Use Of Convolutional Neural Networks Research Articles

Comparative analysis of traditional and transfer learning algorithms for coal spoil classification via close-range imagery

ABSTRACT The characterisation of materials is a prerequisite for evaluating and predicting the stability of mining waste dumps. Over the past three decades, the BHP Mitsubishi Alliance Coal framework has been a cornerstone in Australian coal mines for characterising waste dump materials. However, its reliance on subjective human observations has introduced potential inaccuracies and subjectivity into the process. In response to these limitations, this study proposes an innovative approach to classify coal spoil attributes by remotely acquiring images through phones/tablets. Automated image-based classification relies on feature extraction and a substantial amount of data. Nevertheless, the inherent complexity of geological factors contributing to the formation of both rare and dominant materials leads to imbalanced data. Recognising the need for classification mechanisms to overcome these challenges in spoil classification, the study explores and compares the use of convolutional neural networks, hybrid deep learning, and traditional techniques. Among the sixteen models evaluated in this study, the ResNet18-support vector machine model emerges as a powerful tool in geotechnical characterisation. However, it is essential to address issues of interpretability and adaptability to diverse datasets. As this study evolves, the field of geotechnical characterisation of spoil can anticipate the development of more robust methods in the future.

Deep Learning for Solar Panels Defect Classification Using Data Augmentation Strategies

The inspection and maintenance of solar panels face significant challenges due to the dangers involved in checking for potential defects in photovoltaic panels. However, the advancement of studies to facilitate this verification is limited by the lack of image datasets for training algorithms that can perform this task. Based on this principle, the use of Convolutional Neural Networks (CNNs) for training and Data Augmentation for creating artificial data from real images is common. With that said, this work aims to explore configurations and models of Data Augmentation for the classification of defects in solar panels using CNNs. The proposed methodology consists of four experimental stages, where the application of Zoom, rotation, horizontal displacement, and vertical displacement transformations are evaluated, followed by filtering the best results and combining them to find the ideal classification model. The performance of the model proved to be most effective when using a 35-degree rotation for creating artificial images, thus achieving an 88% F1-Score and 87.64% accuracy.

Advanced Image Classification Using Convolutional Neural Networks

Deep learning is essential for computer vision and object detection. This project explores the use of Convolutional Neural Networks (CNNs) for image classification using the CIFAR-10 dataset, which is widely used and provides a robust benchmark for evaluating image classification models. The CNN model, comprising various convolutional, pooling, and fully connected layers, is trained on a portion of the dataset and tested on another portion to evaluate its performance. Techniques such as data augmentation, dropout, and specific activation functions are employed to enhance the model’s efficacy. The study details the CNN architecture and reports the results, including accuracy metrics, to demonstrate the model’s effectiveness.

Tree sequences as a general-purpose tool for population genetic inference.

As population genetics data increases in size new methods have been developed to store genetic information in efficient ways, such as tree sequences. These data structures are computationally and storage efficient, but are not interchangeable with existing data structures used for many population genetic inference methodologies such as the use of convolutional neural networks (CNNs) applied to population genetic alignments. To better utilize these new data structures we propose and implement a graph convolutional network (GCN) to directly learn from tree sequence topology and node data, allowing for the use of neural network applications without an intermediate step of converting tree sequences to population genetic alignment format. We then compare our approach to standard CNN approaches on a set of previously defined benchmarking tasks including recombination rate estimation, positive selection detection, introgression detection, and demographic model parameter inference. We show that tree sequences can be directly learned from using a GCN approach and can be used to perform well on these common population genetics inference tasks with accuracies roughly matching or even exceeding that of a CNN-based method. As tree sequences become more widely used in population genetics research we foresee developments and optimizations of this work to provide a foundation for population genetics inference moving forward.

Data-driven Smart Farming to Grade and Classify Tomatoes using CNN and FFNN for Agricultural Innovation

Identifying images poses a challenge in computer vision, but the use of deep learning methods has greatly enhanced the performance of image classification systems. In this research, Convolutional Neural Networks (CNN) and Feed Forward Neural Networks (FFNN) have been utilized for image classification. CNN is extremely effective in picture classification, which extracts relevant information from images using convolutional and pooling layers to minimize the dimensionality of the derived features, while FFNN algorithm is a classic neural network with fully linked layers. It can be used to further process the features extracted by CNN. The study makes use of CNN and FFNN models to train a huge dataset of tomato images to categorize them based on their type, ripeness, and damage status. CNN is found to be more effective in the case of tomato classification as compared to FFNN algorithm in all the use cases. The accuracy for classification of an image (tomato or not) using CNN is 95.83%, type classification using CNN is 81.52%, whereas using FFNN is 66.30%; ripeness grading for CNN is 92.86%, whereas for FFNN it is 57.14%; and damage status grading is 92.86% using CNN and 67.86% using FFNN. Therefore, it can be concluded that quality processing of tomatoes can be improved using CNN.

A solution for the automatic detection of expansion joints in dam stilling pools using underwater robots

This research addresses the critical challenges of detecting and measuring underwater concrete structure expansion joints (CSEJ), with a particular focus on dam stilling basin environments. A full set of algorithms for detecting expansion joints with underwater robots is presented by the research, enabling real-time and offline analysis along with visualization. These algorithms introduce the use of Convolutional Neural Networks (CNNs) for image feature extraction and segmentation in CSEJ detection, significantly improving accuracy and adaptability over traditional methods. An automated approach known as the Underwater Dam Crack Detection Network (UDCD Network) was introduced by this research. And advanced attention gates and hybrid dilated convolutions to enhance its detection capabilities are incorporated in this system. Additionally, a three-phase deep transfer learning strategy during training was utilized in this research, which significantly improves the model's performance. The research suggests an underwater image restoration algorithm that uses a dark channel prior and dynamic white balance correction to improve image clarity. Furthermore, unique underwater environmental challenges were tackled by this research, such as non-contact measurement of expansion joint widths using a view-geometry-based algorithm and comprehensive detection of dam wall surfaces. Experimental validation demonstrates that the advanced imaging technologies of the proposed underwater robots hold significant potential for autonomous detection of expansion joints in dam stilling basins.

АНАЛІЗ ЗАСТОСУВАННЯ ЗГОРТКОВИХ НЕЙРОННИХ МЕРЕЖ ДЛЯ ПІДВИЩЕННЯ НАДІЙНОСТІ У РОБОТІ ФЕС

This work is dedicated to the study of applying convolutional neural networks (CNN) for automatic recognition of photovoltaic panel conditions based on images. The focus is placed on the impact of CNN hyperparameter tuning, such as the number of layers, kernel size, learning rate, and activation function, on the accuracy of classi-fying various panel conditions, including physical damage, electrical defects, dust contamination, and clean pan-els. The use of CNNs is driven by their ability to automatically detect complex spatial patterns in images, which is crucial for accurately identifying different types of defects and anomalies on the panels. The main challenges include the variability in panel conditions and external factors, such as weather conditions, that affect data quali-ty. The influence of CNN hyperparameters on classification accuracy was analyzed, and their optimal values were determined to achieve high model accuracy. The relevance of the research lies in the growing role of CNNs in monitoring photovoltaic panel conditions, which enables timely defect detection and optimization of mainte-nance. The results demonstrate that proper hyperparameter tuning of CNNs significantly improves classification accuracy, contributing to increased efficiency and stability of photovoltaic stations. The paper emphasizes the importance of using neural networks for image analysis in the field of photovoltaic system maintenance, ensuring their effective integration into energy networks. The goal of the study is to improve the accuracy of photovoltaic panel condition classification by developing and tuning CNN models with optimal hyperparameters, enabling timely detection of panel defects and enhancing the efficiency of photovoltaic system maintenance.

Tree sequences as a general-purpose tool for population genetic inference.

As population genetics data increases in size new methods have been developed to store genetic information in efficient ways, such as tree sequences. These data structures are computationally and storage efficient, but are not interchangeable with existing data structures used for many population genetic inference methodologies such as the use of convolutional neural networks (CNNs) applied to population genetic alignments. To better utilize these new data structures we propose and implement a graph convolutional network (GCN) to directly learn from tree sequence topology and node data, allowing for the use of neural network applications without an intermediate step of converting tree sequences to population genetic alignment format. We then compare our approach to standard CNN approaches on a set of previously defined benchmarking tasks including recombination rate estimation, positive selection detection, introgression detection, and demographic model parameter inference. We show that tree sequences can be directly learned from using a GCN approach and can be used to perform well on these common population genetics inference tasks with accuracies roughly matching or even exceeding that of a CNN-based method. As tree sequences become more widely used in population genetics research we foresee developments and optimizations of this work to provide a foundation for population genetics inference moving forward.

Revealing the Relationship of Batik Motifs Using Convolutional Neural Network

This study explores the use of Convolutional Neural Network to identify and classify regional batik motifs, a significant aspect of Indonesian cultural heritage. The CNN model was optimized with Adam optimizer and used to extract distinctive features from the batik patterns. Subsequently, a hierarchical clustering method was employed to construct a relationship tree depicting the link between batik motifs based on their region. The research findings demonstrate that the CNN model effectively classifies batik motifs with an accuracy of up to 88%. The study provides insights into the intricate connections between regional batik designs and contributes to the preservation and understanding of Indonesia's cultural heritage.

Development of an Algorithm for Semantic Segmentation of Earth Remote Sensing Data to Determine Phytoplankton Populations

Introduction. Computer vision is widely used for semantic segmentation of Earth remote sensing (ERS) data. The method allows monitoring ecosystems, including aquatic ones. Algorithms that maintain the quality of semantic segmentation of ERS images are in demand, specifically, to identify areas with phytoplankton, where water blooms— the cause of suffocation — are possible. The objective of the study is to create an algorithm that processes satellite data as input information for the formation and checking of mathematical models of hydrodynamics, which are used to monitor the state of water bodies. Various algorithms for semantic segmentation are described in the literature. New research focuses on enhancing the reliability of recognition — often using neural networks. This approach is modified in the presented work. To develop the direction, a new set of information from open sources and synthetic data are proposed. They are aimed at improving the generalization ability of the model. For the first time, the contour area of the phytoplankton population is compared to the database — and thus the boundary conditions are formed for the implementation of mathematical models and the construction of boundary-adaptive grids.Materials and Methods. The set of remote sensing images was supplemented with the author's augmentation algorithm in Python. Computer vision segmented areas of phytoplankton populations in the images. The U-Net convolutional neural network (CNN) was trained on the basis of NVIDIA Tesla T4 computing accelerators.Results. To automate the detection of phytoplankton distribution areas, a computer vision algorithm based on the U-Net CNN was developed. The model was evaluated by the calculated values of the main quality metrics related to segmentation tasks. The following metric values were obtained: Precision = 0.89, Recall = 0.88, F1 = 0.87, Dice = 0.87, and IoU = 0.79. Graphical visualization of the results of CNN learning on the training and validation sets showed good quality of model learning. This is evidenced by small changes in the loss function at the end of training. The segmentation performed by the model turned out to be close to manual marking, which indicated the high quality of the proposed solution. The area of the segmented region of the phytoplankton population was calculated by the area of one pixel. The result obtained for the original image was 51202.5 (based on information about the number of pixels related to the bloom of blue-green algae). The corresponding result of the modeling was 51312.Discussion and Conclusion. The study expands theoretical and practical knowledge on the use of convolutional neural networks for semantic segmentation of space imagery data. Given the results of the work, it is possible to assess the potential for automating the process of semantic segmentation of remote sensing data to determine the boundaries of phytoplankton populations using artificial intelligence. The use of the proposed computer vision model to obtain contours of water bloom due to phytoplankton will provide for the creation of databases — the basis for environmental monitoring of water resources and predictive modeling of hydrobiological processes.

Comparison of Faster R-CNN, YOLO, and SSD for Third Molar Angle Detection in Dental Panoramic X-rays.

The use of artificial intelligence algorithms (AI) has gained importance for dental applications in recent years. Analyzing AI information from different sensor data such as images or panoramic radiographs (panoramic X-rays) can help to improve medical decisions and achieve early diagnosis of different dental pathologies. In particular, the use of deep learning (DL) techniques based on convolutional neural networks (CNNs) has obtained promising results in dental applications based on images, in which approaches based on classification, detection, and segmentation are being studied with growing interest. However, there are still several challenges to be tackled, such as the data quality and quantity, the variability among categories, and the analysis of the possible bias and variance associated with each dataset distribution. This study aims to compare the performance of three deep learning object detection models-Faster R-CNN, YOLO V2, and SSD-using different ResNet architectures (ResNet-18, ResNet-50, and ResNet-101) as feature extractors for detecting and classifying third molar angles in panoramic X-rays according to Winter's classification criterion. Each object detection architecture was trained, calibrated, validated, and tested with three different feature extraction CNNs which are ResNet-18, ResNet-50, and ResNet-101, which were the networks that best fit our dataset distribution. Based on such detection networks, we detect four different categories of angles in third molars using panoramic X-rays by using Winter's classification criterion. This criterion characterizes the third molar's position relative to the second molar's longitudinal axis. The detected categories for the third molars are distoangular, vertical, mesioangular, and horizontal. For training, we used a total of 644 panoramic X-rays. The results obtained in the testing dataset reached up to 99% mean average accuracy performance, demonstrating the YOLOV2 obtained higher effectiveness in solving the third molar angle detection problem. These results demonstrate that the use of CNNs for object detection in panoramic radiographs represents a promising solution in dental applications.

Baby Cry Sound Detection: A Comparison of Mel Spectrogram Image on Convolutional Neural Network Models

Baby cries contain patterns that indicate their needs, such as pain, hunger, discomfort, colic, or fatigue. This study explores the use of Convolutional Neural Network (CNN) architectures for classifying baby cries using Mel Spectrogram images. The primary objective of this research is to compare the effectiveness of various CNN architectures such as VGG-16, VGG-19, LeNet-5, AlexNet, ResNet-50, and ResNet-152 in detecting baby needs based on their cries. The datasets used include the Donate-a-Cry Corpus and Dunstan Baby Language. The results show that AlexNet achieved the best performance with an accuracy of 84.78% on the Donate-a-Cry Corpus dataset and 72.73% on the Dunstan Baby Language dataset. Other models like ResNet-50 and LeNet-5 also demonstrated good performance although their computational efficiency varied, while VGG-16 and VGG-19 exhibited lower performance. This research provides significant contributions to the understanding and application of CNN models for baby cry classification. Practical implications include the development of baby cry detection applications that can assist parents and healthcare provide.

Lung Cancer Detection Using Convolutional Neural Networks

Lung cancer is among the most prevalent and deadly cancers worldwide. Accurate diagnosis and early detection are critical for improving lung cancer patient outcomes and survival rates. Thanks to developments in medical imaging technology, computer-aided diagnosis (CAD) systems have shown a great deal of promise in helping radiologists identify and diagnose lung cancer from medical images. Here, we present the use of convolutional neural networks (CNNs) to create an early detection system (CAD) for lung cancer. The suggested approach uses lung computed tomography (CT) scans as input and use a CNN architecture to extract high-level features from the pictures. We use transfer learning to enhance a CNN model trained on a large dataset of CT images. The CNN model has been taught to determine if a specific CT image contains lung cancer or not. We evaluate the performance of the proposed CAD system on a dataset of CT scans of the lungs from different institutions. The trial's results show that our CNN-based CAD system can reliably and precisely identify lung cancer from CT scans. We also show the comparative performance of our proposed system against the state-of-the-art machine learning methods for lung cancer prediction. In conclusion, the suggested CNN-based deep learning-based CAD system has produced encouraging results for lung cancer detection from CT scans. The approach might help radiologists identify and classify lung cancer early on, leading to better patient outcomes and survival rates. The viability and usefulness of the suggested approach in clinical practice require more study

The use of convolutional neural networks for abnormal behavior recognition in crowd scenes

This study introduces the Abnormality Converging Scene Analysis Method (ACSAM) to detect abnormal group behavior using monitored videos or CCTV images in crowded scenarios. Abnormal behavior recognition involves classifying activities and gestures in continuous scenes, which traditionally presents significant computational challenges, particularly in complex crowd scenes, leading to reduced recognition accuracy. To address these issues, ACSAM employs a convolutional neural network (CNN) enhanced with Abnormality and Crowd Behavior Training layers to accurately detect and classify abnormal activities, regardless of crowd density. The method involves extracting frames from the input scene and using CNN to perform conditional validation of abnormality factors, comparing current values with previous high values to maximize detection accuracy. As the abnormality factor increases, the identification rate improves with higher training iterations. The system was tested on 26 video samples and trained on 34 samples, demonstrating superior performance to other approaches like DeepROD, MSI-CNN, and PT-2DCNN. Specifically, ACSAM achieved a 12.55% improvement in accuracy, a 12.97% increase in recall, and a 10.23% enhancement in convergence rate. These results suggest that ACSAM effectively overcomes the computational challenges inherent in crowd scene detection, offering a robust solution for real-time abnormal behavior recognition in crowded environments.

Robust mosquito species identification from diverse body and wing images using deep learning

Mosquito-borne diseases are a major global health threat. Traditional morphological or molecular methods for identifying mosquito species often require specialized expertise or expensive laboratory equipment. The use of convolutional neural networks (CNNs) to identify mosquito species based on images may offer a promising alternative, but their practical implementation often remains limited. This study explores the applicability of CNNs in classifying mosquito species. It compares the efficacy of body and wing depictions across three image collection methods: a smartphone, macro-lens attached to a smartphone and a professional stereomicroscope. The study included 796 specimens of four morphologically similar Aedes species, Aedes aegypti, Ae. albopictus, Ae. koreicus and Ae. japonicus japonicus. The findings of this study indicate that CNN models demonstrate superior performance in wing-based classification 87.6% (95% CI: 84.2–91.0) compared to body-based classification 78.9% (95% CI: 77.7–80.0). Nevertheless, there are notable limitations of CNNs as they perform reliably across multiple devices only when trained specifically on those devices, resulting in an average decline of mean accuracy by 14%, even with extensive image augmentation. Additionally, we also estimate the required training data volume for effective classification, noting a reduced requirement for wing-based classification compared to body-based methods. Our study underscores the viability of both body and wing classification methods for mosquito species identification while emphasizing the need to address practical constraints in developing accessible classification systems.Graphical abstract

CNN applied to ultrasonic guided wave spectrum image classification

Abstract Osteoporosis is a worldwide problem associated with an increasing number of fragility fractures. Currently, the standard for identifying patients at risk of fragility fracture is through Dual X-ray Absorptiometry (DXA). Different altenatives have been proposed, such as magnetic resonance imaging (MRI), three-dimensional X-rays, ultrasound or algorithms providing scores from clinical data. Among ultrasonic techniques, Bi-Directional Axial Transmission (BDAT) has been used to classify patients with or without fragility fractures, initially using ”classical” ultrasound parameters, such as velocities and latter using Support Vector Machine and automatic features, with performances close to the gold standard DXA. The aim of this study was to investigate the use of Convolutional Neural Networks (CNN) applied to patient classification using ultrasonic guided wave spectrum images, using a previous database of post menopausal women with or without fragility fractures. Two networks will be tested, a reference one, ResNet, successfully applied in classification and diagnosis in medical images, and a tailored one, denoted BDAT-Net, which hyperparameters will be optimized through a grid approach. The obtained accuracy, using BDAT-Net and clinical data (age, body mass index, cortisone intake) was found equal to 0.66 [0.64-0.69] comparable with the one obtained with DXA and significantly better than the one obtained with ResNet. These encouraging results open the door to the use of robust ultrasonic devices for fracture risk assessment, in particular in countries where DXA is not widely available.

Reusable Architecture Growth for Continual Stereo Matching.

The remarkable performance of recent stereo depth estimation models benefits from the successful use of convolutional neural networks to regress dense disparity. Akin to most tasks, this needs gathering training data that covers a number of heterogeneous scenes at deployment time. However, training samples are typically acquired continuously in practical applications, making the capability to learn new scenes continually even more crucial. For this purpose, we propose to perform continual stereo matching where a model is tasked to 1) continually learn new scenes, 2) overcome forgetting previously learned scenes, and 3) continuously predict disparities at inference. We achieve this goal by introducing a Reusable Architecture Growth (RAG) framework. RAG leverages task-specific neural unit search and architecture growth to learn new scenes continually in both supervised and self-supervised manners. It can maintain high reusability during growth by reusing previous units while obtaining good performance. Additionally, we present a Scene Router module to adaptively select the scene-specific architecture path at inference. Comprehensive experiments on numerous datasets show that our framework performs impressively in various weather, road, and city circumstances and surpasses the state-of-the-art methods in more challenging cross-dataset settings. Further experiments also demonstrate the adaptability of our method to unseen scenes, which can facilitate end-to-end stereo architecture learning and practical deployment.

Scalable CNN-based classification of selective sweeps using derived allele frequencies.

Selective sweeps can successfully be distinguished from neutral genetic data using summary statistics and likelihood-based methods that analyze single nucleotide polymorphisms (SNPs). However, these methods are sensitive to confounding factors, such as severe population bottlenecks and old migration. By virtue of machine learning, and specifically convolutional neural networks (CNNs), new accurate classification models that are robust to confounding factors have been recently proposed. However, such methods are more computationally expensive than summary-statistic-based ones, yielding them impractical for processing large-scale genomic data. Moreover, SNP data are frequently preprocessed to improve classification accuracy, further exacerbating the long analysis times. To this end, we propose a 1D CNN-based model, dubbed FAST-NN, that does not require any preprocessing while using only derived allele frequencies instead of summary statistics or raw SNP data, thereby yielding a sample-size-invariant, scalable solution. We evaluated several data fusion approaches to account for the variance of the density of genetic diversity across genomic regions (a selective sweep signature), and performed an extensive neural architecture search based on a state-of-the-art reference network architecture (SweepNet). The resulting model, FAST-NN, outperforms the reference architecture by up to 12% inference accuracy over all challenging evolutionary scenarios with confounding factors that were evaluated. Moreover, FAST-NN is between 30× and 259× faster on a single CPU core, and between 2.0× and 6.2× faster on a GPU, when processing sample sizes between 128 and 1000 samples. Our work paves the way for the practical use of CNNs in large-scale selective sweep detection. https://github.com/SjoerdvandenBelt/FAST-NN.

An integrated approach to Bayesian weight regulations and multitasking learning methods for generating emotion-based content in the metaverse

This paper introduces an integrated model designed to analyze topics and sentiments in textual data simultaneously, recognizing their interdependence. By tackling challenges such as data scarcity, missing information, and biased distributions, the proposed model effectively captures the dynamic interactions between topics and sentiments within complex datasets. The method combines the use of Convolutional Neural Networks (CNN) for detecting topic-related patterns with a revised Recurrent Neural Network (RNN) to trace the emotional flow within contexts, leveraging the strengths of sequential data processing. These components are integrated within a Bayesian probability framework, modeling the conditional probabilities of sentences expressing specific sentiments and documents being associated with particular topics. The combined feature and state vectors from the CNN and revised RNN within this Bayesian setup enable precise classification and prediction of topics and sentiments. Furthermore, this paper explores innovative research avenues, including sentiment analysis in 3D virtual reality environments and the development of new algorithms that reflect content creation techniques in the metaverse, offering dynamic system construction. This integrated approach not only enhances data-driven decision-making but also unlocks new possibilities for advanced multi-text analysis.

A secured deep learning based smart home automation system

With the expansion of modern technologies and the Internet of Things (IoT), the concept of smart homes has gained tremendous popularity with a view to making people’s lives easier by ensuring a secured environment. Several home automation systems have been developed to report suspicious activities by capturing the movements of residents. However, these systems are associated with challenges such as weak security, lack of interoperability and integration with IoT devices, timely reporting of suspicious movements, etc. Therefore, the given paper proposes a novel smart home automation framework for controlling home appliances by integrating with sensors, IoT devices, and microcontrollers, which would in turn monitor the movements and send notifications about suspicious movements on the resident’s smartphone. The proposed framework makes use of convolutional neural networks (CNNs) for motion detection and classification based on pre-processing of images. The images related to the movements of residents are captured by a spy camera installed in the system. It helps in identification of outsiders based on differentiation of motion patterns. The performance of the framework is compared with existing deep learning models used in recent studies based on evaluation metrics such as accuracy (%), precision (%), recall (%), and f-1 measure (%). The results show that the proposed framework attains the highest accuracy (98.67%), thereby surpassing the existing deep learning models used in smart home automation systems.