Use Of Data Augmentation Techniques Research Articles

MRI-Based Brain Tumor Classification using ResNet-50 and Optimized Softmax Regression

Accurate classification of brain tumors is crucial for effective treatment planning and patient management. This study presents a new hybrid deep learning classification method based on transfer learning by feature extraction to automate the categorization of MRI brain image datasets into four classes: meningioma, glioma, pituitary tumor, and no tumor. The proposed method combines a finely-tuned ResNet-50 model, a state-of-the-art convolutional neural network architecture, with optimized Softmax Regression (SR) for classification. The study explores the use of data augmentation techniques and evaluates the model's performance on both augmented and unaugmented images. The results demonstrate that the proposed method achieves an impressive accuracy of 98.4%, outperforming existing methods for automatic brain tumor detection. Furthermore, a detailed comparative analysis is presented to evaluate the proposed model's accuracy and efficiency against previous state-of-the-art hybrid models for brain tumor classification. The study suggests that the proposed methodology could be employed as a diagnostic tool to aid radiologists in identifying questionable brain regions, potentially improving the accuracy and efficiency of brain tumor diagnosis.

Classification of Avocado Ripeness Levels Using CNN Method

This research aims to develop a model for classifying the ripeness level of avocados using the Convolutional Neural Network (CNN) algorithm. The dataset comprises images of avocados categorized into three classes: unripe, ripe, and overripe. The CNN model is trained to classify the images into one of these three categories. The results indicate that the developed model can classify avocado images with high accuracy. The primary tool used for developing and implementing this method is MATLAB R2022a. The CNN algorithm is utilized to recognize and classify the ripeness level of avocados. This process involves several image processing steps, starting with preprocessing, image enhancement, and segmentation to isolate the avocado area. The dataset used in this research consists of 452 images distributed in 3 classes (unripe with 142, ripe with 66, and rotten with 244), with 80% used for training and 20% for testing. After 10 accuracy tests, the results indicate an accuracy rate of 90%. Additionally, features extracted from the images include color, shape, size, and texture characteristics, such as Mean, Standard Deviation, Kurtosis, Skewness, Variance, Entropy Value, Maximum Pixel, and Minimum Pixel. This research contributes to the field of agricultural technology by providing a robust method for the automatic classification of avocado ripeness. The findings are expected to facilitate accurate and efficient recognition of avocado ripeness, thereby supporting agricultural practices and market operations. Future research could explore the use of data augmentation techniques to further improve the accuracy and generalization of this model.

Audio Stream Analysis for Deep Fake Threat Identification

This article introduces a novel approach for the identification of deep fake threats within audio streams, specifically targeting the detection of synthetic speech generated by text-to-speech (TTS) algorithms. At the heart of this system are two critical components: the Vocal Emotion Analysis (VEA) Network, which captures the emotional nuances expressed within speech, and the Supervised Classifier for Deepfake Detection, which utilizes the emotional features extracted by the VEA to distinguish between authentic and fabricated audio tracks. The system capitalizes on the nuanced deficit of deepfake algorithms in replicating the emotional complexity inherent in human speech, thus providing a semantic layer of analysis that enhances the detection process. The robustness of the proposed methodology has been rigorously evaluated across a variety of datasets, ensuring its efficacy is not confined to controlled conditions but extends to realistic and challenging environments. This was achieved through the use of data augmentation techniques, including the introduction of additive white noise, which serves to mimic the variabilities encountered in real-world audio processing. The results have shown that the system's performance is not only consistent across different datasets but also maintains high accuracy in the presence of background noise, particularly when trained with noise-augmented datasets. By leveraging emotional content as a distinctive feature and applying sophisticated machine learning techniques, it presents a robust framework for safeguarding against the manipulation of audio content. This methodological contribution is poised to enhance the integrity of digital communications in an era where synthetic media is proliferating at an unprecedented rate.

YOLOv8 Analysis for Vehicle Classification Under Various Image Conditions

Purpose: The purpose of this research is to detect vehicle types in various image conditions using YOLOv8n, YOLOv8s, and YOLOv8m with augmentation.Methods: This research utilizes the YOLOv8 method on the DAWN dataset. The method involves using pre-trained Convolutional Neural Networks (CNN) to process the images and output the bounding boxes and classes of the detected objects. Additionally, data augmentation applied to improve the model's ability to recognize vehicles from different directions and viewpoints.Result: The mAP values for the test results are as follows: Without data augmentation, YOLOv8n achieved approximately 58%, YOLOv8s scored around 68.5%, and YOLOv8m achieved roughly 68.9%. However, after applying horizontal flip data augmentation, YOLOv8n's mAP increased to about 60.9%, YOLOv8s improved to about 62%, and YOLOv8m excelled with a mAP of about 71.2%. Using horizontal flip data augmentation improves the performance of all three YOLOv8 models. The YOLOv8m model achieves the highest mAP value of 71.2%, indicating its high effectiveness in detecting objects after applying horizontal flip augmentation. Novelty: This research introduces novelty by employing the latest version of YOLO, YOLOv8, and comparing its performance with YOLOv8n, YOLOv8s, and YOLOv8m. The use of data augmentation techniques, such as horizontal flip, to increase data variation is also novel in expanding the dataset and improving the model's ability to recognize objects.

Effective time-series Data Augmentation with Analytic Wavelets for bearing fault diagnosis

In the realm of rotary machine maintenance, rolling bearings emerge as crucial yet frequently vulnerable components. Ensuring their operational integrity is pivotal for the overall safety and performance of the machinery. Traditional diagnostic methods have utilized a variety of signal processing techniques, ranging from blind source separation to wavelet analysis, and more recently, deep neural networks (DNNs). However, a significant challenge in this domain is the scarcity of labeled failure data in real-world applications, leading to a reliance on data augmentation (DA) techniques to enhance the quality and quantity of the training data. Compounding this problem is the use of DA techniques tailored for images, which are not suitable for the time–frequency representations. This study proposes an innovative Data Augmentation with Analytic Wavelets (DAAW) approach, tailored specifically for time-series (TS) data classification in bearing fault diagnosis. The essence of DAAW lies in its ability to generate synthetic scalogram samples that closely mirror the properties of original samples. The proposed DAAW is directly applied at the input stage of a learning model by successively adjusting predefined parameters to generate scalograms, without the need for auxiliary DA algorithms. This technique is versatile in generating synthetic time–frequency data samples, and the parameterization enables it to be tailored to different time series datasets. Through experimentation with publicly available datasets, we show that combining a simple Convolutional Neural Network (CNN) with the proposed approach results in improved performance. It outperforms existing methods, achieving a mean classification accuracy of 99.49%. The critical distance for average accuracy, as determined by the Nemenyi post-hoc test, is 0.4619. Our results also show enhanced performance with accuracy rates of 90.14%, 78.94%, and 62.86% even under low signal-to-noise ratio (SNR) conditions, using 80%, 60%, and 50% of the dataset, respectively, for model training. Notably, its efficacy remains robust even under challenging low SNR conditions. The proposed DAAW significantly improves classification accuracy in bearing fault diagnosis, simultaneously obviating the need for proxy DA tasks.

Data Augmentation for Deep Receivers

Deep neural networks (DNNs) allow digital receivers to learn to operate in complex environments. To do so, DNNs should preferably be trained using large labeled data sets with a similar statistical relationship as the one under which they are to infer. For DNN-aided receivers, obtaining labeled data conventionally involves pilot signalling at the cost of reduced spectral efficiency, typically resulting in access to limited data sets. In this paper, we study how one can enrich a small set of labeled pilots data into a larger data set for training deep receivers. Motivated by the widespread use of data augmentation techniques for enriching visual and text data, we propose dedicated augmentation schemes that exploits the characteristics of digital communication data. We identify the key considerations in data augmentations for deep receivers as the need for domain orientation, class (constellation) diversity, and low complexity. Following these guidelines, we devise three complementing augmentations that exploit the geometric properties of digital constellations. Our combined augmentation approach builds on the merits of these different augmentations to synthesize reliable data from a momentary channel distribution, to be used for training deep receivers. Furthermore, we exploit previous channel realizations to increase the reliability of the augmented samples. The superiority of our approach is numerically evaluated for training several deep receiver architectures in different channel conditions. We consider both linear and non-linear synthetic channels, as well as the COST 2100 channel generator, for both single-input single-output and multiple-input multiple-output scenarios. We show that our combined augmentations approach allows DNN-aided receivers to achieve gains of up to 1 dB in bit error rate and of up to ×3 in spectral efficiency, compared to regular non-augmented training. Moreover, we demonstrate that our augmentations benefit training even as the number of pilots increases, and perform an ablation study on the different augmentations, which shows that the combined approach surpasses each individual augmentation technique.

Analysis and training of a traffic sign recognition neural network model

Objective. The purpose of the research is to develop and train a neural network model based on convolutional neural networks for effective recognition of road signs in images.Method. Deep learning methods were used, namely convolutional neural networks, which allow you to automatically extract image characteristics and train on a large data set. The research methodology included the following steps: collecting and preparing a variety of road sign data, creating and training a neural network model based on convolutional layers, applying data augmentation methods to improve model performance, and evaluating the model’s effectiveness on a test data set.Result. A neural network model is developed that can classify various types of road signs based on input images with high accuracy. The model was trained on diverse and high-quality data, allowing it to generalize and recognize road signs in different lighting conditions and camera angles. The use of data augmentation techniques significantly increased the model’s performance and improved its generalization ability.Conclusion. The study highlights the importance of using diverse and high-quality data to train a model and applying data augmentation techniques to improve its performance. The study confirms the effectiveness of using neural networks, especially convolutional neural networks, for the task of recognizing road signs in images.

A Review on Neural Network Based Models for Short Term Solar Irradiance Forecasting

The accuracy of solar energy forecasting is critical for power system planning, management, and operation in the global electric energy grid. Therefore, it is crucial to ensure a constant and sustainable power supply to consumers. However, existing statistical and machine learning algorithms are not reliable for forecasting due to the sporadic nature of solar energy data. Several factors influence the performance of solar irradiance, such as forecasting horizon, weather classification, and performance evaluation metrics. Therefore, we provide a review paper on deep learning-based solar irradiance forecasting models. These models include Long Short-Term Memory (LTSM), Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN), Convolutional Neural Network (CNN), Generative Adversarial Networks (GAN), Attention Mechanism (AM), and other existing hybrid models. Based on our analysis, deep learning models perform better than conventional models in solar forecasting applications, especially in combination with some techniques that enhance the extraction of features. Furthermore, the use of data augmentation techniques to improve deep learning performance is useful, especially for deep networks. Thus, this paper is expected to provide a baseline analysis for future researchers to select the most appropriate approaches for photovoltaic power forecasting, wind power forecasting, and electricity consumption forecasting in the medium term and long term.

COVID-19 Diagnosis in Computerized Tomography (CT) and X-ray Scans Using Capsule Neural Network.

This study proposes a deep-learning-based solution (named CapsNetCovid) for COVID-19 diagnosis using a capsule neural network (CapsNet). CapsNets are robust for image rotations and affine transformations, which is advantageous when processing medical imaging datasets. This study presents a performance analysis of CapsNets on standard images and their augmented variants for binary and multi-class classification. CapsNetCovid was trained and evaluated on two COVID-19 datasets of CT images and X-ray images. It was also evaluated on eight augmented datasets. The results show that the proposed model achieved classification accuracy, precision, sensitivity, and F1-score of 99.929%, 99.887%, 100%, and 99.319%, respectively, for the CT images. It also achieved a classification accuracy, precision, sensitivity, and F1-score of 94.721%, 93.864%, 92.947%, and 93.386%, respectively, for the X-ray images. This study presents a comparative analysis between CapsNetCovid, CNN, DenseNet121, and ResNet50 in terms of their ability to correctly identify randomly transformed and rotated CT and X-ray images without the use of data augmentation techniques. The analysis shows that CapsNetCovid outperforms CNN, DenseNet121, and ResNet50 when trained and evaluated on CT and X-ray images without data augmentation. We hope that this research will aid in improving decision making and diagnostic accuracy of medical professionals when diagnosing COVID-19.

Impact of Data Augmentation Techniques on the Implementation of a Combination Model of Convolutional Neural Network (CNN) and Multilayer Perceptron (MLP) for the Detection of Diseases in Rice Plants

Detection of disease in rice plants is important to avoid damage and reduce yields. In this study, the influence of data augmentation techniques on the application of a combination model of Convolutional Neural Network (CNN) and Multilayer Perceptron (MLP) was carried out to detect diseases in rice plants. This study uses a dataset of rice images for disease detection which contains 6034 images of rice with five categories, namely Bacterial Leaf Bligh, Blast, Brown Spot, normal, and tungro. This dataset is divided into three parts, namely training data, validation data and test data. The data augmentation techniques used are rotation, brightness and zoom on rice images. The combination model of CNN and MLP is built using the Python programming language and the TensorFlow deep learning framework. In measuring the success rate of the built model, it can be measured using the accuracy, precision and recall values obtained in the model test. Several scenarios were carried out to produce the best model, namely the use of data augmentation techniques, the number of layers and the number of iterations (epochs). From the experiments that have been carried out which have been tested with data as many as 25 digital images, the best model is obtained with an testing accuracy of 92%, 94% precision and 92% recall. This model applies a random zoom augmentation technique with a value of 0.5 – 1.0, CNN+MLP with 3 layers and a dataset ratio of 80:20 and an epoch early stop, This result has increased by more than 10%.

Automatic 3D Building Reconstruction from OpenStreetMap and LiDAR Using Convolutional Neural Networks

This paper presents the implementation of an automatic method for the reconstruction of 3D building maps. The core innovation of the proposed method is the supplementation of OpenStreetMap data with LiDAR data to reconstruct 3D urban environments automatically. The only input of the method is the area that needs to be reconstructed, defined by the enclosing points in terms of the latitude and longitude. First, area data are requested in OpenStreetMap format. However, there are certain buildings and geometries that are not fully received in OpenStreetMap files, such as information on roof types or the heights of buildings. To complete the information that is missing in the OpenStreetMap data, LiDAR data are read directly and analyzed using a convolutional neural network. The proposed approach shows that a model can be obtained with only a few samples of roof images from an urban area in Spain, and is capable of inferring roofs in other urban areas of Spain as well as other countries that were not used to train the model. The results allow us to identify a mean of 75.57% for height data and a mean of 38.81% for roof data. The finally inferred data are added to the 3D urban model, resulting in detailed and accurate 3D building maps. This work shows that the neural network is able to detect buildings that are not present in OpenStreetMap for which in LiDAR data are available. In future work, it would be interesting to compare the results of the proposed method with other approaches for generating 3D models from OSM and LiDAR data, such as point cloud segmentation or voxel-based approaches. Another area for future research could be the use of data augmentation techniques to increase the size and robustness of the training dataset.

Towards global scale segmentation with OpenStreetMap and remote sensing

Land Use Land Cover (LULC) segmentation is a famous application of remote sensing in an urban environment. Up-to-date and complete data are of major importance in this field. Although with some success, pixel-based segmentation remains challenging because of class variability. Due to the increasing popularity of crowd-sourcing projects, like OpenStreetMap, the need for user-generated content has also increased, providing a new prospect for LULC segmentation. We propose a deep-learning approach to segment objects in high-resolution imagery by using semantic crowdsource information. Due to satellite imagery and crowdsource database complexity, deep learning frameworks perform a significant role. This integration reduces computation and labor costs. Our methods are based on a fully convolutional neural network (CNN) that has been adapted for multi-source data processing. We discuss the use of data augmentation techniques and improvements to the training pipeline. We applied semantic (U-Net) and instance segmentation (Mask R-CNN) methods and, Mask R–CNN showed a significantly higher segmentation accuracy from both qualitative and quantitative viewpoints. The conducted methods reach 91% and 96% overall accuracy in building segmentation and 90% in road segmentation, demonstrating OSM and remote sensing complementarity and potential for city sensing applications.

Textual Augmentation Techniques Applied to Low Resource Machine Translation: Case of Swahili

In this work we investigate the impact of applying textual data augmentation tasks to low resource machine translation. There has been recent interest in investigating approaches for training systems for languages with limited resources and one popular approach is the use of data augmentation techniques. Data augmentation aims to increase the quantity of data that is available to train the system. In machine translation, majority of the language pairs around the world are considered low resource because they have little parallel data available and the quality of neural machine translation (NMT) systems depend a lot on the availability of sizable parallel corpora. We study and apply three simple data augmentation techniques popularly used in text classification tasks; synonym replacement, random insertion and contextual data augmentation and compare their performance with baseline neural machine translation for English-Swahili (En-Sw) datasets. We also present results in BLEU, ChrF and Meteor scores. Overall, the contextual data augmentation technique shows some improvements both in the EN -> SW and SW -> EN directions. We see that there is potential to use these methods in neural machine translation when more extensive experiments are done with diverse datasets.

Harnessing the Power of Convolutional Neural Network for Exoplanet Discovery

The discovery of planets apart from Earth that can sustain lives has always been fascinating as well as challenging. Discussion around such planets, popularly termed as "Exoplanets" have been doing the rounds for quite some time now. These exoplanets are often considered to be "Earth-like" or "habitable" because they may have conditions that could potentially support life. This work focuses on how Deep Learning techniques can be useful in identifying potential exoplanets. To do so, astronomical data gathered by space telescopes such as Kepler and BRITE have been utilized. The method employed to detect exoplanets is Transit Photometry along with Convolutional Neural Network. The study highlights the limitations of small training datasets and suggests the use of data augmentation techniques to increase the size of the training dataset, and the transfer learning approach to improve the performance of the classification models. The research offers valuable insights into the nature and diversity of exoplanets and may open avenues for future discoveries. With a performance accuracy of 96.67%, the proposed approach showcases merit and hence can prove to be a harbinger in exploring planetary habitability in the colossal space.

Tree species classification from complex laser scanning data in Mediterranean forests using deep learning

Abstract Recent advances in terrestrial laser scanning (TLS) technology have enabled the automatic capture of three‐dimensional vegetation structure at high resolution, but the scalability of using these data for large‐scale forest monitoring is limited by reliance on intensive manual data processing, including the use of stem maps generated in the field to determine tree species. New methods from data science have the capacity to automate this identification process, reducing the hurdles towards automated inventories with TLS. In particular, contemporary developments in point cloud processing methods, alongside large increases in the computing power of consumer‐level graphics processing units, provide new opportunities. Here, we apply a deep learning‐based approach, based on joint classification from multiple viewpoints for each stem, to automatically classify tree species directly from laser scanning data obtained in structurally complex Mediterranean forests. We also explore the use of data augmentation techniques to maximise performance for a fixed number of manually labelled stems. Our method does not require expensive pre‐processing such as leaf‐wood separation or quantitative reconstructions. Using modern network architectures and data augmentation techniques, and without extensive pre‐processing, we are able to achieve high overall and per‐species accuracy that is comparable or higher than in existing work while using data from a water‐limited ecosystem complicated by structural convergence and multi‐stem trees. Our findings demonstrate the power of deep learning to remove a major TLS data processing obstacle—individual species identification—and to minimise the bottleneck created by manual data labelling requirements in the use of TLS for standard forest monitoring.

Use of Data Augmentation Techniques in Detection of Antisocial Behavior Using Deep Learning Methods

The work presented in this paper focuses on the use of data augmentation techniques applied in the domain of the detection of antisocial behavior. Data augmentation is a frequently used approach to overcome issues related to the lack of data or problems related to imbalanced classes. Such techniques are used to generate artificial data samples used to improve the volume of the training set or to balance the target distribution. In the antisocial behavior detection domain, we frequently face both issues, the lack of quality labeled data as well as class imbalance. As the majority of the data in this domain is textual, we must consider augmentation methods suitable for NLP tasks. Easy data augmentation (EDA) represents a group of such methods utilizing simple text transformations to create the new, artificial samples. Our main motivation is to explore EDA techniques’ usability on the selected tasks from the antisocial behavior detection domain. We focus on the class imbalance problem and apply EDA techniques to two problems: fake news and toxic comments classification. In both cases, we train the convolutional neural networks classifier and compare its performance on the original and EDA-extended datasets. EDA techniques prove to be very task-dependent, with certain limitations resulting from the data they are applied on. The model’s performance on the extended toxic comments dataset did improve only marginally, gaining only 0.01 improvement in the F1 metric when applying only a subset of EDA methods. EDA techniques in this case were not suitable enough to handle texts written in more informal language. On the other hand, on the fake news dataset, the performance was improved more significantly, boosting the F1 score by 0.1. Improvement was most significant in the prediction of the minor class, where F1 improved from 0.67 to 0.86.

Data Augmentation-aided Convolutional Neural Network for Detection of Abnormalities in Digital Mammography

Background: The use of data augmentation techniques to addressing the challenge of network overfitting and classification error is important in deep learning. Insufficient sample data for training have the tendency to bias the trained model so that it fails to generalize well. Several studies have proposed different augmentation techniques to solve this problem. But there are some peculiarities identified with the nature of datasets when applying augmentation methods. The subtle nature of some abnormalities in digital mammography often makes it difficult to transform such datasets into different form, while preserving the structure of the abnormality. Aim: To address this, this study aims to apply a combination of carefully selected data augmentation operations on digital mammography.

Towards a device for helping deaf people to dance: estimation of forro bar length using artificial neural network

Dance has the potential to improve people's quality of life, as well as assist to decrease depression and anxiety. However, the lack of technologies capable of exploring alternative senses of hearing limits music and dance's beneficial effects on listeners. In order to find a model capable of being implemented in accessible devices, this work evaluated the use of a model based on neural networks to estimate the forro music bar length. Model variations were trained for seven datasets composed of mixes of music samples without noise, with real noise and with white noise. For each dataset, the best variation was selected and these were evaluated for the same real noise samples. The model variations that were presented to samples with real noise in the training estimated the bar duration with an average percentage error of less than 7% in the test step, being significantly better the model trained only with real sample. The evaluated model was able to estimate the length of the forro music bar length, even in real scenarios, as long as it was presented in this scenario during training. Increased database diversity and the use of data augmentation techniques can lead to improvements in the generalizability of the model. The simplicity of the evaluated model and its ability to learn when properly trained, indicate its potential to be used, in real time, on a mobile device to pass the rhythm of forro music to deaf and hard of hearing (D/HH) people.

Performance Evaluation of Convolutional Neural Network Using Synthetic Medical Data Augmentation Generated by GAN

Data augmentation is widely used in image processing and pattern recognition problems in order to increase the richness in diversity of available data. It is commonly used to improve the classification accuracy of images when the available datasets are limited. Deep learning approaches have demonstrated an immense breakthrough in medical diagnostics over the last decade. A significant amount of datasets are needed for the effective training of deep neural networks. The appropriate use of data augmentation techniques prevents the model from over-fitting and thus increases the generalization capability of the network while testing afterward on unseen data. However, it remains a huge challenge to obtain such a large dataset from rare diseases in the medical field. This study presents the synthetic data augmentation technique using Generative Adversarial Networks to evaluate the generalization capability of neural networks using existing data more effectively. In this research, the convolutional neural network (CNN) model is used to classify the X-ray images of the human chest in both normal and pneumonia conditions; then, the synthetic images of the X-ray from the available dataset are generated by using the deep convolutional generative adversarial network (DCGAN) model. Finally, the CNN model is trained again with the original dataset and augmented data generated using the DCGAN model. The classification performance of the CNN model is improved by 3.2% when the augmented data were used along with the originally available dataset.

Multiclass Image Classification Using GANs and CNN Based on Holes Drilled in Laminated Chipboard.

The multiclass prediction approach to the problem of recognizing the state of the drill by classifying images of drilled holes into three classes is presented. Expert judgement was made on the basis of the quality of the hole, by dividing the collected photographs into the classes: “very fine,” “acceptable,” and “unacceptable.” The aim of the research was to create a model capable of identifying different levels of quality of the holes, where the reduced quality would serve as a warning that the drill is about to wear down. This could reduce the damage caused by a blunt tool. To perform this task, real-world data were gathered, normalized, and scaled down, and additional instances were created with the use of data-augmentation techniques, a self-developed transformation, and with general adversarial networks. This approach also allowed us to achieve a slight rebalance of the dataset, by creating higher numbers of images belonging to the less-represented classes. The datasets generated were then fed into a series of convolutional neural networks, with different numbers of convolution layers used, modelled to carry out the multiclass prediction. The performance of the so-designed model was compared to predictions generated by Microsoft’s Custom Vision service, trained on the same data, which was treated as the benchmark. Several trained models obtained by adjusting the structure and hyperparameters of the model were able to provide better recognition of less-represented classes than the benchmark.