models, while the test dataset was used to evaluate the performance of the algorithm. Here, two particular approaches have been proposed: Grammatical Evolution alone, and an ensemble of Grammatical Evolution with Differential Evolution. Under these four different forms are developed, viz, Grammatical Evolution with a recursive grammar (M1), an ensemble of Grammatical evolution executed on a linear grammar and Differential Evolution (M2), an ensemble of Grammatical evolution executed on a quadratic grammar and Differential Evolution (M3), and, Grammatical Evolution with a recursive grammar and Differential Evolution (M4). Moreover, the present approaches were also compared for estimation accuracy against the previously published DE models. It was substantiated that the M4 proposal exhibited the best performance towards estimation. It is therefore established that the current approach exhibits a better estimation capability (with RMSE of 2.2002), compared to the models previously available in the literature. M4 approach is then employed to predict the future energy demand using the same set of socio-economic inputs and the results demonstrated high prediction accuracy with an RMSE of 2.2278.

Read more…

In this paper, Compact bandpass filters have been designed. A single bandpass filter was designed using novel triple concentric complementary split-ring resonators placed along the microstrip line in the ground plane. Gaps and via were placed on the microstrip line to control electromagnetic characteristics, resulting in a single bandpass filter. In turn, spiral resonators were attached to the microstrip transmission line at the gaps in the transmission line to obtain a compact dual passband filter. Stepped impedance microstrip line and T-shaped stubs were attached to the microstrip line in between spiral resonators. The structure designed resulted in a Triple bandpass filter. A fractional bandwidth of 3% was achieved at the center frequency of 3GHz. The filter had a 1.5dB insertion loss which is the minimum value in the operating frequency band. The filter resonance frequency was 1.32 GHz and 2.47GHz which have a fractional bandwidth of 7.5% and 4.85% respectively and the corresponding insertion loss was 1.3dB and 1.8dB respectively. The triple bandpass filter had a fractional bandwidth of 1.16%, 11.4%, and 1.86%, centered at 1.29 GHz, 2.27 GHz, and 3.21GHz with 1.6dB, 1.3dB, and 1.8 dB insertion loss at the respective frequencies. The proposed bandpass filters are useful for GPS, WLAN, WiMAX, and radar applications.

Read more…

There are different factors which contribute to the amount of output power which can be delivered by solar photovoltaic (PV) module at any time of the year. The factors include but not limited to solar irradiation, ambient temperature, relative humidity, wind velocity, position of sun in the sky, geographical position of installed solar PV module and others. Apparent position of the sun in the sky contribute to the amount of electromagnetic radiation from the sun reaching the module’s surface area. With the sun further away from the module and irradiance reaching the module surface area at an angle non perpendicular to the surface leads to low output power delivered by the module. In southern hemisphere the PV module experience high output power around November/December which are summer months and low output power around June/July which are winter months. This paper develops performance adjustment factor of fixed solar PV module to adjust PV module output power such that the PV system can deliver required amount of power during winter season. The results show that the value of performance adjustment factor for fixed solar PV module or system was established and can be used to adjust performance or output power for winter periods.

Read more…

The aim of the article is to analyze the annual reports on cyber security of Central European countries, i.e. the Czech Republic, Slovakia, Poland, Germany, and Austria. The article focuses on the development of the state of cyber security, actors of threats in cyberspace, cyber threats, and the most common types of attacks. The article evaluates the objectives of cyber-attacks from the point of view of state institutions, organizations, and state and private companies, and they have listed the follow-up measures here. The method used is a critical verbal evaluation with comments and comparative analysis to find the strengths and weaknesses of the evaluated cyber security strategies and learn from them. The experiment of the cyber defense against phishing attacks is mentioned as an example of the cyber defense of individuals. The rules in Microsoft Outlook were used by filtering incoming email messages. The result is promising by stopping 88 % of phishing emails. The discussion and conclusion state that COVID-19 played a big role in the cyber security situation in countries to the analyzed documents.

Read more…

The false nearest neighbors (FNN) method estimates the variables of a system by sequentially embedding a time series into a higher-dimensional delay coordinate system and finding an embedding dimension in which the neighborhood of the delay coordinate vector in the lower dimension does not extend into the higher, that is, a dimension in which no false neighbors or neighborhoods exist. However, the FNN method requires an arbitrary threshold value to distinguish false neighborhoods, which must be considered each time for each time series to be analyzed. In this study, we propose a robust method to estimate the minimum embedding dimension, which eliminates the arbitrariness of threshold selection. We applied the proposed approach to the van der Pol and Lorenz equations as representative examples of chaotic time series. The results verified the accuracy of the proposed variable estimation method, which showed a lower error rate compared to the minimum dimension estimates for most of the thresholding intervals set by the FNN method.

Read more…

Radon transform is a mathematical tool widely applied in various domains, including biophysics and computer tomography. Previously, it was discovered that applying the Radon transform to a binary image comprising circle forms resulted in discontinuity. As a result, the line detection approach based on it became discontinued. The d-Radon transform is a modified version of the Radon transform that is presented as a solution to this problem. The properties of the circle cause the Radon transform to be discontinuous. This work extends this finding by looking into the Radon transform’s regularity property and a proposed modification to a convex shape. We discovered that regularity in the Radon space is determined by the regularity of the shape’s point. This leads to the continuity condition for the line detection method.

Read more…

In the time past, virtually all the contemporary communication systems depend on distinct source and channel encoding schemes for data transmission. Irrespective of the recorded success of the distinct schemes, the new developed scheme known as joint source channel coding technique has proven to have technically outperformed the conventional schemes. The aim of the study is centered in developing an enhanced joint source-channel coding scheme that could mitigate some of the limitations observed in the contemporary joint source channel coding schemes. The study tends to leverage on recent developments in machine learning known as deep learning techniques for robust and enhanced scheme, devoid of explicit code dependence for the signal compression and as well in error correction but learn automatically on end-to-end mapping structure for the source signals. It primarily aimed at providing an improved channel performance approach for wireless communication network. A deep learning algorithm was implemented in the study, the scheme focused on improving the Bit Error Rate (BER) performance while reducing latency and the processing complexity in Joint Source Channel Coding systems. The deep learning autoencoder model was deployed to compare with the hamming code, convolution code, and uncoded systems. JSCC using neural networks were simulated based on BER performance over a range of energy per symbol to noise ratio (Eb/No). Training and test error for the fully connected neural network autoencoder models on channels with 0.0dB and 8.0dB were carried out. The results obtained showed that the autoencoder model had a better BER performance when compared with the convolution code and uncoded systems, it also outperformed the uncoded BFSK with an approximately equal BER performance when compared with the hamming code (soft decision) decoding system.

Read more…

A nested Mach-Zehnder interferometer (MZI) configuration whose phase shifters are placed in parallel is suitable for silicon-silica hybrid structure to realize a high-speed optical switch. Even when the signal wavelength deviates from an optimal wavelength, the crosstalk of the nested MZI optical switch can be suppressed by employing phase generating couplers (PGCs) in place of directional couplers. We calculate the characteristics of a 4-stage nested MZI switch with PGCs, and show that crosstalk is lower than −40 dB over a wavelength range of as wide as 200 nm from 1450 to 1650 nm in six output ports. We also examine the scalability of the multi-stage nested MZI switch, and deduce the required number of switch stages for given output port counts with low crosstalk.

Read more…

This paper builds on the realization that since mobile devices have become a common tool for researchers to collect, process, and analyze large quantities of data, we are now entering a generation where the creation of solutions to difficult real-world problems will mostly come in the form of mobile device apps. One such relevant real-life problem is to accurately and cheaply detect the over-consumption of alcohol, since it can lead to many problems including fatalities. Today, there are several expensive and/or tedious alternative procedures in the market that are used to test subjects’ Blood Alcohol Content (BAC). This paper explores a cheaper and more effective alternative to address this problem by classifying if subjects have consumed too much alcohol by using accelerometer data from the subjects’ mobile devices while they perform physical activity. In order to create the most accurate classification system, we conduct experiments with five different supervised machine learning methods and use them on two features derived from accelerometer data of two different male subjects. We then share our experiment results that support why “Decision Tree Learning” is the supervised machine learning method that is best suited for our mobile device sobriety classification system.

Read more…