Augmentation Of Images Research Articles

Visual augmentation of live-streaming images in virtual reality to enhance teleoperation of unmanned ground vehicles

First-person view (FPV) technology in virtual reality (VR) can offer in-situ environments in which teleoperators can manipulate unmanned ground vehicles (UGVs). However, non-experts and expert robot teleoperators still have trouble controlling robots remotely in various situations. For example, obstacles are not easy to avoid when teleoperating UGVs in dim, dangerous, and difficult-to-access areas with environmental obstacles, while unstable lighting can cause teleoperators to feel stressed. To support teleoperators’ ability to operate UGVs efficiently, we adopted construction yellow and black lines from our everyday life as a standard design space and customised the Sobel algorithm to develop VR-mediated teleoperations to enhance teleoperators’ performance. Our results show that our approach can improve user performance on avoidance tasks involving static and dynamic obstacles and reduce workload demands and simulator sickness. Our results also demonstrate that with other adjustment combinations (e.g., removing the original image from edge-enhanced images with a blue filter and yellow edges), we can reduce the effect of high-exposure performance in a dark environment on operation accuracy. Our present work can serve as a solid case for using VR to mediate and enhance teleoperation operations with a wider range of applications.

Real-time augmentation of diagnostic nasal endoscopy video using AI-enabled edge computing.

AI-enabled augmentation of nasal endoscopy video images is feasible in the clinical setting. Edge computing hardware can interface with existing nasal endoscopy equipment. Real-time AI performance can achieve an acceptable balance of accuracy and efficiency.

Cross-Modality Person Re-Identification Method with Joint-Modality Generation and Feature Enhancement.

In order to minimize the disparity between visible and infrared modalities and enhance pedestrian feature representation, a cross-modality person re-identification method is proposed, which integrates modality generation and feature enhancement. Specifically, a lightweight network is used for dimension reduction and augmentation of visible images, and intermediate modalities are generated to bridge the gap between visible images and infrared images. The Convolutional Block Attention Module is embedded into the ResNet50 backbone network to selectively emphasize key features sequentially from both channel and spatial dimensions. Additionally, the Gradient Centralization algorithm is introduced into the Stochastic Gradient Descent optimizer to accelerate convergence speed and improve generalization capability of the network model. Experimental results on SYSU-MM01 and RegDB datasets demonstrate that our improved network model achieves significant performance gains, with an increase in Rank-1 accuracy of 7.12% and 6.34%, as well as an improvement in mAP of 4.00% and 6.05%, respectively.

GAN based augmentation using a hybrid loss function for dermoscopy images

Dermatology is the most appropriate field to utilize pattern recognition-based automated techniques for objective, accurate, and rapid diagnosis because diagnosis mainly relies on visual examinations of skin lesions. Recent approaches utilizing deep learning techniques have shown remarkable results in this field. However, they necessitate a substantial quantity of images and the availability of dermoscopy images is often limited. Also, even if enough images are available, their labeling requires expert knowledge and is time-consuming. To overcome these issues, an efficient augmentation approach is needed to expand training datasets from input images. Therefore, in this work, a generative adversarial network has been developed using a new hybrid loss function constructed with traditional loss functions to enhance the generation power of the architecture. Also, the effect of the proposed approach and different generative network-based augmentations, which have been used with dermoscopy images in the literature, on the classification of skin lesions has been investigated. Therefore, the main contributions of this work are: (i) introducing a new generative model for the augmentation of dermoscopy images; (ii) presenting the effect of the proposed model on the classification of the images; (iii) comparative evaluations of the effectiveness of different generative network-based augmentations in the classification of seven forms of skin lesions. The classification accuracy when the proposed augmentation is used is 93.12%, which is higher than its counterparts. Experimental results indicate the significance of augmentation techniques in the classification of skin lesions and the efficiency of the proposed structure in improving the classification accuracy.

Attribute Feature Perturbation-Based Augmentation of SAR Target Data.

Large-scale, diverse, and high-quality data are the basis and key to achieving a good generalization of target detection and recognition algorithms based on deep learning. However, the existing methods for the intelligent augmentation of synthetic aperture radar (SAR) images are confronted with several issues, including training instability, inferior image quality, lack of physical interpretability, etc. To solve the above problems, this paper proposes a feature-level SAR target-data augmentation method. First, an enhanced capsule neural network (CapsNet) is proposed and employed for feature extraction, decoupling the attribute information of input data. Moreover, an attention mechanism-based attribute decoupling framework is used, which is beneficial for achieving a more effective representation of features. After that, the decoupled attribute feature, including amplitude, elevation angle, azimuth angle, and shape, can be perturbed to increase the diversity of features. On this basis, the augmentation of SAR target images is realized by reconstructing the perturbed features. In contrast to the augmentation methods using random noise as input, the proposed method realizes the mapping from the input of known distribution to the change in unknown distribution. This mapping method reduces the correlation distance between the input signal and the augmented data, therefore diminishing the demand for training data. In addition, we combine pixel loss and perceptual loss in the reconstruction process, which improves the quality of the augmented SAR data. The evaluation of the real and augmented images is conducted using four assessment metrics. The images generated by this method achieve a peak signal-to-noise ratio (PSNR) of 21.6845, radiometric resolution (RL) of 3.7114, and dynamic range (DR) of 24.0654. The experimental results demonstrate the superior performance of the proposed method.

Improving Computer-Aided Medical Diagnosis Using Generative Adversarial Networks for Carotid Artery Ultrasound Image Data Augmentation and Classification

Cardiovascular diseases are a leading cause of death globally, making early detection of atherosclerosis critical for prevention. Carotid artery ultrasound imaging is a common diagnostic tool; however, the limited availability of labelled medical images hinders the training of deep learning models. This study examines generative adversarial networks (GANs) for data augmentation and classification of carotid artery Doppler images to improve computer-aided medical diagnosis. Four convolutional neural networks (CNNs) – AlexNet, VGGNet, GoogleNet, and CifarNet – are evaluated for their classification performance on original and extended datasets. AlexNet outperforms the other models, achieving a classification accuracy of 94.18% on the extended dataset. The GAN implementation for data augmentation and overfitting reduction demonstrates the potential of generative models in enhancing the performance of deep learning models in medical image analysis, particularly in the "common artery carotid" class. This research contributes to understanding GANs as a valuable tool for data augmentation and classification in the context of carotid artery ultrasound images.

LARLUS: laparoscopic augmented reality from laparoscopic ultrasound.

This research endeavors to improve tumor localization in minimally invasive surgeries, a challenging task primarily attributable to the absence of tactile feedback and limited visibility. The conventional solution uses laparoscopic ultrasound (LUS) which has a long learning curve and is operator-dependent. The proposed approach involves augmenting LUS images onto laparoscopic images to improve the surgeon's ability to estimate tumor and internal organ anatomy. This augmentation relies on LUS pose estimation and filtering. Experiments conducted with clinical data exhibit successful outcomes in both the registration and augmentation of LUS images onto laparoscopic images. Additionally, noteworthy results are observed in filtering, leading to reduced flickering in augmentations. The outcomes reveal promising results, suggesting the potential of LUS augmentation in surgical images to assist surgeons and serve as a training tool. We have used the LUS probe's shaft to disambiguate the rotational symmetry. However, in the long run, it would be desirable to find more convenient solutions.

Synthetic data augmentation for high-resolution X-ray welding defect detection and classification based on a small number of real samples

Deep learning has become the dominant technology in most computer vision tasks. These methods often rely on a large number of labeled sample datasets for training, and in the field of non-destructive testing of welds in industrial manufacturing, weld images with defects are very scarce, and it is still a challenging challenge to construct high-resolution weld defect datasets that meet the requirements. To overcome this limitation, a new data augmentation method for high-resolution X-ray welding defect classification and synthesis based on a small number of real samples is proposed to realize the data augmentation of industrial nondestructive inspection X-ray film defect images. Firstly, to overcome the scarcity of the weld X-ray defect classification dataset, the weld X-ray defect classification dataset (Weld Defect Classification, WDC) is constructed. Secondly, the performance of 16 common deep classification models on WDC datasets is explored. Then, the images of the real local welding defects and the non-defective weld area are fused at random locations, and two data augmentation modes, (Single Image Single Defect, SISD) and (Single Image Multi Defects, SIMD), can generate defect files and annotation files (Visual Object Classes, VOC) at the same time, which can save a lot of time for manual marking. Finally, compared with the traditional data augmentation method, the proposed method can effectively improve the accuracy of defect detection and generalization, the mAP (Mean Average Precision, mAP) @0.5 of YOLOV8X (You Only Look Once, YOLO) and YOLOV5.6.1X is 66.6% and 72.8%, which provides an effective solution for data sample generation in the industrial field.

Effective augmentation of front opening unified pod filter images

The front opening unified pods (FOUPs) are popular carriers in semi-conductor industry, and their automatic optical inspection relies on the sufficiency of images. Filter papers at the bottom of FOUP require frequent inspection, while obtaining their images is time-consuming. This work thus investigates augmenting images of filter papers using conventional interpolation and a recent generative adversarial network with variations, including numbers of epochs, usage of progressive learning, loss functions, and cross frame calibration. Drawbacks of interpolated images are exhibited. Results of sixteen models are quantitatively evaluated using the Fréchet inception distance (FID). The best FID among all models is fid = 0.102, and the distribution of generated images is close to that of real images. A user study is also conducted to investigate the authenticity of generated images. Our experiments reveal that given a limited number of training images, after proper geometric calibration, images synthesized by a generative adversarial network are qualified for data augmentation and substantially benefit industrial image classification tasks.

Sample Augmentation Method for Side-Scan Sonar Underwater Target Images Based on CBL-sinGAN

The scarcity and difficulty in acquiring Side-scan sonar target images limit the application of deep learning algorithms in Side-scan sonar target detection. At present, there are few amplification methods for Side-scan sonar images, and the amplification image quality is not ideal, which is not suitable for the characteristics of Side-scan sonar images. Addressing the current shortage of sample augmentation methods for Side-scan sonar, this paper proposes a method for augmenting single underwater target images using the CBL-sinGAN network. Firstly, considering the low resolution and monochromatic nature of Side-scan sonar images while balancing training efficiency and image diversity, a sinGAN network is introduced and designed as an eight-layer pyramid structure. Secondly, the Convolutional Block Attention Module (CBAM) is integrated into the network generator to enhance target learning in images while reducing information diffusion. Finally, an L1 loss function is introduced in the network discriminator to ensure training stability and improve the realism of generated images. Experimental results show that the accuracy of shipwreck target detection increased by 4.9% after training with the Side-scan sonar sample dataset augmented by the proposed network. This method effectively retains the style of the images while achieving diversity augmentation of small-sample underwater target images, providing a new approach to improving the construction of underwater target detection models.

AUTOMATED DETECTION OF PARKINSON’S DISEASE BASED ON HYBRID CNN AND QUANTUM MACHINE LEARNING TECHNIQUES IN MRI IMAGES

Parkinson’s disease (PD) is a long-term neurological condition that causes severe neuronal degeneration in the motor cortex. Because of the constraints of medical applications determining the severity of PD, forecasting its development can be difficult. Recently research has focused on artificial intelligence to automatically diagnose PD from MRI images. The proposed study aims (a) to incorporate automated method for augmentation of the MRI images using the deep convolutional generative adversarial networks (DCGAN), (b) to perform feature extraction and classification by employing the pre-trained models such as VGG16, Xception, and InceptionV3 models, (c) to develop a hybrid model by fusing the InceptionV3 features and classification using QSVM model. A total of 60 real-time MRIs of normal (N = 30) and PD (N = 30) patients were included in this study. To increase the dataset, the proposed study employed the DCGAN approach. Thus, the dataset for this study was increased to 1000 (normal and PD) images. Automatic feature extraction and classification were performed utilizing different pre-trained models such as VGG16, Xception, and InceptionV3. Among the pre-trained models InceptionV3 architecture provided high accuracy of 74% compared to other models. The proposed work developed a hybrid model by fusing InceptionV3 features and a quantum support vector machine (QSVM) for the detection of PD and healthy groups. The proposed model attained a prediction accuracy of 87.5% and high precision of 95%, respectively. Additionally, the hybrid architecture provided high recall and F1-measure of 84% and 89%, respectively. The hybrid model provided lowest false negative and false positive of 4 and 1, respectively. Therefore, the hybrid model could be an effective diagnostic tool for the automated prediction of PD.

Deep Learning Approach for Automated Data Augmentation and Multi-class Classification of Pap Smear Images

AI is now used to analyze large datasets, which can subsequently be used to predict outcomes and provide patient insights. In this paper, we aim to address the challenge of insufficient medical data, which in turn hinders ML and AI from being used in medicine. By producing additional and related medical data, data augmentation will make it easier to develop machine learning or deep learning models to automate diagnosis. This study aims to utilize Generative Adversarial Networks to create synthetic images and to develop a model to effectively classify the dataset, including the newly generated images, into NILM, HSIL, LSIL, or SCC, which were the four types of cancer cells provided in the dataset. The LBC dataset for multi-class classification has been classified with a state-of-the-art accuracy of 99.1% in our study. The testing accuracy achieved using traditional methods is 98.8%, whereas using GAN is 99.1%, demonstrating that employing synthetic images created using GANs can be a better augmentation method than conventional ways.

Competitive advantage in healthcare based on augmentation of clinical images with artificial intelligence: case study of the 'Sambias' project

Competitive advantage in healthcare based on augmentation of clinical images with artificial intelligence: case study of the 'Sambias' project

Weakly-supervised learning based automatic augmentation of aerial insulator images

The automatic detection of insulator defects with UAV and CNN-based detectors has become a popular paradigm in recent years. However, insufficient insulator data has always been the bottleneck of detector performance. The existing augmentation method either performs a whole image transformation that lacks new semantic features or generates samples with massive manual annotation. Therefore, this paper proposes an automatic augmentation method called Weakly-Supervised Segmentation Mix (WSSM), where a Foreground Segmentation Network (FSN) is trained under the supervision of the bounding box label to extract the insulators for new sample synthesis. In the FSN training process, UnionMix is designed to generate hard samples based on the pseudo-label, thus facilitating the FSN segmentation ability for insignificant insulator boundaries. Oriented Muti-Instance Loss (OMIL) is proposed to extract supervision from the oriented bounding box so that FSN can be fully trained to handle the diverse angle distribution of insulators. The experiments conducted on the Aerial Insulator Dataset (AID) indicate that the synthesized images of WSSM can achieve stable improvement on multiple mainstream detectors. Both in-domain and cross-domain backgrounds can be used in WSSM to promote the network. For example, the AP of YOLOv5-m can be improved from 68.14 to 69.73 with 2600 COCO images, which exceeds the AP of bassline YOLOv5-l (69.69). To further verify the foreground extraction capability, this paper takes the FSN result as the pseudo-label and trains the instance segmentation network on iSAID. The comparison with existing SOTA methods proves the superior quality of the proposed method.

HFRAS: design of a high-density feature representation model for effective augmentation of satellite images

AbstractEfficiently extracting features from satellite images is crucial for classification and post-processing activities. Many feature representation models have been created for this purpose. However, most of them either increase computational complexity or decrease classification efficiency. The proposed model in this paper initially collects a set of available satellite images and represents them via a hybrid of long short-term memory (LSTM) and gated recurrent unit (GRU) features. These features are processed via an iterative genetic algorithm, identifying optimal augmentation methods for the extracted feature sets. To analyse the efficiency of this optimization process, we model an iterative fitness function that assists in incrementally improving the classification process. The fitness function uses an accuracy & precision-based feedback mechanism, which helps in tuning the hyperparameters of the proposed LSTM & GRU feature extraction process. The suggested model used 100 k images, 60% allocated for training and 20% each designated for validation and testing purposes. The proposed model can increase classification precision by 16.1% and accuracy by 17.1% compared to conventional augmentation strategies. The model also showcased incremental accuracy enhancements for an increasing number of training image sets.

Tomato Plant Leaf Disease Detection Using Convolutional Neural Network

Our country is growing country, India is mostly depended on the agriculture field, and Agriculture is pillar of our country’s economy. Because of this India has to improve in agriculture filed, in many of situations’ farmer face difficulties to detect crop disease in a case large area farm, or late in detection of crop disease in such cases farmer lose their all crop and face the huge loss, to avoid this, in agricultural field is detection of plant diseases is very important. And it is very difficult tasks. In regular normal procedure it requires a more time period and trained person to detect accurate plant disease. In this paper, we proposed effective way for plant disease detection using computer science and machine learning model. Disease transmission from unhealthy plant to all other healthy plants in farm is one of the major damages to crop farm. And these diseases spread like forest fire and have the possible to impact the whole operation if not identified in early on. Now Plant disease detection methods helps to identify infected plants in a very early stage and also help us to identify plant disease in a wide range of area of crops in a cost-effective manner. The aim of this project model is to implement machine learning models, in our proposed system we take a plant leaf image on that leaf images we predict the plant disease using Convolution Neural Networks (CNN) model, in that we build a such a model to predict the plant disease with maximum accuracy and it is for plant disease detection for tomato plant. This machine learning model is analyzing different image metrics pixels data to determine the best performance of network. For that dataset were used around 7around 8016 images we going to use to train the model. We going to use 14 layers CNN models to get better accuracy results. Model consist various layers like convolution, pooling, flatten and dense. Early two layers that is preprocessing and augmentation of images Finally, we get the result of which disease that plant have.

Breast Ultrasound Images Augmentation and Segmentation Using GAN with Identity Block and Modified U-Net 3.

One of the most prevalent diseases affecting women in recent years is breast cancer. Early breast cancer detection can help in the treatment, lower the infection risk, and worsen the results. This paper presents a hybrid approach for augmentation and segmenting breast cancer. The framework contains two main stages: augmentation and segmentation of ultrasound images. The augmentation of the ultrasounds is applied using generative adversarial networks (GAN) with nonlinear identity block, label smoothing, and a new loss function. The segmentation of the ultrasounds applied a modified U-Net 3+. The hybrid approach achieves efficient results in the segmentation and augmentation steps compared with the other available methods for the same task. The modified version of the GAN with the nonlinear identity block overcomes different types of modified GAN in the ultrasound augmentation process, such as speckle GAN, UltraGAN, and deep convolutional GAN. The modified U-Net 3+ also overcomes the different architectures of U-Nets in the segmentation process. The GAN with nonlinear identity blocks achieved an inception score of 14.32 and a Fréchet inception distance of 41.86 in the augmenting process. The GAN with identity achieves a smaller value in Fréchet inception distance (FID) and a bigger value in inception score; these results prove the model's efficiency compared with other versions of GAN in the augmentation process. The modified U-Net 3+ architecture achieved a Dice Score of 95.49% and an Accuracy of 95.67%.

Skin Cancer Classification: A Transfer Learning Approach Using Inception-v3

In the human body, the skin serves as the primary layer of defense for essential organs. However, as a result of ozone layer degradation, exposure to UV radiation, fungal and viral infections. Skin cancer is becoming more common. This study proposes a novel deep learning-based framework for the multi-classification of eight different types of skin cancer. The suggested framework is divided into several steps. The initial phase is the data augmentation of images. In the second step, deep models are fine-tuned. The model is opted, for Inception-v3, and updated their layers. In the third step, The suggested model has been applied to train both fine-tuned on augmented datasets. After optimization, the pre-trained model performs well for classifying skin tumors, with Inception-v3 having accuracy and an F-score of 81% and 81%, respectively.

Data Augmentation of X-ray Images for Automatic Cargo Inspection of Nuclear Items.

As part of establishing a management system to prevent the illegal transfer of nuclear items, automatic nuclear item detection technology is required during customs clearance. However, it is challenging to acquire X-ray images of major nuclear items (e.g., nuclear fuel and gas centrifuges) loaded in cargo with which to train a cargo inspection model. In this work, we propose a new means of data augmentation to alleviate the lack of X-ray training data. The proposed augmentation method generates synthetic X-ray images for the training of semantic segmentation models combining the X-ray images of nuclear items and X-ray cargo background images. To evaluate the effectiveness of the proposed data augmentation technique, we trained representative semantic segmentation models and performed extensive experiments to assess its quantitative and qualitative performance capabilities. Our findings show that multiple item insertions to respond to actual X-ray cargo inspection situations and the resulting occlusion expressions significantly affect the performance of the segmentation models. We believe that this augmentation research will enhance automatic cargo inspections to prevent the illegal transfer of nuclear items at airports and ports.

Generative Pipeline for Data Augmentation of Unconstrained Document Images with Structural and Textural Degradation (Student Abstract)

Generative Pipeline for Data Augmentation of Unconstrained Document Images with Structural and Textural Degradation (Student Abstract)