Data Augmentation Research Articles

Enhancing speech emotion recognition with deep learning using multi-feature stacking and data augmentation

This study evaluates the effectiveness of data augmentation on 1D convolutional neural network (CNN) and transformer models for speech emotion recognition (SER) on the Ryerson audio-visual database of emotional speech and song (RAVDESS) dataset. The results show that data augmentation has a positive impact on improving emotion classification accuracy. Techniques such as noising, pitching, stretching, shifting, and speeding are applied to increase data variation and overcome class imbalance. The 1D CNN model with data augmentation achieved 94.5% accuracy, while the transformer model with data augmentation performed even better at 97.5%. This research is expected to contribute better insights for the development of accurate emotion recognition methods by using data augmentation with these models to improve classification accuracy on the RAVDESS dataset. Further research can explore larger and more diverse datasets and alternative model approaches.

Automatic speech recognition for Indonesian medical dictation in cloud environment

<span lang="EN-US">This paper introduces SPWPM, an automatic speech recognition (ASR) system designed specifically for Indonesian medical dictation. The main objective of SPWPM is to assist medical professionals in producing medical reports and diagnosing patients. Deployed within a cloud computing service architecture, SPWPM strives to achieve a minimum speech recognition accuracy of 95%. The ASR model of SPWPM is developed using Kaldi and PyChain technologies—creating a comprehensive training dataset involving collaboration with PT Dua-Empat-Tujuh and Harapan Kita Hospital. Several optimization techniques were applied, including language modeling with smoothing, lexicon generation using the Grapheme-to-Phoneme Converter, and data augmentation. The readiness of this technology to assist hospital users was assessed through two evaluations: the SPWPM architecture test and the SPWPM speech recognition test. The results demonstrate the system's preparedness in accurately transcribing medical dictation, showcasing its potential to enhance medical reporting for healthcare professionals in hospital environments.</span>

Malaria cell identification using improved machine learning and modified deep learning architecture

Malaria continues to be a serious problem for public health because of its occurrence in tropical and subtropical areas with inadequate healthcare systems and few resources. For prompt intervention and treatment of malaria, effective and precise diagnosis is essential. Professional pathologists examine blood smear films by hand to get a microscopic diagnosis and another way they will do a rapid antigen malaria test which produces the result of 50% accuracy. Convolutional neural network (CNN) is a type of deep learning (DL) model that has been effectively used for a variety of image recognition applications. Our suggested approach uses, improved machine learning (IML) methods like support vector machine (SVM)+principal component analysis (PCA) fit, SVM+t-distributed stochastic neighbor embedding (t-SNE) fit, and CNN architecture with an accuracy of 86.23%, 88.27%, and 97.16% accuracy respectively, to combine feature extraction, data augmentation, and modify the layers by including the SVM algorithm in the final layer of the CNN architecture. The proposed method will significantly reduce pathologists' burden by automating the identification of malaria and improving diagnosis accuracy in resourceconstrained contexts

A general-purpose pointer meter reading detection method based on adaptive feature fusion and data augmentation

Pointer meters continue to serve as the primary monitoring instruments within substations. To solve the issue of poor model accuracy and low reading accuracy when reading multiple types of pointer meters, this paper proposes a universal cascaded convolutional neural network for accurate reading of various types of pointer instruments. The proposed detection method consists of two stages: instrument digital positioning and instrument pointer fitting. In the digital(figure of a number) positioning stage, an instrument data augmentation method for “large background and small target” is proposed to improve the model's sensitivity to digital position and reduce the interference of redundant background information to the model. In addition, an adaptive feature fusion pyramid structure is proposed for more effective feature fusion and enhanced fuzzy small target recognition. Compared with the baseline model, the method introduced in this study realizes a 0.47 % improvement in target detection accuracy. In our proposed dataset, the average recognition accuracy for the digital position location phase is 98.65 %,the average reading errors in square instruments, circular instruments, and instruments with uneven scales are 0.76 %, 0.73 %, and 0.67 % respectively. The proposed universal detection method in this paper achieves ultra-high precision and can be applied to reading pointer instruments of different types and scales. Compared with existing pointer meter reading methods, the proposed detection method showcases superior simplicity and effectiveness. Furthermore, it showcases a reduced measurement error, thereby bolstering its practical utility in industrial contexts.

A back propagation neural network based respiratory motion modelling method.

This study presents the development of a backpropagation neural network-based respiratory motion modelling method (BP-RMM) for precisely tracking arbitrary points within lung tissue throughout free respiration, encompassing deep inspiration and expiration phases. Internal and external respiratory data from four-dimensional computed tomography (4DCT) are processed using various artificial intelligence algorithms. Data augmentation through polynomial interpolation is employed to enhance dataset robustness. A BP neural network is then constructed to comprehensively track lung tissue movement. The BP-RMM demonstrates promising accuracy. In cases from the public 4DCT dataset, the average target registration error (TRE) between authentic deep respiration phases and those forecasted by BP-RMM for 75 marked points is 1.819mm. Notably, TRE for normal respiration phases is significantly lower, with a minimum error of 0.511mm. The proposed method is validated for its high accuracy and robustness, establishing it as a promising tool for surgical navigation within the lung.

Coffee bean graded based on deep net models

Coffee is a widely consumed beverage, and sorting coffee beans is a critical process that ensures high-quality graded coffee products. Coffee beans were graded into nine grades in robusta types. To automate the grading process, a deep learning-based approach was developed using a large dataset of high-resolution images and data augmentation techniques. In contrast to previous studies focusing on robusta type graded into six coffee bean grads, our research extends this framework by employing robusta type into nine grades with an outperformed accuracy. The proposed work uses four deep learning models, namely residual network 34(Resnet34), inception version 3 (Inception v3), efficient network bayesian optimization (EfficientNet-B0), and visual geometry group-16(VGG-16), where trained and evaluated for coffee bean classification into nine grades. The EfficientNet-B0 model exhibited outperformed accuracy, achieving 100% in distinguishing good and bad coffee beans, even in challenging lighting and background conditions.

Palembang songket fabric motif image detection with data augmentation based on ResNet using dropout

A good way to spread knowledge about Palembang songket woven cloth patterns is to use information technology, especially artificial intelligence technology. This study's main goal is to develop a ResNet model with dropout regularization methods and find out how dropout regularization affects the ResNet model for detecting Palembang songket fabric motif with more data. Data was collected in places like tujuh saudara songket, Zainal songket, songket PaSH, AMS songket, and batik, Ernawati songket, Nabilah collections, Ilham songket, and Marissa songket. We used eight class of data for this research. A dataset of 7,680 data for training, 960 data for validation, and 960 data for testing is a dataset that has been prepared to be implemented in experiments. In the final results, the experimental results for DResNet demonstrated that accuracy at the training stage was 92.16%, accuracy at the validation stage was 78.60%, and accuracy at the submission stage was 80.3%. The experimental results also show that dropouts are able to increase the accuracy of the ResNet model by adding +1.10% accuracy in the training process, adding +1.80% accuracy in the validation process, and adding +0.40% accuracy in the testing process.

An investigation of machine learning algorithms and data augmentation techniques for diabetes diagnosis using class imbalanced BRFSS dataset

Diabetes is a prevalent chronic condition that poses significant challenges to early diagnosis and identifying at-risk individuals. Machine learning plays a crucial role in diabetes detection by leveraging its ability to process large volumes of data and identify complex patterns. However, imbalanced data, where the number of diabetic cases is substantially smaller than non-diabetic cases, complicates the identification of individuals with diabetes using machine learning algorithms. This study focuses on predicting whether a person is at risk of diabetes, considering the individual’s health and socio-economic conditions while mitigating the challenges posed by imbalanced data. We employ several data augmentation techniques, such as oversampling (Synthetic Minority Over Sampling for Nominal Data, i.e.SMOTE-N), undersampling (Edited Nearest Neighbor, i.e. ENN), and hybrid sampling techniques (SMOTE-Tomek and SMOTE-ENN) on training data before applying machine learning algorithms to minimize the impact of imbalanced data. Our study sheds light on the significance of carefully utilizing data augmentation techniques without any data leakage to enhance the effectiveness of machine learning algorithms. Moreover, it offers a complete machine learning structure for healthcare practitioners, from data obtaining to machine learning prediction, enabling them to make informed decisions.

A Convolution Neural Network-based Approach for Metal Surface Roughness Evaluation

Background: Metal surface roughness detection is an essential step of quality control in the metal processing industry. Due to the high manual involvement and poor efficiency of traditional roughness testing, rapid automated vision detection has received increasing attention in product quality control. Many methods have focused on extracting features related to roughness from images by means of mathematical statistics. However, these methods often rely on extensive experiments and complex calculations, while being sensitive to external environmental disturbances. Methods: In this paper, a convolution neural network-based approach for metal surface roughness evaluation has been proposed. The convolutional neural network was initialized using a transfer learning strategy, and the data augmentation technique was applied to the benchmark dataset for sample expansion. Results: To evaluate this approach, samples of 4 types of roughness classes were prepared. The samples were divided into a training set, validation set, and test set in the ratio of 7:2:1. The accuracy of the neural network on the test set was found to be above 86%. Conclusion: The effectiveness of the proposed approach and its superiority over manual detection have been demonstrated in the experiments.

Tomato pose estimation using the association of tomato body and sepal

In facility horticulture smart farms, harvesting robotic systems to automate harvesting tasks are challenging due to the complex environment, irregular growth and fruits pose. To harvest a fruits, the vision information required by a harvesting robot is accurate position and pose. Especially, fruit pose information is essential for planning paths to avoid damaging stems, leaves, branches, and obstacles, preventing damage to the fruit and harvesting robots, and planning efficient harvest sequences. This paper presents a method that uses information from tomato sepals to estimate the orientation of tomatoes, which are a commonly grown crop in horticulture. First, we train a YOLOv8 model to detect and segment bodies and sepals of tomatoes. For robust training of the model, the training data is constructed using effective data augmentation, which synthesizes segmented foreground objects and inserting them into the background. Then, for finding association between the body and sepal of a tomato, we apply IoU score based matching and the Hungarian algorithm. Consequently, we obtain point clouds for both parts using RGB-D data. Finally, we compute the center point of each object by using spherical fitting and the statistics of the point cloud, respectively. Then, we estimate the pose of the tomato as a vector between two center points of the body and sepal. To accurately and practically evaluate the proposed method, we generated ground truth data using calibration patterns in tomato greenhouse. As the experimental results show, The segmentation results show that AP50,sepal is 94.7, AP50,tomato is 96.3, and mAP is 61.5. The result of the pose estimation for the pose validation dataset is a mean error angle of 6.79 ± 3.18 and angle errors of less than 10 degrees account for 87.2%. The total algorithm proposed in this paper requires about 0.038s for each tomato greenhouse image. As a result, our proposed pose estimation can be practically utilized in robot systems for tomato harvesting.

Description and analysis of Sigfox received signal strength indicator dataset by using statistical techniques

Low power wide area network (LPWAN) technology has expanded and is essential in the development of applications for the internet of things (IoT). The Sigfox LPWAN network is characterized by its long-range coverage, low cost and power consumption. In this article, a set of 5174 values is analyzed, containing 1606 null RSSI data, obtained with the Sipy module and MicroPython, which provide a coverage map of several points with a resolution of 200 meters deployed in Quito–Ecuador. It is evaluated the type of distribution to which the set of network measurements is adjusted and an optimal 900 MHz propagation model in suburban environments is determined from the measurements obtained from the known base station. As a result, the lost values of RSSI were predicted using the inverse normal distribution method in the original values, observing that they conform to a logistic distribution. The data from the base station were subjected to a data augmentation algorithm designed in MATLAB, determining that the stanford university interim (SUI) model reduces the precision error in the trend of the curve by not presenting changes greater than 5 dB, achieving a precision of 97% with respect to the fit of the curve of the data.

Leaf Disease Detection Using Convolutional Neural Network

Leaf Disease Detection Using Convolutional Neural Network

Deep learning based Election Results Prediction using Twitter Data

Abstract: This project presents an intelligent system political sentiment analysis with information from Twitter. . The system utilizes deep learning techniques, specifically a BiLSTM model featuring an attention mechanism, to predict the political affiliation (BJP or Congress) of tweets. The data preprocessing involves cleaning text, data augmentation, tokenization, and indexing. A dataset is used to train the model. consisting of tweets from both parties, getting very accurate results in predicting the sentiment of tweets. Additionally, the software offers functionality for data visualization, including displaying sample tweets, sentiment analysis results, and the winning party. Additionally, an admin page is available for further customization and management. Overall, this project the possibilities of Using deep learning to analyze sentiment of political content on social media data.

Real-Time Face Mask Detection Using MobileNet V2

Abstract: The rapid disruption of international trade and travel caused by the COVID-19 epidemic has had an impact on our daily lives. The custom of donning a face mask for protection has evolved. In the near future, a lot of public service providers will demand that clients wear appropriate masks in order to use their services. Identification of face masks is becoming a crucial duty to support world culture. This paper outlines a condensed method for accomplishing this goal utilizing certain fundamental machine learning tools, such as TensorFlow, Keras, OpenCV, and Scikit-Learn. The project's goal is to identify face masks at a public event or gathering. MobileNet V2 is the algorithm employed in the project to accomplish the goal. There is a picture of a few people with and without masks. As an input dataset, a picture of a few individuals wearing and not wearing masks is employed. Pre- processing, data augmentation, training, testing, and image segmentation are some of the steps that go into reaching the project's goal. Creating a model that can identify people who are not wearing masks in public settings is the goal of this project. To guarantee adherence to the standards for public safety, this work can be combined with real-time applications at airports, train stations, businesses,schools, and other public locations.

Traffic Sign Recognition using HIICNN Stack Ensemble Method

Abstract: This project introduces a novel deep convolutional neural network (CNN) model aimed at enhancing traffic sign recognition for autonomous vehicle technology and road safety. Employing a stacking ensemble approach, we combine multiple CNNs with diverse improvement techniques, including preprocessing for adverse weather conditions and shooting errors, and data augmentation to handle class imbalance. Through adjustments in learning rates during model training, we mitigate overfitting. Our ensemble model achieves an exceptional 99.4 percent test accuracy on the German Traffic Sign Recognition Benchmark (GTSRB) dataset, surpassing prior studies. We utilize gradient-weighted Class Activation Mapping (Grad-CAM) for model explain- ability and an evidential deep learning approach to quantify classification uncertainty. Our framework underscores the efficacy of combining preprocessing, ensemble learning, and transfer learning for superior perfor- mance and reliability in traffic sign recognition systems, contributing significantly to autonomous driving and road safety.

Space-time image velocimetry in blurred scenes based on BSTI-DCGAN data augmentation

Due to the limited sample quantity and the complex data collection process of the blurred space-time image (BSTI) dataset, the deep learning-based space-time image velocimetry (STIV) results in larger errors when applied to blurry videos. To enhance the measurement accuracy, we propose the use of STIV in blurred scenes based on BSTI-deep convolutional generative adversarial network (DCGAN) data augmentation. Firstly, BSTI-DCGAN is developed based on the DCGAN. This network utilizes a bilinear interpolation-convolution module for upsampling and integrates coordinated attention and multi-concatenation attention to enhance the resemblance between generated and real images. Next, further expanding the dataset by using artificially synthesized space-time images subsequently, all space-time images are transformed into spectrograms to create a training dataset for the classification network. Finally, the primary spectral direction is detected using the classification network. The experimental results indicate that our approach effectively augments the dataset and improves the accuracy of practical measurements. Under the condition of video blur, the relative errors of the average flow velocity and discharge are 3.92% and 2.72%, respectively.

Physics-guided neural network for grinding temperature prediction

Creep-feed grinding is a high-efficiency, high-precision grinding process widely used in the manufacturing of aviation engines. However, the workpiece burn and other quality issues caused by high processing temperature limit the yield of grinding. Therefore, the accurate model of grinding temperature has become the key to improving processing efficiency and quality. Different from the traditional physical or data-driven models, this paper attempts to combine both perspectives based on Physics-Guided Neural Networks (PGNN) to accurately predict grinding temperature with a small number of experiments. At the level of data acquisition, real grinding experiment data was obtained and a data augmentation method had been proposed. At the level of neural network structure, optimisation processes were implemented to enhance prediction performance, and a physics-guided loss function was inserted to guide network training. The experiment results shows that PGNN had better prediction accuracy than the physical model, while also mitigating the limitations of data-driven models on small sample sets. PGNN also performed better with noisy data and predictions out of the training data range, this reveals the benefits of PGNN for small sample problems in processing scenarios.

Deep Learning-Based Surgical Treatment Recommendation and Nonsurgical Prognosis Status Classification for Scaphoid Fractures by Automated X-ray Image Recognition

Biomedical information retrieval for diagnosis, treatment and prognosis has been studied for a long time. In particular, image recognition using deep learning has been shown to be very effective for cancers and diseases. In these fields, scaphoid fracture recognition is a hot topic because the appearance of scaphoid fractures is not easy to detect. Although there have been a number of recent studies on this topic, no studies focused their attention on surgical treatment recommendations and nonsurgical prognosis status classification. Indeed, a successful treatment recommendation will assist the doctor in selecting an effective treatment, and the prognosis status classification will help a radiologist recognize the image more efficiently. For these purposes, in this paper, we propose potential solutions through a comprehensive empirical study assessing the effectiveness of recent deep learning techniques on surgical treatment recommendation and nonsurgical prognosis status classification. In the proposed system, the scaphoid is firstly segmented from an unknown X-ray image. Next, for surgical treatment recommendation, the fractures are further filtered and recognized. According to the recognition result, the surgical treatment recommendation is generated. Finally, even without sufficient fracture information, the doctor can still make an effective decision to opt for surgery or not. Moreover, for nonsurgical patients, the current prognosis status of avascular necrosis, non-union and union can be classified. The related experimental results made using a real dataset reveal that the surgical treatment recommendation reached 80% and 86% in accuracy and AUC (Area Under the Curve), respectively, while the nonsurgical prognosis status classification reached 91% and 96%, respectively. Further, the methods using transfer learning and data augmentation can bring out obvious improvements, which, on average, reached 21.9%, 28.9% and 5.6%, 7.8% for surgical treatment recommendations and nonsurgical prognosis image classification, respectively. Based on the experimental results, the recommended methods in this paper are DenseNet169 and ResNet50 for surgical treatment recommendation and nonsurgical prognosis status classification, respectively. We believe that this paper can provide an important reference for future research on surgical treatment recommendation and nonsurgical prognosis classification for scaphoid fractures.

Industrial data-driven modeling for imbalanced fault diagnosis

PurposeMotivated by recent research indicating the significant challenges posed by imbalanced datasets in industrial settings, this paper presents a novel framework for Industrial Data-driven Modeling for Imbalanced Fault Diagnosis, aiming to improve fault detection accuracy and reliability.Design/methodology/approachThis study addressing the challenge of imbalanced datasets in predicting hard drive failures is both innovative and comprehensive. By integrating data enhancement techniques with cost-sensitive methods, the research pioneers a solution that directly targets the intrinsic issues posed by imbalanced data, a common obstacle in predictive maintenance and reliability analysis.FindingsIn real industrial environments, there is a critical demand for addressing the issue of imbalanced datasets. When faced with limited data for rare events or a heavily skewed distribution of categories, it becomes essential for models to effectively mine insights from the original imbalanced dataset. This involves employing techniques like data augmentation to generate new insights and rules, enhancing the model’s ability to accurately identify and predict failures.Originality/valuePrevious research has highlighted the complexity of diagnosing faults within imbalanced industrial datasets, often leading to suboptimal predictive accuracy. This paper bridges this gap by introducing a robust framework for Industrial Data-driven Modeling for Imbalanced Fault Diagnosis. It combines data enhancement and cost-sensitive methods to effectively manage the challenges posed by imbalanced datasets, further innovating with a bagging method to refine model optimization. The validation of the proposed approach demonstrates superior accuracy compared to existing methods, showcasing its potential to significantly improve fault diagnosis in industrial applications.

Sample-Adaptive Classification Inference Network

Existing pre-trained models have yielded promising results in terms of computational time reduction. However, these models only focus on pruning simple sentences or less salient words, while neglecting the treatment of relatively complex sentences. It is frequently these sentences that cause the loss of model accuracy. This shows that the adaptation of the existing models is one-sided. To address this issue, in this paper, we propose a sample-adaptive training and inference model. Specifically, complex samples are extracted from the training datasets and a dedicated data augmentation module is trained to extract global and local semantic information of complex samples. During inference, simple samples can exit the model via the Sample Adaptive Exit Mechanism, Normal samples pass through the whole backbone model before inference, while complex samples are processed by the Characteristic Enhancement Module after passing through the backbone model. In this way, all samples are processed adaptively. Our extensive experiments on classification tasks datasets in the field of Natural Language Processing demonstrate that our method enhances model accuracy and reduces model inference time for multiple datasets. Moreover, our method is transferable and can be applied to multiple pre-trained models.