Artificial Neural Network Classifier Research Articles

Tracking of moving human in different overlapping cameras using Kalman filter optimized

Tracking objects is a crucial problem in image processing and machine vision, involving the representation of position changes of an object and following it in a sequence of video images. Though it has a history in military applications, tracking has become increasingly important since the 1980s due to its wide-ranging applications in different areas. This study focuses on tracking moving objects with human identity and identifying individuals through their appearance, using an Artificial Neural Network (ANN) classification algorithm. The Kalman filter is an important tool in this process, as it can predict the movement trajectory and estimate the position of moving objects. The tracking error is reduced by weighting the filter using a fuzzy logic algorithm for each moving human. After tracking people, they are identified using the features extracted from the histogram of images by ANN. However, there are various challenges in implementing this method, which can be addressed by using Genetic Algorithm (GA) for feature selection. The simulations in this study aim to evaluate the convergence rate and estimation error of the filter. The results show that the proposed method achieves better results than other similar methods in tracking position in three different datasets. Moreover, the proposed method performs 8% better on average than other similar algorithms in night vision, cloud vision, and daylight vision situations.

Improvement of the classical artificial neural network simulation model of the parabolic trough solar collector outlet temperature and thermal efficiency using the conformable activation functions

The outlet temperature and the thermal efficiency predictions of the PTSC are essential parameters in the solar thermal power system. Therefore, it is crucial to have a prediction model that can predict its spatiotemporal behavior to the greatest extent possible. For that, the present study provides improved models of the classical ANN model by using ELU, swish, softplus, logsig, and tansig functions through the exponential transfer function to improve the outlet temperature prediction performance and thermal efficiency of PTSC. The PTSC’s outlet temperature simulation showed that the ANN model of topology 6-2-1 with the conformable swish transfer function (cswish) was the best model among nine other studied models, with an R2 and Radj2 of 0.9988, and MAE of 0.0052. The simulation of PTSC’s thermal efficiency proved that the conformable functions logsig, tansig, ELU and the classical ELU function involved in the ANN model for 6-3-1 achieved better results with approximately 95% precision. With the non-integer activation functions, was reduced a neuron in the hidden layer, which leads to the simplicity of the prediction model compared to the use of the classical activation functions. The proposed activation functions do not require high-performance features according to their execution times. Therefore, it could be a good alternative for engineers to apply in other thermal energy systems to make their prediction models more accessible and practical in the control and optimization process.

Mapping sugarcane residue burnt areas in smallholder farming systems using machine learning approaches

Satellite remote sensing methods have been proven to quantify the burnt areas resulting from different fire activities reliably. However, they were reported to perform poorly to identify crop residue burnt areas, particularly in smallholder systems due to lack of frequent and high spatial resolution satellite data. In this study, we used Harmonized Landsat Sentinel-2 (HLS) observations and evaluated two machine learning classifiers (i.e., Support Vector Machines (SVM) and Artificial Neural Networks (ANN)) to map sugarcane burnt areas for the 2019–20 season in a smallholder farming region in Thailand.Results showed that both classifiers performed well in identifying the spatial patterns of sugarcane burnt areas in the region. The ANN outperformed SVM at both pixel and regional scales. At pixel level, ANN accuracy was 93.4 % while SVM's best-performing Polynomial kernel accuracy was 82.7 %. The ANN estimated average percent burnt area (51.1 %) in the region was closer to reported value (48.7 %) by Thailand's Office of Cane and Sugar Board (OCSB), compared to the SVM estimate (62.9 %). The total estimated burnt areas by ANN and SVM (315 and 418 thousand ha, respectively) deviated more from OCSB's area (240 thousand ha) compared to percent burnt area. However, area estimates from classifiers had significantly better accuracy than the estimates of MODIS burnt products.Overall, this study demonstrated that HLS observations provided required spectral information to build promising models to map burnt areas in smallholder systems with higher accuracy than global products. Our mapping algorithm using the ANN classifier showed the potential to monitor sugarcane burnt areas reliably, and contribute to the successful implementation of regulatory policies in Thailand.

Spatial Analysis of Soil Fertility Using Geostatistical Techniques And Artificial Neural Networks

Information on the spatial variation of soil fertility attributes is an essential input for precision agriculture and soil management decision making. In this study, soil fertility assessment was carried out through the spatial distribution of thematic maps of individual properties and the subsequent integration into a digital mapping model of local fertility classes, as fundamental bases for the implementation of fertilization and amendment plans adjusted to soil status and crop requirements. For the evaluation of fertility, a systematic surface sampling was carried out in 70 sites in the "Agronomy" production field of the National University of the Central Plains "Romulo Gallegos", El Castrero sector, Juan German Roscio municipality, Guárico state, Venezuela. Ten soil variables were analyzed: pH (1:2.5), electrical conductivity (1:5), organic matter, available phosphorus, assimilable potassium, available calcium and magnesium, and the relative amounts of sand, silt and clay. Soil property maps were produced by geostatistical analysis and interpolation by ordinary kriging, and artificial intelligence techniques based on an artificial neural network classification system were applied to generate soil fertility classes using the Fuzzy Kohonen Clustering Network (FKCN) algorithm by interpolating the values of the membership function for each of the classes. The reliability of the individual maps of each soil variable was obtained by cross validation with a reliability level higher than 90%, with the exception of the variables % Clay and % Silt that presented a reliability higher than 85%. The integration of the soil attribute maps and the combination of the values of belonging to each class produced a map integrated by five soil fertility categories. The final model of digital soil fertility classes presented a reliability equivalent to 86%, which indicated a high degree of homogeneity within the soil classes obtained for fertility purposes.

In situ or accreted? Using deep learning to infer the origin of extragalactic globular clusters from observables

ABSTRACT Globular clusters (GCs) are powerful tracers of the galaxy assembly process, and have already been used to obtain a detailed picture of the progenitors of the Milky Way (MW). Using the E-MOSAICS cosmological simulation of a (34.4 Mpc)3 volume that follows the formation and co-evolution of galaxies and their star cluster populations, we develop a method to link the origin of GCs to their observable properties. We capture this complex link using a supervised deep learning algorithm trained on the simulations, and predict the origin of individual GCs (whether they formed in the main progenitor or were accreted from satellites) based solely on extragalactic observables. An artificial neural network classifier trained on ∼50 000 GCs hosted by ∼700 simulated galaxies successfully predicts the origin of GCs in the test set with a mean accuracy of 89 per cent for the objects with $\rm [Fe/H]\lt -0.5$ that have unambiguous classifications. The network relies mostly on the alpha-element abundances, metallicities, projected positions, and projected angular momenta of the clusters to predict their origin. A real-world test using the known progenitor associations of the MW GCs achieves up to 90 per cent accuracy, and successfully identifies as accreted most of the GCs in the inner Galaxy associated to the Kraken progenitor, as well as all the Gaia-Enceladus GCs. We demonstrate that the model is robust to observational uncertainties, and develop a method to predict the classification accuracy across observed galaxies. The classifier can be optimized for available observables (e.g. to improve the accuracy by including GC ages), making it a valuable tool to reconstruct the assembly histories of galaxies in upcoming wide-field surveys.

Extracting relations from texts using vector language models and a neural network classifier.

The article investigates the possibility of identifying the presence of SKOS (Simple Knowledge Organization System) relations between concepts represented by terms on the base of their vector representation in general natural language models. Several language models of the Word2Vec and GloVe families are considered, on the basis of which an artificial neural network (ANN) classifier of SKOS relations is formed. To train and test the efficiency of the classifier, datasets formed on the basis of the DBPedia and EuroVoc thesauri are used. The experiments performed have shown the high efficiency of the classifier trained using GloVe family models, while training it with use of Word2Vec models looks impossible in the bounds of considered ANN-based classifier architecture. Based on the results, a conclusion is made about the key role of taking into account the global context of the use of terms in the text for the possibility of identifying SKOS relations.

Artificial Neural Network for Classification of Dengue Fever Using Backpropagation Algorithm

Fever is an increase in body temperature to higher than usual. Normal human body temperature is at 37oC, if the body temperature is more than this figure, it indicates a fever caused by infectious or non-infectious factors. The main symptom of Dengue hemorrhagic fever is high fever with a temperature between 30oC - 40oC which appears suddenly, the fever lasts for 7 days and occurs continuously, body temperature can be normal or low, then will rise slowly every day and can reach 40oC . These two diseases are still a public health problem in urban areas, including in the cities of Binjai and Medan. The problem that has occurred so far is that people in general cannot differentiate the symptoms of Dengue Fever from Malaria, so the treatment given only provides ordinary fever medicine, so that within three days there is no change and the high body temperature makes the patient know that someone has dengue fever. Therefore, the solution provided in this research is to find out the physical characteristics experienced by the sufferer before further diagnosis is carried out. If someone has a fever above 38oC, the body has red spots, irregular breathing, immediately go to the doctor because these symptoms indicate symptoms of dengue hemorrhagic fever or malaria fever. Artificial neural networks are an information processing system designed to imitate the workings of the human brain by carrying out a learning process through changing the weights of synapses. The human brain consists of millions of interconnected neurons known as biological neurons. Each neuron consists of a cell that has a number of dendrites (input) and an axon (output). Axons connect to other neurons through connecting pathways that produce chemical reactions when responding to incoming input. The input required includes the number of input variables, input variable values, weights, learning rate, threshold, maximum epoh and target (output) with the error value classification used is Mean Absolute Error (MAE), there are 2 types of disease with fever symptoms used. The types of disease are dengue hemorrhagic fever and malaria and the system will be designed using the Visual Basic 2010 programming language. From the results of the research that has been carried out, classification results are obtained with a value of 0.893619481 or rounded to equal 1 and classified as dengue hemorrhagic fever.

Effective uncertain fault diagnosis technique for wind conversion systems using improved ensemble learning algorithm

This paper introduces a pioneering fault diagnosis technique termed Interval Ensemble Learning based on Sine Cosine Optimization Algorithm (IEL- SCOA), tailored to tackle uncertainties prevalent in wind energy conversion (WEC) systems. The approach unfolds in three integral phases. Firstly, the establishment of interval centers and ranges, employing upper and lower bounds, effectively manages the inherent uncertainties arising from noise and measurement errors intrinsic to the wind system. Subsequently, the dataset undergoes processing via the Sine-Cosine Optimization Algorithm (SCOA), enabling the extraction of the most pertinent attributes. The culmination of predictive precision and classification performance is achieved through the integration of the refined dataset into an ensemble learning paradigm, harmonizing bagging, boosting techniques, and an artificial neural network classifier. The principal aim of the IEL-SCOA approach is to discern the spectrum of operational conditions within WEC systems, encompassing a healthy mode alongside six distinct faulty modes. These anomalies, encompassing short circuits, open circuits, and wear-out incidents, are deliberately induced at diverse locations and facets of the system, notably the generator and grid sides. Empirical results underscore the robustness and efficiency of the proposed methodology, showcasing an exceptional accuracy rate of 99.76 %. These outcomes definitively establish the IEL-SCOA approach as a potent and efficacious tool for precise fault diagnosis in uncertain WEC systems.

Revealing the diagnostic value and immune infiltration of senescence-related genes in endometriosis: a combined single-cell and machine learning analysis.

Introduction: Endometriosis is a prevalent and recurrent medical condition associated with symptoms such as pelvic discomfort, dysmenorrhea, and reproductive challenges. Furthermore, it has the potential to progress into a malignant state, significantly impacting the quality of life for affected individuals. Despite its significance, there is currently a lack of precise and non-invasive diagnostic techniques for this condition. Methods: In this study, we leveraged microarray datasets and employed a multifaceted approach. We conducted differential gene analysis, implemented weighted gene co-expression network analysis (WGCNA), and utilized machine learning algorithms, including random forest, support vector machine, and LASSO analysis, to comprehensively explore senescence-related genes (SRGs) associated with endometriosis. Discussion: Our comprehensive analysis, which also encompassed profiling of immune cell infiltration and single-cell analysis, highlights the therapeutic potential of this gene assemblage as promising targets for alleviating endometriosis. Furthermore, the integration of these biomarkers into diagnostic protocols promises to enhance diagnostic precision, offering a more effective diagnostic journey for future endometriosis patients in clinical settings. Results: Our meticulous investigation led to the identification of a cluster of genes, namely BAK1, LMNA, and FLT1, which emerged as potential discerning biomarkers for endometriosis. These biomarkers were subsequently utilized to construct an artificial neural network classifier model and were graphically represented in the form of a Nomogram.

State of temperature detection of Li-ion batteries by intelligent gray box model

Li-Ion batteries are among the key enablers of more sustainable use of energy. However, they need to be supervised and undergo continuous maintenance to assure safety and longevity. This paper focuses on the sensorless detection of the State of Temperature (SOT) of the Li-ion batteries during the operational life cycle of the battery irrespective of its state of charge. The paper presents the new Intelligent Gray Box Model (IGBM) to detect the SOT of Li-ion cells: that combines the three most powerful diagnostic tools Electrochemical Impedance Spectroscopy (EIS), Equivalent Circuit Model (ECM), and Artificial Neural Network Classifier (NNC). The work introduces the experimental test bench capable of emulating real-world and embedded constraints to conduct EIS onboard, its data preprocessing, and useful information extraction for the entire frequency spectrum. Furthermore, this paper presents a new hybrid parameter identification that combines the Whale Optimization Algorithm (WOA) and Levenberg Marquardt algorithm (LM) to identify the fractional order ECM. Finally, a neural network classifier is designed, optimized, and compared with different feature scaling techniques to evaluate its accuracy and robustness to detect and classify exact battery temperatures in real time from experimental data.

Reliable detection of blast disease in rice plant using optimized artificial neural network

Early plant disease diagnosis is necessary for several purposes, including reducing yield losses, monitoring and predicting infections, detecting host resistance, and studying basic host–pathogen biological processes. However, early detection has been limited by a trained workforce and the ability to identify the problem early in the growing season. Artificial intelligence/machine learning algorithms can help fill this gap. Automatic leaf detection using machine learning is proposed in this study. The presented approach consists of three stages: pre‐processing, feature extraction, and classification. Initially, the input image is transformed into the red, green, and blue formatand the noise in the green band is removed using a median filter. Then important features of the green band are extracted. After feature extraction, the extracted features are fed to the optimized artificial neural network classifier to classify an image as normal or diseased. To improve artificial neural network (ANN) performance, the ANN parameters are chosen optimally using the adaptive sunflower optimization (ASFO) algorithm. Then, the infected region is separated using a level set segmentation algorithm. The efficiency of our work is analyzed based on accuracy, sensitivity, and specificity; the proposed method reached the maximum accuracy of 97.94% for plant disease prediction.

Monitoring the impacts of crop residue cover on agricultural productivity and soil chemical and physical characteristics

To the best of our knowledge, the impacts of crop residue cover (CRC) on agricultural productivity and soil fertility have not been studied by previous researchers. In this regard, this study aims to apply an integrated approach of remote sensing and geospatial analysis to detect CRC and monitor the effects of CRC on agricultural productivity, as well as soil chemical and physical characteristics. To achieve this, a series of Landsat images and 275 ground control points (GCPs) collected from the study areas for the years 2013, 2015, and 2021 were used. A convolutional neural network (CNN), a class of artificial neural network has commonly applied to analyze visual imagery, was employed in this study for CRC detection in two classes (Not-CRC and CRC) for the years 2013, 2015, and 2021. To assess the effects of CRC, the Normalized Difference Vegetation Index (NDVI) was applied to Landsat image series for the years 2015 (22 images), 2019 (20 images), and 2022 (23 images). Furthermore, this study evaluates the impacts of CRC on soil fertility based on collected field observation data. The results show a high performance (Accuracy of > 0.95) of the CNN for CRC detection and mapping. The findings also reveal positive effects of CRC on agricultural productivity, indicating an increase in vegetation density by about 0.1909 and 0.1377 for study areas 1 and 2, respectively, from 2015 to 2022. The results also indicate an increase in soil chemical and physical characteristics, including EC, PH, Na, Mg, HCO3, K, silt, sand, and clay from 2015 to 2022, based on physical examination. In general, the findings underscore that the value of an integrated approach of remote sensing and geospatial analysis for detecting CRC and monitoring its impacts on agricultural productivity and soil fertility. This research can offer valuable insight to researchers and decision-makers in the field of soil science, land management and agriculture.

Dissipative quantum many-body dynamics in (1+1)D quantum cellular automata and quantum neural networks

Classical artificial neural networks, built from elementary units, possess enormous expressive power. Here we investigate a quantum neural network (QNN) architecture, which follows a similar paradigm. It is structurally equivalent to so-called (1+1)D quantum cellular automata, which are two-dimensional quantum lattice systems on which dynamics takes place in discrete time. Information transfer between consecutive time slices—or adjacent network layers—is governed by local quantum gates, which can be regarded as the quantum counterpart of the classical elementary units. Along the time-direction an effective dissipative evolution emerges on the level of the reduced state, and the nature of this dynamics is dictated by the structure of the elementary gates. We show how to construct the local unitary gates to yield a desired many-body dynamics, which in certain parameter regimes is governed by a Lindblad master equation. We study this for small system sizes through numerical simulations and demonstrate how collective effects within the quantum cellular automaton can be controlled parametrically. Our study constitutes a step towards the utilization of large-scale emergent phenomena in large QNNs for machine learning purposes.

Determination of moisture content and contaminated blank dried figs (Ficus carica L.) using dielectric property and artificial neural network

AbstractDried figs are a garden produce that must be graded after harvesting. Moisture levels and contaminated blank are two of the most critical effective elements on the marketability of dried figs, and they are highly related to fig quality. In the present research, an intelligent system was employed to classify dried figs based on moisture content levels and infected blank fruits. Capacitance characteristics, average diameter, and fruit area were all taken into account in this study. The dried fig dielectric constant was measured at six different frequency levels: 12, 22, 32, 42, 52, and 62 MHz. The best frequency was then chosen using the improved distance evaluation feature selection approach. Image processing was also used to determine the average diameter and area of the figures. Following that, the dielectric constant of the most effective frequency, the average diameter, and the area of the fruit were used as input parameters in the artificial neural network classification model to classify and describe the moisture and porosity level of the dried fig. The most essential dielectric constant information relating to moisture and porosity level was at frequencies of 22 and 52 MHz, respectively. Finally, classification accuracy of 95.7% for moisture and 91.3% for porosity level was attained. The results demonstrated the excellent performance and capabilities of the proposed approach for rating the internal quality of dried figs.

Data Mining with Supervised Instance Selection Improves Artificial Neural Network Classification Accuracy

Data Mining with Supervised Instance Selection Improves Artificial Neural Network Classification Accuracy

Flexible triboelectric nanogenerators of Au-g-C3N4/ZnO hierarchical nanostructures for machine learning enabled body movement detection

Here we report the development of triboelectric nanogenerator (TENG) based self-powered human motion detector with chemically developed Au-g-C3N4/ZnO based nanocomposite on common cellulose paper platform. Compared to bare g-C3N4, the nanocomposite in the form of hierarchical morphology is found to exhibit higher output voltage owing to the contribution of Au and ZnO in increasing the dielectric constant and surface roughness. While generating power ∼3.5 μW cm−2 and sensitivity ∼3.3 V N−1, the flexible TENG, is also functional under common biomechanical stimuli to operate as human body movement sensor. When attached to human body, the flexible TENG is found to be sensitive towards body movement as well as the frequency of movement. Finally upon attaching multiple TENG devices to human body, the nature of body movement has been traced precisely using machine learning (ML) techniques. The execution of the learning algorithms like artificial neural network and random forest classifier on the data generated from these multiple sensors can yield an accuracy of 99% and 100% respectively to predict body movement with great deal of precision. The exhibition of superior sensitivity and ML based biomechanical motion recognition accuracy by the hierarchical structure based flexible TENG sensor are the prime novelties of the work.

Retracted: Artificial Neural Network in Classification of Multisource Remote Sensing Images

Retracted: Artificial Neural Network in Classification of Multisource Remote Sensing Images

An enhanced artificial neural network for hand gesture recognition using multi-modal features

ABSTRACT The development of automatic application fields enhanced the advancement in Hand Gesture Recognition (HGR), which enhances human-computer interaction. In particular, hand gestures find a valuable application in hand games, virtual reality, and gesture-based signalling systems. Thus, detecting and recognising hand gestures gain technical significance. Although existing models highlight remarkable recognition outcomes, handling dynamic gestures is a challenging task. Hence, in this research, an enhanced flashfly optimisation-based artificial neural network is developed for the effective identification of hand gestures, and to overcome the existing challenges, the enhanced flash fly optimisation-based ANN technique is proposed. The main objective of the optimisation strategy is to effectively tune the parameters involved in the artificial neural network (ANN) by the attained optimal solution. The optimised ANN classifier is well trained by the pre-processed statistical and the Residual neural Networks (ResNet) features for consistently identifying hand movements. The attained performance of the enhanced flash fly optimisation-based ANN classifier during the representation of hand gestures based on the precision, recall, and F1 score is 0.940 %, 0.791 %, and 0.927 % for the training percentages 50 and 90. Hence, it is illustrated that the proposed HGR model exceeds the competent technique, demonstrating its effectiveness.

Assessment of Soybean Lodging Using UAV Imagery and Machine Learning.

Plant lodging is one of the most essential phenotypes for soybean breeding programs. Soybean lodging is conventionally evaluated visually by breeders, which is time-consuming and subject to human errors. This study aimed to investigate the potential of unmanned aerial vehicle (UAV)-based imagery and machine learning in assessing the lodging conditions of soybean breeding lines. A UAV imaging system equipped with an RGB (red-green-blue) camera was used to collect the imagery data of 1266 four-row plots in a soybean breeding field at the reproductive stage. Soybean lodging scores were visually assessed by experienced breeders, and the scores were grouped into four classes, i.e., non-lodging, moderate lodging, high lodging, and severe lodging. UAV images were stitched to build orthomosaics, and soybean plots were segmented using a grid method. Twelve image features were extracted from the collected images to assess the lodging scores of each breeding line. Four models, i.e., extreme gradient boosting (XGBoost), random forest (RF), K-nearest neighbor (KNN) and artificial neural network (ANN), were evaluated to classify soybean lodging classes. Five data preprocessing methods were used to treat the imbalanced dataset to improve classification accuracy. Results indicate that the preprocessing method SMOTE-ENN consistently performs well for all four (XGBoost, RF, KNN, and ANN) classifiers, achieving the highest overall accuracy (OA), lowest misclassification, higher F1-score, and higher Kappa coefficient. This suggests that Synthetic Minority Oversampling-Edited Nearest Neighbor (SMOTE-ENN) may be a good preprocessing method for using unbalanced datasets and the classification task. Furthermore, an overall accuracy of 96% was obtained using the SMOTE-ENN dataset and ANN classifier. The study indicated that an imagery-based classification model could be implemented in a breeding program to differentiate soybean lodging phenotype and classify lodging scores effectively.

Speed and Accuracy Trade-off ANN/SVM Based Sleep Apnea Detection with FPGA Implementation

ABSTRACT During sleep, some people experience breathing difficulties, leading to a condition known as sleep apnoea, which can result in suffocation. This study focuses on detecting sleep apnoea using Artificial Neural Network (ANN) and Support Vector Machine (SVM) classifiers, incorporating discrete wavelet transform for feature extraction. The proposed method aims to denoise Electrocardiography (ECG) signals and extract four statistical features, focusing on the gamma-band of the ECG signal. By comparing accuracy and speed of the models with varying bit representations on Field Programmable Gate Array (FPGA) technology, the study achieved high accuracy (92% for ANN and 88% for SVM with 32 bits). Implementing the models on FPGA allows real-world applications like home-based sleep monitoring systems. Utilizing wavelet transform and statistical features hold potential to further improve the detection accuracy. Overall, this study showcases an effective approach for sleep apnoea detection.