ShuffleDetect: Detecting Adversarial Images against Convolutional Neural Networks

in Applied Sciences (2023), 13(6), 4068

Recently, convolutional neural networks (CNNs) have become the main drivers in many image recognition applications. However, they are vulnerable to adversarial attacks, which can lead to disastrous ... [more ▼]

Recently, convolutional neural networks (CNNs) have become the main drivers in many image recognition applications. However, they are vulnerable to adversarial attacks, which can lead to disastrous consequences. This paper introduces ShuffleDetect as a new and efficient unsupervised method for the detection of adversarial images against trained convolutional neural networks. Its main feature is to split an input image into non-overlapping patches, then swap the patches according to permutations, and count the number of permutations for which the CNN classifies the unshuffled input image and the shuffled image into different categories. The image is declared adversarial if and only if the proportion of such permutations exceeds a certain threshold value. A series of 8 targeted or untargeted attacks was applied on 10 diverse and state-of-the-art ImageNet-trained CNNs, leading to 9500 relevant clean and adversarial images. We assessed the performance of ShuffleDetect intrinsically and compared it with another detector. Experiments show that ShuffleDetect is an easy-to-implement, very fast, and near memory-free detector that achieves high detection rates and low false positive rates. [less ▲]

A strategy creating high-resolution adversarial images against convolutional neural networks and a feasibility study on 10 CNNs

in Journal of Information and Telecommunication (2022), 7(1), 89-119

To perform image recognition, Convolutional Neural Networks (CNNs) assess any image by first resizing it to its input size. In particular, high-resolution images are scaled down, say to 224×244 for CNNs ... [more ▼]

To perform image recognition, Convolutional Neural Networks (CNNs) assess any image by first resizing it to its input size. In particular, high-resolution images are scaled down, say to 224×244 for CNNs trained on ImageNet. So far, existing attacks, aiming at creating an adversarial image that a CNN would misclassify while a human would not notice any difference between the modified and unmodified images, proceed by creating adversarial noise in the 224×244 resized domain and not in the high-resolution domain. The complexity of directly attacking high-resolution images leads to challenges in terms of speed, adversity and visual quality, making these attacks infeasible in practice. We design an indirect attack strategy that lifts to the high-resolution domain any existing attack that works efficiently in the CNN's input size domain. Adversarial noise created via this method is of the same size as the original image. We apply this approach to 10 state-of-the-art CNNs trained on ImageNet, with an evolutionary algorithm-based attack. Our method succeeded in 900 out of 1000 trials to create such adversarial images, that CNNs classify with probability ≥0.55 in the adversarial category. Our indirect attack is the first effective method at creating adversarial images in the high-resolution domain. [less ▲]

Evolutionary Algorithm-based images, humanly indistinguishable and adversarial against Convolutional Neural Networks: efficiency and filter robustness

in IEEE Access (2021)

Emotion Recognition Based on Facial Expressions Using Convolutional Neural Network (CNN)

in Proceedings - 2020 International Conference on Computing, Networking, Telecommunications and Engineering Sciences Applications, CoNTESA 2020 (2020)

Over the last few years, there has been an increasing number of studies about facial emotion recognition because of the importance and the impact that it has in the interaction of humans with computers ... [more ▼]

Over the last few years, there has been an increasing number of studies about facial emotion recognition because of the importance and the impact that it has in the interaction of humans with computers. With the growing number of challenging datasets, the application of deep learning techniques have all become necessary. In this paper, we study the challenges of Emotion Recognition Datasets and we also try different parameters and architectures of the Conventional Neural Networks (CNNs) in order to detect the seven emotions in human faces, such as: anger, fear, disgust, contempt, happiness, sadness and surprise. We have chosen iCV MEFED (Multi-Emotion Facial Expression Dataset) as the main dataset for our study, which is relatively new, interesting and very challenging. © 2020 IEEE. [less ▲]

Detailed Analysis of IRIS Recognition Performance

in 2019 International Conference on Computing, Electronics Communications Engineering (iCCECE) (2019)

Iris recognition is a well-known biometric identification system which distinguishes authentic and imposter individuals based on the features of their irides. It employs stringent statistical analyses of ... [more ▼]

Iris recognition is a well-known biometric identification system which distinguishes authentic and imposter individuals based on the features of their irides. It employs stringent statistical analyses of the features of irides due to the fact that each person has a unique iris, just like a fingerprint. In this work, the approach adopted towards the iris recognition problem is through an exhaustive and careful analysis of the statistical properties of the iris images and the randomness of spurious noise effects. The ability to differentiate two different templates from each other improves with the increase in the number of the degrees of freedom (DOF). The DOF depends on the encoding schemes utilized and moreover, it is hypothesized that the encoding schemes used in themselves could influence the recognition performance. The CASIA (Chinese Academy of Sciences Institute of Automation) version 1 database of iris images used in this study has been modified by the addition of artificial noise in order to simulate practical real life in situ noisy iris capture environments. The classical and state-of-the-art segmentation techniques have been compared, determining whether they are superior to the others under several conditions. The 1D, 2D Gabor filters and the short window implementation were all tested. The conclusion was that the 2D Gabor Filters produce a lower equal error rate (EER), higher accuracy and decidability than by using the one-dimensional log Gabor filter. After modifying the one-dimensional log Gabor filters, a lower EER and higher accuracy was found as the noise level increased. This makes the modified 1D log Gabor Filters a better proposition in noisy conditions. The generated iris templates have a predetermined theoretical value of DOF and from the statistical analysis, an experimental value can be determined. The relation between these values can be used as a metric to compare different databases. [less ▲]

Dynamic virtual bats algorithm (dvba) for minimization of supply chain cost with embedded risk

in 2014 European Modelling Symposium (2014, October 21)

Dynamic Virtual Bats Algorithm (DVBA) is a new optimization algorithm, which is tested on several benchmark functions for global optimization. However it has not been tested on a real world problem yet ... [more ▼]

Dynamic Virtual Bats Algorithm (DVBA) is a new optimization algorithm, which is tested on several benchmark functions for global optimization. However it has not been tested on a real world problem yet. In this paper DVBA has been applied to minimize the supply chain cost with other well known algorithms, Particle Swarm Optimization (PSO), Bat Algorithm (BA), Genetic Algorithm (GA) and Tabu Search (TS). Optimization of supply chain is considered as a real challenge by researchers because of its complexity. Big number of parameters to be controlled and their distributions, interconnections between parameters and dynamism are the main factors that increase the complexity of a supply chain. The result of the case study showed that the DVBA is much superior to other algorithms in terms of accuracy and efficiency. [less ▲]