The imperative for robust detection mechanisms has grown in the face of increasingly sophisticated Distributed Denial of Service (DDoS) attacks. This paper introduces DDoSBERT, an innovative approach harnessing transformer text classification for DDoS detection. The methodology conducts a detailed exploration of feature selection methods, emphasizing the selection of critical techniques, including Correlation, Mutual Information , and Univariate Feature Selection. Motivated by the dynamic landscape of DDoS attacks, DDoSBERT confronts contemporary challenges such as binary and multi-attack classification and imbalance attack classification. The methodology delves into diverse text transformation techniques for feature selection and employs three transformer classification models : distilbert-base-uncased, prunebert-base-uncased-6-finepruned-w-distil-mnli, and distilbert-base-uncased-finetuned-sst-2-english. Additionally, the paper outlines a comprehensive framework for assessing the importance of features in the context of five DDoS datasets, comprised of APA-DDoS, CRCDDoS2022, DDoS Attack SDN , CIC-DDoS-2019, and BCCC-cPacket-Cloud-DDoS-2024 datasets. The experimental results, rigorously evaluated against relevant benchmarks, affirm the efficacy of DDoSBERT, underscoring its significance in enhancing the resilience of systems against text-based transformation DDoS attacks. The discussion section interprets the results, highlights the implications of the findings, and acknowledges limitations while suggesting avenues for future research.

Adversarial camouflage has garnered attention for its ability to attack object detectors from any viewpoint by covering the entire object’s surface. However, universality and robustness in existing methods often fall short as the transferability aspect is often overlooked, thus restricting their application only to a specific target with limited performance. To address these challenges, we present Adversarial Camouflage for Transferable and Intensive Vehicle Evasion (ACTIVE), a state-of-the-art physical camouflage attack framework designed to generate universal and robust adversarial camouflage capable of concealing any 3D vehicle from detectors. Our framework incorporates innovative techniques to enhance universality and robustness, including a refined texture rendering that enables common texture application to different vehicles without being constrained to a specific texture map, a novel stealth loss that renders the vehicle undetectable, and a smooth and camouflage loss to enhance the naturalness of the adversarial camouflage. Our extensive experiments on 15 different models show that ACTIVE consistently outperforms existing works on various public detectors, including the latest YOLOv7. Notably, our universality evaluations reveal promising transferability to other vehicle classes, tasks (segmentation models), and the real world, not just other vehicles.

To perform adversarial attacks in the physical world, many studies have proposed adversarial camouflage, a method to hide a target object by applying camouflage patterns on 3D object surfaces. For obtaining optimal physical adversarial camouflage, previous studies have utilized the so-called neural renderer, as it supports differentiability. However, existing neural renderers cannot fully represent various real-world transformations due to a lack of control of scene parameters compared to the legacy photo-realistic renderers. In this paper, we propose the Differentiable Transformation Attack (DTA), a framework for generating a robust physical adversarial pattern on a target object to camouflage it against object detection models with a wide range of transformations. It utilizes our novel Differentiable Transformation Network (DTN), which learns the expected transformation of a rendered object when the texture is changed while preserving the original properties of the target object. Using our attack framework, an adversary can gain both the advantages of the legacy photo-realistic renderers including various physical-world transformations and the benefit of white-box access by offering differentiability. Our experiments show that our camouflaged 3D vehicles can successfully evade state-of-the-art object detection models in the photo-realistic environment (i.e., CARLA on Unreal Engine). Furthermore, our demonstration on a scaled Tesla Model 3 proves the applicability and transferability of our method to the real world.