Synthetic ID Card Image Generation for Improving Presentation Attack Detection

Currently, it is ever more common to access online services for activities which formerly required physical attendance. From banking operations to visa applications, a significant number of processes have been digitised, especially since the advent of the COVID-19 pandemic, requiring remote biometric authentication of the user. On the downside, some subjects intend to interfere with the normal operation of remote systems for personal profit by using fake identity documents, such as passports and ID cards. Deep learning solutions to detect such frauds have been presented in the literature. However, due to privacy concerns and the sensitive nature of personal identity documents, developing a dataset with the necessary number of examples for training deep neural networks is challenging. This work explores three methods for synthetically generating ID card images to increase the amount of data while training fraud-detection networks. These methods include computer vision algorithms and Generative Adversarial Networks. Our results indicate that databases can be supplemented with synthetic images without any loss in performance for the print/scan Presentation Attack Instrument Species (PAIS) and a loss in performance of 1% for the screen capture PAIS. Author

Medical Image Augmentation Framework for Resolving Chest X-Ray Data Imbalance

Deep learning techniques, particularly convolutional neural networks (CNNs), have led to an enormous breakthrough in the field of medical imaging. Since the onset of the COVID-19 pandemic, studies based on deep learning systems have shown excellent results for diagnosis through the use of Chest X-rays. However, these methods are data sensitive, and their effectiveness depends on the availability and reliability of data. Models trained on a class-imbalanced dataset tend to be biased towards the majority class. The class-imbalanced datasets can be balanced by augmenting them with synthetically generated images. This paper proposes a method for generating synthetic COVID-19 Chest X-Rays images using Generative Adversarial Networks (GANs). The images generated using the proposed GAN were augmented to three imbalanced datasets of real images. It was observed that the performance of the CNN model for COVID-19 classification improved with the augmented images. Significant improvement was seen in the sensitivity or recall, which is a very critical metric. The sensitivity achieved by adding GAN-generated synthetic images to each of the imbalanced datasets matched the sensitivity levels of the balanced dataset. Hence, the proposed solution can be used to generate images that boost the sensitivity of COVID-19 diagnosis to the level of a balanced dataset. Furthermore, this approach of synthetic data augmentation can be used in other medical classification applications for improved diagnosis recommendations. © 2022 University of Bahrain. All rights reserved.

Generating Synthetic Images & Data to Improve Object Detection in CCTV Footage from Public Transport

Passenger behaviour on public transport has become a source of great interest in the wake of the COVID-19 pandemic. Operators are interested in employing new methods to monitor vehicle utilisation and passenger behaviour. One way to do this is through the use of Machine Learning, using the CCTV footage that is already being captured from the vehicles. However, one of the limitations of Machine Learning is that it requires large amounts of annotated training data, which is not always available. In this poster, we present a technique that uses 3D models to generate synthetic training images/data and discuss the effect that training with the synthetic data had on the Machine Learning models when applied to real-world CCTV footage. © 2022 IEEE.

DC-GAN-based synthetic X-ray images augmentation for increasing the performance of EfficientNet for COVID-19 detection.

Currently, many deep learning models are being used to classify COVID-19 and normal cases from chest X-rays. However, the available data (X-rays) for COVID-19 is limited to train a robust deep-learning model. Researchers have used data augmentation techniques to tackle this issue by increasing the numbers of samples through flipping, translation, and rotation. However, by adopting this strategy, the model compromises for the learning of high-dimensional features for a given problem. Hence, there are high chances of overfitting. In this paper, we used deep-convolutional generative adversarial networks algorithm to address this issue, which generates synthetic images for all the classes (Normal, Pneumonia, and COVID-19). To validate whether the generated images are accurate, we used the k-mean clustering technique with three clusters (Normal, Pneumonia, and COVID-19). We only selected the X-ray images classified in the correct clusters for training. In this way, we formed a synthetic dataset with three classes. The generated dataset was then fed to The EfficientNetB4 for training. The experiments achieved promising results of 95% in terms of area under the curve (AUC). To validate that our network has learned discriminated features associated with lung in the X-rays, we used the Grad-CAM technique to visualize the underlying pattern, which leads the network to its final decision.