Current Detection Systems Research Articles

Sinusoidal Current Signal-Based Fire Detection System with Automatic Address Assignment

In this paper, a novel sinusoidal current signal-based fire detection system is proposed with automatic address assignment. The system model employs a conventional power line to embed fire information, i.e., the address, rather than using an additional communication line. At the transmitter, different frequencies of the sinusoidal current signal are combined and transmitted through a power line. At the receiver, fast Fourier transform (FFT) is applied to distinguish the frequency bins, which can represent the addresses of fire detectors. The proposed system model is implemented and the numerical results are presented in terms of measurements.

EGD-Net: Edge-guided and differential attention network for surface defect detection

Establishing an automated defect detection system is a critical task for industrial production, but the current defect detection system still faces great challenges, especially for defects with blurred edges and weak defects in complex backgrounds. To solve these problems, we propose an edge-guided and differential attention network (EGD-Net), which can highlight the defect areas by strengthening edge information and effectively eliminating background clutter. In the proposed network, the multi-scale features are first extracted. Then, a specially designed edge prediction module is used to extract defect edge information from shallow layers. Three multi-scale feature fusion modules are employed to fuse context information in the deep layers. Following this, the edge fusion module is constructed to complement the edge information and context information for better guiding defect segmentation. Finally, a differential attention module (DAM) is designed to perform top-down attention and produce the final prediction results. The DAM can effectively eliminate clutter in the background area caused by connecting the edge features. In the experiment, we collected a packaging box dataset with complex background patterns from the practical industrial field to verify the proposed model performance. Moreover, four public datasets were also employed to validate the model. Experimental results (mIoU/mPA) (DAGM2007: 85.53%/88.19%, CrackForest: 87.58%/91.32%, AITEX defect: 78.31%/82.29%, MT defect: 77.08%/81.19%, box defect 94.39%/96.51%:) show that our proposed method outperforms other state-of-the-art methods, especially for the detection of complex background defects.

A subject-independent SSVEP-based BCI target detection system based on fuzzy ordering of EEG task-related components

BCI systems are recognized as a means of interaction between the brain and the computer. In this research, a subject-independent BCI target detection system based on SSVEP was developed using the information in the task-related components and their fuzzy memberships to the regarding task. To assess the proposed system, an online available SSVEP data with four stimulus frequencies of 5, 6, 7 and 8 Hz from five participants were used. First, the task-related components of the SSVEP data were extracted using the TRCA algorithm. Then the membership of each task-related component was determined to each stimulus frequency, and the components were ordered fuzzily based on these membership values. Finally, a new concept named “Isolated Group” was defined and the final data label was decided by the identification of the isolated groups. Based on the results, observation of the detected isolated groups, showed that the isolated groups’ components were mostly related to the stimulus frequency, based on their FFT plots. The classification results obtained using Leave-One-Subject-Out method were 100 %, 79.17 %, 70.83 %, 70.83 % and 95.83 % (mean: 83.33 %) for five participants, respectively. In summary, the fuzzy ordering of task-related components was used to pick the components with the most important information for target detection. Although the detection of stimulus frequency in BCI applications has been widely studied, most current target detection systems require an intensive training process for every individual. In contrast, in this research‘s method, only a concise training process for determining criteria weights in terms of an optimization step, is needed.

Feature extraction for machine learning-based intrusion detection in IoT networks

A large number of network security breaches in IoT networks have demonstrated the unreliability of current Network Intrusion Detection Systems (NIDSs). Consequently, network interruptions and loss of sensitive data have occurred, which led to an active research area for improving NIDS technologies. In an analysis of related works, it was observed that most researchers aim to obtain better classification results by using a set of untried combinations of Feature Reduction (FR) and Machine Learning (ML) techniques on NIDS datasets. However, these datasets are different in feature sets, attack types, and network design. Therefore, this paper aims to discover whether these techniques can be generalised across various datasets. Six ML models are utilised: a Deep Feed Forward (DFF), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Decision Tree (DT), Logistic Regression (LR), and Naive Bayes (NB). The accuracy of three Feature Extraction (FE) algorithms is detected; Principal Component Analysis (PCA), Auto-encoder (AE), and Linear Discriminant Analysis (LDA), are evaluated using three benchmark datasets: UNSW-NB15, ToN-IoT and CSE-CIC-IDS2018. Although PCA and AE algorithms have been widely used, the determination of their optimal number of extracted dimensions has been overlooked. The results indicate that no clear FE method or ML model can achieve the best scores for all datasets. The optimal number of extracted dimensions has been identified for each dataset, and LDA degrades the performance of the ML models on two datasets. The variance is used to analyse the extracted dimensions of LDA and PCA. Finally, this paper concludes that the choice of datasets significantly alters the performance of the applied techniques. We believe that a universal (benchmark) feature set is needed to facilitate further advancement and progress of research in this field.

Deep and transfer learning for building occupancy detection: A review and comparative analysis

The building internet of things (BIoT) is quite a promising concept for curtailing energy consumption, reducing costs, and promoting building transformation. Besides, integrating artificial intelligence (AI) into the BIoT is essential for data analysis and intelligent decision-making. Thus, data-driven approaches to infer occupancy patterns usage are gaining growing interest in BIoT applications. Typically, analyzing big occupancy data gathered by BIoT networks helps significantly identify the causes of wasted energy and recommend corrective actions. Within this context, building occupancy data aids in the improvement of the efficacy of energy management systems, allowing the reduction of energy consumption while maintaining occupant comfort. Occupancy data might be collected using a variety of devices. Among those devices are optical/thermal cameras, smart meters, environmental sensors such as carbon dioxide (CO2), and passive infrared (PIR). Even though the latter methods are less precise, they have generated considerable attention owing to their inexpensive cost and low invasive nature. This article provides an in-depth survey of the strategies used to analyze sensor data and determine occupancy. The article’s primary emphasis is on reviewing deep learning (DL), and transfer learning (TL) approaches for occupancy detection. This work investigates occupancy detection methods to develop an efficient system for processing sensor data while providing accurate occupancy information. Moreover, the paper conducted a comparative study of the readily available algorithms for occupancy detection to determine the optimal method in regards to training time and testing accuracy. The main concerns affecting the current occupancy detection system in terms of privacy and precision were thoroughly discussed. For occupancy detection, several directions were provided to avoid or reduce privacy problems by employing forthcoming technologies such as edge devices, Federated learning, and Blockchain-based IoT.

A Deep Convolutional Neural Network Stacked Ensemble for Malware Threat Classification in Internet of Things

Malicious attacks to software applications are on the rise as more people use Internet of things (IoT) devices and high-speed internet. When a software system crash happens caused by malicious action, a malware imaging method can examine the application. In this study, we present a novel malware classification method that captures suspected operations in a variety of discrete size image features, allowing us to identify such IoT device malware families. To decrease deep neural network training time, essential local and global image features are selected using a combined local and global feature descriptor (LBP-GLCM). The classification performance of the proposed deep learning model is improved by combining the predictions of weak learners (CNNs) and using them as knowledge input to a multi-layer perceptron meta learner. This is a neural network ensemble with stacked generalization that is used to improve network generalization ability. The public dataset used for performance evaluation contains 5472 samples from 11 different malware families. In order to compare the proposed methodology to current malware detection systems, we developed a baseline experiment. The proposed approach improved malware classification results to 98.5% accuracy and 98.4% accuracy when using [Formula: see text] and [Formula: see text] image sizes, respectively. Overall, the results showed that the stacked generalization ensemble with multi-step extracting features is a more effective method for classification performance and response time.

Design of a hydrodynamic cavitation system for the extraction and detection of Escherichia coli (O157:H7) from ground beef

Pathogen detection in food is essential for mitigating a community's adverse health and economic impacts. Pathogen extraction and release from food matrices is a bottleneck for current detection systems. Homogenization via stomaching is the best available method for pathogen extraction from meat. However, the recovery of pathogens is limited. This paper presents the benefit of adding seven minutes of cavitation to the stand-alone stomaching method for the homogenization of ground beef and subsequent downstream detection of E. coli O157:H7 in spiked ground beef. To the best of our knowledge, this is the first instance of the utilization of stomaching followed by cavitation for the extraction of pathogens from food matrices. In this study, the homogenization of ground beef is characterized using particle size measurements. Results show that homogenization of the ground beef increases as cavitation intensity increases, leading to an enhanced release of E. coli O157:H7 compared to stand-alone stomaching. However, much higher intensity cavitation (cavitation number = 0.84) would damage E. coli O157:H7, leading to a 20.5% reduction in the viable E. coli O157:H7, adversely affecting downstream detection. Therefore, cavitation parameters were optimized to maximize the release of E. coli O157:H7 from ground beef and concomitantly minimize the damage to the pathogen. The addition of seven-minute cavitation to stomaching was 74 % more effective than the current gold standard, stand-alone stomaching for detecting E. coli O157:H7 from ground beef. The work presented here has substantial implications for extracting various pathogens from different types of food.

Vehicle Over Speeding Detection

Our proposed project aims to develop a system that detects cars driving at speeds over specified limit and inform concerned authorities immediately. Road accidents occurrences have increased recently so there needs to be a system that allows to detect over-speeding cars. Current speed detection systems are handheld guns held by police personnel that allow them to check car speed and then manually inform authorities about the vehicle. Whereas this proposed system does not need any human interception and records car speed as well as wirelessly informs authorities of over-speeding detections. The system first calculates the time required by the specific car for moving from first point to the second. Based on this data it calculates the car speed. This data is gathered and then transmitted by the system wirelessly to concerned authorities at a remote location. The mechanism consists of IT transmitter- receiver pair that work in combination for vehicle detection purpose. The microcontroller is now used to process this data and calculate the time required by vehicle to travel from one point to the other. Depending upon this time it now calculates vehicle speed as well as displays this on an LCD display. The system also sends this data wirelessly. It sounds a buzzer alarm if an over speed vehicle is detected. Over speeding of vehicles is the major cause of accidents in recent times. Monitoring of such over speeding vehicles especially on highways is of prime importance. Many manual methods are being used to detect the over speeding vehicles by traffic control team. However, these methods require lot of man power and continuous monitoring by traffic personnel. In this study, an attempt is made to develop an automatic speed monitoring system, which provides a simple way to monitoring speeds of all the vehicles from a centralized control room. This system calculates the instantaneous speed of vehicle with help of sensors and the over speeding vehicle is detected using an image processing technique using Python programming language. The developed model is validated with real world traffic data and comparative analysis of speeds obtained by manual method and developed model shows that model truly represents the field condition.

Abnormal Data Monitoring and Analysis Based on Data Mining and Neural Network

In order to solve the problems of low efficiency, high consumption of human and time resources, and low degree of intelligence in the current financial abnormal data detection system in computerized accounting, this paper proposes a financial abnormal data monitoring and analysis algorithm based on data mining and neural network. The analysis algorithm uses the method of data mining to process the original financial data, remove invalid information, retain valuable information, and standardize the data to solve the problem of large labor and time consumption. Then, neural network-related algorithms are used to identify the anomalies of standardized data, so as to realize the intelligent early warning of financial abnormal data. Compared with the financial audit algorithm in traditional accounting computerization, this algorithm has the advantages of high efficiency, low energy consumption, and high intelligence. The test results show that the classification accuracy of the proposed algorithm for abnormal data can reach more than 90%. It is proved that the algorithm is effective and improves the efficiency at the same time. The classification error rate of the classifier designed in this paper is 22.5%, and the accuracy rate is 77.5%. Both estimated and actual values represent the number of times, and there is no physical unit. The experiment shows that the main reason for the error rate is the delay of the inspection results of financial abnormalities. Through the example analysis, it can be concluded that the proposed intelligent analysis method of financial abnormal data based on deep learning has good effectiveness and accuracy and has a certain practical value.

VEHICLE SLOT DETECTION AND SMART PARKING SYSTEM USING MACHINE LEARNING

Parking space detection is a major challenge in our cities and drivers waste time when moving from one place to another in search of a free parking space. The current parking space detection systems available are based on sensors which are costly to install and maintain. The sensor systems cannot also be used outdoor environments such as in cities as the sensors can be stolen or vandalized. This study compared the performance of M- RCNN and YOLO algorithm which are a deep learning algorithm used to classify images. YOLO was seen to be the best to use in this study because it was able to run under low comping resources and give accurate predictions. It was thus used to develop a prototype that was used for detecting the status of parking slots as either empty or occupied. The solution was verified by feeding it with a parking area video stream that had vehicles coming and leaving the parking area and monitoring how well its able to identify the vehicle objects from other objects and how well it is able to predict the status of a parking area as either vacant or occupied Keywords- Parking Slot, Deep Learning, Automated Parking, CNN, Mask R-CNN, YOLO, Image Processing

Optimization design of excitation coil for pulsed eddy current testing of laser welding defect structure of power battery bus-bar

Pulsed eddy current testing is suitable for non-contact detection of conductor structural defect, which is an effective method to realize online detection of welding structural defect of power battery pack. Aiming at the circular laser weld structure between the bus-bar and the terminal of the power battery pack, a three-dimensional finite element simulation model of pulsed eddy current testing is established. According to the structural parameters of the pulsed eddy current excitation coil, which are the inner diameter, outer diameter, height and number of turns, the pulsed eddy current voltage responses of the laser welding defect structure of the battery pack bus-bar are calculated, respectively. By analyzing the influence of the excitation coil parameters on the pulsed eddy current detection results, the pulsed eddy current excitation coil parameters are determined for the testing of laser welding structural defects. It provides a method and basis for the optimal design of the pulsed eddy current detection system for the laser weld structural defect of the battery pack bus-bar.

Arsenic ion assisted core–satellites nano-assembly of gold nanoparticles for its colorimetric determination in water

Arsenic contamination in water has threatened lives globally. The capability of arsenic to interact with the sulfhydryl group of amino acids in human body is considered as one of the factors leading to its increased toxicity and there is the dire need for development of better detection platforms. In this study, advantage of natural biomimetic interaction between thiol rich ligands and arsenic ions along with the effect of nanoarchitecture was evaluated by fabricating a heterostructural bioreceptor probe which interacts with arsenic based on affinity interaction to bring about a visual colorimetric response via plasmon coupling. As (III) ions can bind to these selective ligands and self-assemble to form gold nanoparticle networks. The mesh-like nanoarchitecture enhances the available surface area for interaction between selective bioreceptor heads and As (III) ions. The developed chemoprobe was validated by assessing the spiked real water samples with recoveries of 95–103%. The results indicate that the chemical nanoprobe was responsive over a wide linear range of 1 to 200 μg L−1, offering the detection limit of 0.22 μg L−1 and selective for As (III) ions in a presence of 10 times more concentrated solution of the competing ions which indicates at the potential of nano-architecture in improving current detection systems.

Review on Phishing Attack Detection using Recurrent Neural Network

Phishing is a crime that involves the theft of personal information from users. Individuals, corporations, cloud storage, and government websites are all targets for the phishing websites. Anti-phishing technologies based on hardware are commonly utilised, while software-based options are preferred due to cost and operational considerations. Current phishing detection systems have no solution for problems like zero-day phishing assaults. To address these issues, a three-phase attack detection system called the Phishing Attack Detector based on Web Crawler was suggested, which uses a recurrent neural network to precisely detect phishing incidents. Based on the classification of phishing and non-phishing pages, it covers the input features Web traffic, web content, and Uniform Resource Locator (URL).

Evanescent scattering imaging of single protein binding kinetics and DNA conformation changes

Evanescent illumination has been widely used to detect single biological macromolecules because it can notably enhance light-analyte interaction. However, the current evanescent single-molecule detection system usually requires specially designed microspheres or nanomaterials. Here we show that single protein detection and imaging can be realized on a plain glass surface by imaging the interference between the evanescent lights scattered by the single proteins and by the natural roughness of the cover glass. This allows us to quantify the sizes of single proteins, characterize the protein–antibody interactions at the single-molecule level, and analyze the heterogeneity of single protein binding behaviors. In addition, owing to the exponential distribution of evanescent field intensity, the evanescent imaging system can track the analyte axial movement with high resolution, which can be used to analyze the DNA conformation changes, providing one solution for detecting small molecules, such as microRNA. This work demonstrates a label-free single protein imaging method with ordinary consumables and may pave a road for detecting small biological molecules.

Scalable Malware Detection System Using Distributed Deep Learning

The number of complex and novel malware attacks is increasing exponentially in the cyberworld. Malware detection systems are facing new challenges due to the volume, velocity, and complexity of malware. The current malware detection system relies on a time-consuming, resource-intensive, and knowledge-intensive classification approach. Most of the existing malware detection system is ineffective in detecting novel malware attacks. A deep learning approach can be used to build a malware detection system that can effectively detect novel malware attacks without much human intervention. The current circumstance necessitates not just a malware system with excellent accuracy, but also one that can serve a large volume of demand in near real-time. A scalable malware detection system capable of detecting complex attacks is the need of time. This article discusses a scalable and distributed deep learning approach for malware detection using convolutional neural network and bidirectional long short-term memory (CNN-BiLSTM). The deep learning approach has been used to make the system learn and make predictive decisions without human intervention. The performance of the deep learning approach depends on various parameters and training data sets. Hence, different combinations of deep learning algorithms have been used to design and test the models to achieve the desired result. The experimental results show that the double layer of CNN and BiLSTM has better performance than single-layer CNN.

BotCap: Machine Learning Approach for Botnet Detection Based on Statistical Features

In this paper, we describe a detailed approach to develop a botnet detection system using machine learning (ML)techniques. Detecting botnet member hosts, or identifying botnet traffic has been the main subject of manyresearch efforts. This research aims to overcome two serious limitations of current botnet detection systems:First, the need for Deep Packet Inspection-DPI and the need to collect traffic from several infected hosts. Toachieve that, we have analyzed several botware samples of known botnets. Based on this analysis, we haveidentified a set of statistical features that may help to distinguish between benign and botnet malicious traffic.Then, we have carried several machine learning experiments in order to test the suitability of ML techniques andalso to pick a minimal subset of the identified features that provide best detection. We have implemented ourapproach in a tool called BotCap whose test results showed its proven ability to detect individually infected hostsin a local network.

A Novel Psychotherapy Effect Detector of Public Art Based on ResNet and EEG Imaging.

Methods The EEG emotion dataset seed is used for feature extraction with DE, and the emotion is recognized by ResNet. Adam optimizer is used to classify the extracted DE through ResNet50 model. Each batch is set as 5 groups of data and is trained for 50 rounds, then the model is optimized, and the accuracy rate is 76.47%, which output the probability of good emotion through the model. We put the model optimized by ResNet into the intelligent module and visualize it with numerical value. Results The detector designed by EEG data and ResNet50 optimization model has high accuracy. The results show that the error between the detector data and the questionnaire interview data is small, the average error is 2.77, and the accuracy is 97%. The closer the subject's emotion before the test is to neutral emotion, the closer the questionnaire result is to the test result of the tester, and the smaller the error is. The difference between the tester data and the survey questionnaire data is 0.2, which is in good agreement and has small error. It can be seen that the detector has high accuracy. Conclusion Our proposed public art psychotherapy effect detector has good accuracy in detecting users' emotions. It can detect the group psychotherapy effect of public art and can classify and screen a large number of public arts in the city by quantitative methods. It provides support for further summarizing the practical utility of public art and provides a new way for the optimal design and follow-up evaluation of public art design.

Proprioceptive Slip Detection for Planetary Rovers in Perceptually Degraded Extraterrestrial Environments

Slip detection is of fundamental importance for the safety and efficiency of rovers driving on the surface of extraterrestrial bodies. Current planetary rover slip detection systems rely on visual perception on the assumption that sufficient visual features can be acquired in the environment. However, visual-based methods are prone to suffer in perceptually degraded planetary environments with dominant low terrain features such as regolith, glacial terrain, salt evaporites, and poor lighting conditions such as dark caves and permanently shadowed regions. Relying only on visual sensors for slip detection also requires additional computational power and reduces the rover traversal rate. This paper answers the question of how to detect wheel slippage of a planetary rover without depending on visual perception. In this respect, we propose a slip detection system that obtains its information from a proprioceptive localization framework that is capable of providing reliable, continuous, and computationally efficient state estimation over hundreds of meters. This is accomplished by using zero velocity update, zero angular rate update, and non-holonomic constraints as pseudo-measurement updates on an inertial navigation system framework. The proposed method is evaluated on actual hardware and field tested in a planetary-analog environment. The method achieves greater than 92% slip detection accuracy for distances around 150 m using only an inertial measurement unit and wheel encoders.

A Semi-Supervised Learning Approach for Tackling Twitter Spam Drift

Abstract: Twitter play an important role in accelerating the spread of spam. In order to protect the users, Twitter and the research community have been developing different spam detection systems by applying different machine-learning techniques. However, a recent study showed that the current machine learning-based detection systems are not able to detect spam accurately because spam tweet characteristics vary over time. This issue is called "Twitter Spam Drift". In the proposed system a semi-supervised learning approach (SSLA) has been proposed to tackle this. The new approach uses the unlabeled data to learn the structure of the domain. To handle the drift, live twitter stream of data is taken for the study. The pre-processing of livedownloaded data is labeled and machine learning is applied to detect spam and non-spam users. The data is stored in cloud storage, which can be accessed by user from anywhere. Experimental results were conducted on more than one machine learning algorithm and finds the better for the proposed problem, in-terms of accuracy

Review Study on Development of Fire Detection Technologies

Progresses in identifiers, microelectronics, as well as data innovation, as well as a more noteworthy comprehension of fire physical science, have brought about critical increases in fire identification innovation throughout the most recent ten years. Fire identification innovation actually faces difficulties like diminishing misleading problems, expanding responsiveness with dynamic reaction, and for exceptionally costly and complex establishments to more readily safeguard the general population and consent to evolving regulation. Offering shields. The motivation behind this article is to address ebb and flow innovative work in fire identification innovation, for example, enhancements in sensor frameworks, fire information handling, and screen innovation, including incorporated fire recognition frameworks. This paper sees ongoing advances in fire identification innovation, like arising sensor advances, sign and observing advances as well as consolidated fire recognition frameworks. Some of the concerns with current fire detection systems, as well as potential research efforts, are discussed.