Data Augmentation For Deep Learning Research Articles

Assessing the reliability of point mutation as data augmentation for deep learning with genomic data

BackgroundDeep neural networks (DNNs) have the potential to revolutionize our understanding and treatment of genetic diseases. An inherent limitation of deep neural networks, however, is their high demand for data during training. To overcome this challenge, other fields, such as computer vision, use various data augmentation techniques to artificially increase the available training data for DNNs. Unfortunately, most data augmentation techniques used in other domains do not transfer well to genomic data.ResultsMost genomic data possesses peculiar properties and data augmentations may significantly alter the intrinsic properties of the data. In this work, we propose a novel data augmentation technique for genomic data inspired by biology: point mutations. By employing point mutations as substitutes for codons, we demonstrate that our newly proposed data augmentation technique enhances the performance of DNNs across various genomic tasks that involve coding regions, such as translation initiation and splice site detection.ConclusionSilent and missense mutations are found to positively influence effectiveness, while nonsense mutations and random mutations in non-coding regions generally lead to degradation. Overall, point mutation-based augmentations in genomic datasets present valuable opportunities for improving the accuracy and reliability of predictive models for DNA sequences.

The Effect of Data Augmentation in Deep Learning with Drone Object Detection

Drone object detection is one of the main applications of image processing technology and pattern recognition using deep learning. However, the limited drone image data that can be accessed for training detection algorithms is a challenge in the development of drone object detection technology. Therefore, many studies have been conducted to increase the amount of drone image data using data augmentation techniques. This study aims to evaluate the effect of data augmentation on deep learning accuracy in drone object detection using the YOLOv5 algorithm. The methods used in this research include collecting drone image data, augmenting data with rotate, crop and cutout, training the YOLOv5 algorithm with and without data augmentation, as well as testing and analyzing training results.The results of the study show that data augmentation can't improve the accuracy of the YOLOv5 algorithm in drone object detection. Evidenced by the decreasing value of precision and mAP@0.5 and the relatively constant value of recall and F-1 score. This is caused by too much augmentation can cause loss of important information in the data and improper augmentation can cause noise or distortion in the data.

Domain-guided data augmentation for deep learning on medical imaging.

While domain-specific data augmentation can be useful in training neural networks for medical imaging tasks, such techniques have not been widely used to date. Our objective was to test whether domain-specific data augmentation is useful for medical imaging using a well-benchmarked task: view classification on fetal ultrasound FETAL-125 and OB-125 datasets. We found that using a context-preserving cut-paste strategy, we could create valid training data as measured by performance of the resulting trained model on the benchmark test dataset. When used in an online fashion, models trained on this hybrid data performed similarly to those trained using traditional data augmentation (FETAL-125 F-score 85.33 ± 0.24 vs 86.89 ± 0.60, p-value 0.014; OB-125 F-score 74.60 ± 0.11 vs 72.43 ± 0.62, p-value 0.004). Furthermore, the ability to perform augmentations during training time, as well as the ability to apply chosen augmentations equally across data classes, are important considerations in designing a bespoke data augmentation. Finally, we provide open-source code to facilitate running bespoke data augmentations in an online fashion. Taken together, this work expands the ability to design and apply domain-guided data augmentations for medical imaging tasks.

Random walk-based erasing data augmentation for deep learning

Random walk-based erasing data augmentation for deep learning

Common‐offset domain reflection–diffraction separation with deep learning

AbstractSeparation of diffracted from reflected events in seismic data is still challenging due to the relatively low amplitude of the diffracted wavefield compared to the reflected wavefield as well as the overlap in the kinematics of reflection and diffraction events. A workflow based on deep learning can be a simple and fast alternative, but using training data made by physics‐based modelling is expensive and lacks diversity in terms of noise, amplitude, frequency content and wavelet. This results in poor generalization beyond the training data without retraining and transfer learning. In this paper, we demonstrate successful applications of reflection–diffraction separation using a conventional U‐net architecture. The novelty of our approach is that we do not use synthetic data created from physics‐based modelling, but instead use only synthetic data built from basic geometric shapes. Our domain of application is the pre‐migration common‐offset domain where reflected events resemble local geology and the diffracted wavefield consists of downward convex hyperbolic diffraction patterns. Both patterns were randomly perturbed in many ways while maintaining their intrinsic features. This approach is inspired by the common practice of data augmentation in deep learning for machine vision applications. Since many of the standard data augmentation techniques lack a geophysical motivation, we have instead perturbed our synthetic training data in ways to make more sense from a signal processing perspective or given our ‘domain knowledge’ of the problem at hand. We did not have to retrain the network to show good results on the field data set. The large variety and diversity in examples enabled to trained neural networks to show encouraging results on synthetic and field data sets that were not used in training.

Attenuation of seismic migration smile artifacts with deep learning

Attenuation of migration artifacts on Kirchhoff migrated seismic data can be challenging due to the relatively low amplitude of migration artifacts compared to reflections as well as the overlap in the kinematics of reflection and migration smiles. Several ‘conventional’ filtering methods exist and recently deep learning based workflows have been proposed. A deep learning workflow can be a simple and fast alternative to existing methods. In case of supervised training of a deep neural network using training data made by physics-based modelling or actual migrations is expensive and lacks diversity in terms of noise, amplitude, frequency content and wavelet. This can result in poor generalization beyond the training data without re-training and transfer learning. In this paper we demonstrate successful applications of migration smile separation using a conventional U-net architecture. The novelty in our approach is that we do not use synthetic data created from physics-based modelling, but instead use only synthetic data build form basic geometric shapes. Our domain of application is the migrated common offset domain, or simply the stack of the pre-stack migrated data, where reflections resemble local geology and migration smiles are upward convex hyperbolic patterns. Both patterns were randomly perturbed in many ways while maintaining their intrinsic features. This approach is inspired by the common practice of data augmentation in deep learning for machine vision applications. Since many of the standard data augmentation techniques lack a geophysical motivation, we have instead perturbed our synthetic training data in ways to make more sense for a signal processing perspective or given our ‘domain knowledge’ of the problem at hand. We did not have to retrain the network to produce good results on the field dataset. The large variety and diversity in examples enabled the trained neural network to show encouraging results on synthetic and field datasets that were not used in training.

A Widespread Assessment and Open Issues on Image Captioning Models

Automated generation of image captions is a demanding AI crisis as it necessitates the exploitation of numerous methods from diverse computer science fields. Deep learning (DL) approaches have revealed marvelous results in a lot of diverse appliances. On the other hand, data augmentation in DL that imitates the quantity and the variety of training data without the need of gathering additional data is a hopeful area in machine learning (ML). Producing textual descriptions for a specified image is a demanding task using the computer. This survey makes a critical analysis of about 65 papers regarding image captioning. More particularly, varied performance measures that are contributed in diverse articles are analyzed. In addition, a comprehensive study is made regarding the maximal performances and varied features deployed in each work. Moreover, chronological analysis and dataset analysis are done and finally, the survey extends with the determination of varied research challenges, which might be productive for the analysts to endorse enhanced upcoming works on image captioning.

Fourier transform-based data augmentation in deep learning for diabetic foot thermograph classification

Diabetes Mellitus (DM) belongs to the ten diseases group with the highest mortality rate globally, with an estimated 578 million cases by 2030, according to the World Health Organization (WHO). The disease manifests itself through different disorders, where vasculopathy shows a chronic relationship with diabetic ulceration events in distal extremities, being temperature a biomarker that can quantify the risk scale. According to the above, an analysis is performed with standing thermography images, finding temperature patterns that do not follow a particular distribution in patients with DM. Therefore, the modern medical literature has taken the use of Computer-Aided Diagnosis (CAD) systems as a plausible option to increase medical analysis capabilities. In this sense, we proposed to study three state-of-the-art deep learning (DL) architectures, experimenting with convolutional, residual, and attention (Transformers) approaches to classify subjects with DM from diabetic foot thermography images. The models were trained under three conditions of data augmentation. A novel method based on modifying the images through the change of the amplitude in the Fourier Transform is proposed, being the first work to perform such synergy in the characterization of risk in ulcers through thermographies. The results show that the proposed method allowed reaching the highest values, reaching a perfect classification through the convolutional neural network ResNet50v2, promising for limited data sets in thermal pattern classification problems.

Improve Neural Distinguishers of SIMON and SPECK

Deep learning has played an important role in many fields, which shows significant potential for cryptanalysis. Although these existing works opened a new direction of machine learning aided cryptanalysis, there is still a research gap that researchers are eager to fill. How to further improve neural distinguishers? In this paper, we propose a new algorithm and model to improve neural distinguishers in terms of accuracy and the number of rounds. First, we design an algorithm based on SAT to improve neural distinguishers. With the help of SAT/SMT solver, we obtain new effective neural distinguishers of SIMON using the input differences of high-probability differential characteristics. Second, we propose a new neural distinguisher model using multiple output differences. Inspired by the existing works and data augmentation in deep learning, we use the output differences to exploit more derived features and train neural distinguishers, by splicing output differences into a matrix as a sample. Based on the new model, we construct neural distinguishers of SIMON and SPECK with round and accuracy promotion. Utilizing our neural distinguishers, we can distinguish reduced-round SIMON or SPECK from pseudorandom permutation better.

Data augmentation for deep learning based semantic segmentation and crop-weed classification in agricultural robotics

Deep learning methods such as convolutional neural networks (CNN) have become popular for addressing crops and weeds classification problems in agricultural robotics. However, to have satisfactory performance and avoid overfitting, training deep neural nets typically requires thousands of labeled images. This leads to tedious pixelwise labeling for semantic segmentation. In this paper, we hinge on the recent development in data augmentation and utilize the concept further for semantic segmentation and classification of crops and weeds. To be specific, we propose a novel data augmentation framework, based on the random image cropping and patching (RICAP) method, which is originally designed to augment data for generic image classification. The proposed framework introduces novel enhancements to the original RICAP so that it can be effectively used for data augmentation of semantic segmentation tasks. We evaluate the proposed methodology on two datasets from different farms. Comprehensive experimental evaluations and ablation studies show that the proposed framework can effectively improve segmentation accuracies, and the enhancements made over the original RICAP actually contribute to the performance gain. On average, the proposed method increases the mean accuracy and mean intersection over union (IOU) of the deep neural net with the conventional data augmentation (random flipping, rotation and colour jitter) from 91.01 to 94.02 and from 63.59 to 70.77 respectively for Narrabri dataset, and from 97.99 to 98.51 and from 74.26 to 77.09 respectively for Bonn dataset. The limitation of the proposed method, especially when a large number of training data is available, has also been discussed.

Intraoral image generation by progressive growing of generative adversarial network and evaluation of generated image quality by dentists

Dentists need experience with clinical cases to practice specialized skills. However, the need to protect patient's private information limits their ability to utilize intraoral images obtained from clinical cases. In this study, since generating realistic images could make it possible to utilize intraoral images, progressive growing of generative adversarial networks are used to generate intraoral images. A total of 35,254 intraoral images were used as training data with resolutions of 128 × 128, 256 × 256, 512 × 512, and 1024 × 1024. The results of the training datasets with and without data augmentation were compared. The Sliced Wasserstein Distance was calculated to evaluate the generated images. Next, 50 real images and 50 generated images for each resolution were randomly selected and shuffled. 12 pediatric dentists were asked to observe these images and assess whether they were real or generated. The d prime of the 1024 × 1024 images was significantly higher than that of the other resolutions. In conclusion, generated intraoral images with resolutions of 512 × 512 or lower were so realistic that the dentists could not distinguish whether they were real or generated. This implies that the generated images can be used in dental education or data augmentation for deep learning, without privacy restrictions.

MeshCut data augmentation for deep learning in computer vision.

To solve overfitting in machine learning, we propose a novel data augmentation method called MeshCut, which uses a mesh-like mask to segment the whole image to achieve more partial diversified information. In our experiments, this strategy outperformed the existing augmentation strategies and achieved state-of-the-art results in a variety of computer vision tasks. MeshCut is also an easy-to-implement strategy that can efficiently improve the performance of the existing convolutional neural network models by a good margin without careful hand-tuning. The performance of such a strategy can be further improved by incorporating it into other augmentation strategies, which can make MeshCut a promising baseline strategy for future data augmentation algorithms.

Weakly Labeled Data Augmentation for Deep Learning: A Study on COVID-19 Detection in Chest X-Rays.

The novel severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has caused a pandemic resulting in over 2.7 million infected individuals and over 190,000 deaths and growing. Assertions in the literature suggest that respiratory disorders due to COVID-19 commonly present with pneumonia-like symptoms which are radiologically confirmed as opacities. Radiology serves as an adjunct to the reverse transcription-polymerase chain reaction test for confirmation and evaluating disease progression. While computed tomography (CT) imaging is more specific than chest X-rays (CXR), its use is limited due to cross-contamination concerns. CXR imaging is commonly used in high-demand situations, placing a significant burden on radiology services. The use of artificial intelligence (AI) has been suggested to alleviate this burden. However, there is a dearth of sufficient training data for developing image-based AI tools. We propose increasing training data for recognizing COVID-19 pneumonia opacities using weakly labeled data augmentation. This follows from a hypothesis that the COVID-19 manifestation would be similar to that caused by other viral pathogens affecting the lungs. We expand the training data distribution for supervised learning through the use of weakly labeled CXR images, automatically pooled from publicly available pneumonia datasets, to classify them into those with bacterial or viral pneumonia opacities. Next, we use these selected images in a stage-wise, strategic approach to train convolutional neural network-based algorithms and compare against those trained with non-augmented data. Weakly labeled data augmentation expands the learned feature space in an attempt to encompass variability in unseen test distributions, enhance inter-class discrimination, and reduce the generalization error. Empirical evaluations demonstrate that simple weakly labeled data augmentation (Acc: 0.5555 and Acc: 0.6536) is better than baseline non-augmented training (Acc: 0.2885 and Acc: 0.5028) in identifying COVID-19 manifestations as viral pneumonia. Interestingly, adding COVID-19 CXRs to simple weakly labeled augmented training data significantly improves the performance (Acc: 0.7095 and Acc: 0.8889), suggesting that COVID-19, though viral in origin, creates a uniquely different presentation in CXRs compared with other viral pneumonia manifestations.

Adaptive 3D Model-Based Facial Expression Synthesis and Pose Frontalization.

Facial expressions are one of the important non-verbal ways used to understand human emotions during communication. Thus, acquiring and reproducing facial expressions is helpful in analyzing human emotional states. However, owing to complex and subtle facial muscle movements, facial expression modeling from images with face poses is difficult to achieve. To handle this issue, we present a method for acquiring facial expressions from a non-frontal single photograph using a 3D-aided approach. In addition, we propose a contour-fitting method that improves the modeling accuracy by automatically rearranging 3D contour landmarks corresponding to fixed 2D image landmarks. The acquired facial expression input can be parametrically manipulated to create various facial expressions through a blendshape or expression transfer based on the FACS (Facial Action Coding System). To achieve a realistic facial expression synthesis, we propose an exemplar-texture wrinkle synthesis method that extracts and synthesizes appropriate expression wrinkles according to the target expression. To do so, we constructed a wrinkle table of various facial expressions from 400 people. As one of the applications, we proved that the expression-pose synthesis method is suitable for expression-invariant face recognition through a quantitative evaluation, and showed the effectiveness based on a qualitative evaluation. We expect our system to be a benefit to various fields such as face recognition, HCI, and data augmentation for deep learning.

Methods Used in Computer-Aided Diagnosis for Breast Cancer Detection Using Mammograms: A Review.

According to the American Cancer Society's forecasts for 2019, there will be about 268,600 new cases in the United States with invasive breast cancer in women, about 62,930 new noninvasive cases, and about 41,760 death cases from breast cancer. As a result, there is a high demand for breast imaging specialists as indicated in a recent report for the Institute of Medicine and National Research Council. One way to meet this demand is through developing Computer-Aided Diagnosis (CAD) systems for breast cancer detection and diagnosis using mammograms. This study aims to review recent advancements and developments in CAD systems for breast cancer detection and diagnosis using mammograms and to give an overview of the methods used in its steps starting from preprocessing and enhancement step and ending in classification step. The current level of performance for the CAD systems is encouraging but not enough to make CAD systems standalone detection and diagnose clinical systems. Unless the performance of CAD systems enhanced dramatically from its current level by enhancing the existing methods, exploiting new promising methods in pattern recognition like data augmentation in deep learning and exploiting the advances in computational power of computers, CAD systems will continue to be a second opinion clinical procedure.

Increasing Accuracy of Stock Price Pattern Prediction through Data Augmentation for Deep Learning

Increasing Accuracy of Stock Price Pattern Prediction through Data Augmentation for Deep Learning

A survey on Image Data Augmentation for Deep Learning

Deep convolutional neural networks have performed remarkably well on many Computer Vision tasks. However, these networks are heavily reliant on big data to avoid overfitting. Overfitting refers to the phenomenon when a network learns a function with very high variance such as to perfectly model the training data. Unfortunately, many application domains do not have access to big data, such as medical image analysis. This survey focuses on Data Augmentation, a data-space solution to the problem of limited data. Data Augmentation encompasses a suite of techniques that enhance the size and quality of training datasets such that better Deep Learning models can be built using them. The image augmentation algorithms discussed in this survey include geometric transformations, color space augmentations, kernel filters, mixing images, random erasing, feature space augmentation, adversarial training, generative adversarial networks, neural style transfer, and meta-learning. The application of augmentation methods based on GANs are heavily covered in this survey. In addition to augmentation techniques, this paper will briefly discuss other characteristics of Data Augmentation such as test-time augmentation, resolution impact, final dataset size, and curriculum learning. This survey will present existing methods for Data Augmentation, promising developments, and meta-level decisions for implementing Data Augmentation. Readers will understand how Data Augmentation can improve the performance of their models and expand limited datasets to take advantage of the capabilities of big data.

Geostatistical Simulation of Medical Images for Data Augmentation in Deep Learning

Data augmentation, which is the process for creating alternative copies of each sample in a small training data set, is important for extracting deep-learning features for medical image classification problems. However, data augmentation has not been well explored and existing methods are heuristic. In this paper, a geostatistical simulation of images is introduced as a data augmentation approach for extracting deep-learning features from medical images show that are characterized with texture. The stochastic simulation procedure is to generate realizations of an image by modeling its spatial variability through the generation of multiple equiprobable stochastic realizations. The approach employed for the geostatistical simulation is based on the concepts of regionalized variables and kriging formulation to create multiple textural variations of medical images. The experimental results on classifying two medical-image data sets show that the use of geostatistical simulation for extracting deep-learning features with several popular pre-trained deep-learning models (AlexNet, ResNet-50, and GoogLeNet/Inception) provided better accuracy rates and more balanced results in terms of sensitivity and specificity than either with or without the implementation of conventional data augmentation. The proposed approach can be applied to other pre-trained as well as those without data pre-training for effective deep-feature extraction from medical images. The proposed application of the theory of geostatistics for medical image data augmentation in deep learning is original. The novelty of this approach can also be applied to many other types of data that are inherently textural. The geostatistical simulation opens a new door to the development of state-of-the-art artificial intelligence in feature extraction of medical images.

Data Augmentation to Stabilize Image Caption Generation Models in Deep Learning

Automatic image caption generation is a challenging AI problem since it requires utilization of several techniques from different computer science domains such as computer vision and natural language processing. Deep learning techniques have demonstrated outstanding results in many different applications. However, data augmentation in deep learning, which replicates the amount and the variety of training data available for learning models without the burden of collecting new data, is a promising field in machine learning. Generating textual description for a given image is a challenging task for computers. Nowadays, deep learning performs a significant role in the manipulation of visual data with the help of Convolutional Neural Networks (CNN). In this study, CNNs are employed to train prediction models which will help in automatic image caption generation. The proposed method utilizes the concept of data augmentation to overcome the fuzziness of well-known image caption generation models. Flickr8k dataset is used in the experimental work of this study and the BLEU score is applied to evaluate the reliability of the proposed method. The results clearly show the stability of the outcomes generated through the proposed method when compared to others.