Jovana Jovanović

Dr. Jovana Jovanović


Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
11 June 2025 in Electronics
Verified
Journal Article
Open access
PDF file
IMIvan Martinović
Ivan Martinović
TSTomás de J. Mateo Sanguino
Tomás de J. Mateo Sanguino
JJJovana Jovanović
Jovana Jovanović
MJMihailo Jovanović
Mihailo Jovanović
Milena DjukanovicMilena Djukanovic
Milena Djukanovic
The increasing deployment of autonomous vehicles (AVs) has exposed critical vulnerabilities in traffic sign classification systems, particularly against adversarial attacks that can compromise safety. This study proposes a dual-purpose defense framework based on convolutional autoencoders to enhance robustness against two prominent white-box attacks: Fast Gradient Sign Method (FGSM) and Projected Gradient Descent (PGD). Experiments on the German Traffic Sign Recognition Benchmark (GTSRB) dataset show that, although these attacks can significantly degrade system performance, the proposed models are capable of partially recovering lost accuracy. Notably, the defense demonstrates strong capabilities in both detecting and reconstructing manipulated traffic signs, even under low-perturbation scenarios. Additionally, a feature-based autoencoder is introduced, which—despite a high false positive rate—achieves perfect detection in critical conditions, a tradeoff considered acceptable in safety-critical contexts. These results highlight the potential of autoencoder-based architectures as a foundation for resilient AV perception while underscoring the need for hybrid models integrating visual-language frameworks for real-time, fail-safe operation.

26 August 2025 in Applied Sciences
Verified
Journal Article
Open access
PDF file
JJJovana Jovanović
Jovana Jovanović
Milena DjukanovicMilena Djukanovic
Milena Djukanovic
LRLuka Radunovic
Luka Radunovic
SRSunčica Rogić
Sunčica Rogić
MJMihailo Jovanović
Mihailo Jovanović
In this study, we investigate a novel post-processing approach combining heat annealing and powder coating to enhance both the mechanical performance and surface finish of polylactic acid (PLA) 3D-printed components. Previous work demonstrated that annealing PLA at temperatures between 100 and 120 °C significantly improves its mechanical properties. Building on this, we explore the effects of applying medium-density fiberboard (MDF) powder coating, which cures at a similar temperature range, to simultaneously improve the material’s surface aesthetics. Test specimens were printed with identical parameters and subjected to heat treatment at 120 °C for varying durations (0 to 15 min, in one-minute intervals). Additional observations included dimensional stability and surface uniformity. The results indicate a clear correlation between post-processing time and improvements in both strength and surface appearance, with optimal outcomes observed between 5 and 8 min of curing. This combined post-processing method provides a cost-effective and accessible way to enhance part performance and aesthetics, thereby expanding the applications of PLA-based additive manufacturing, particularly in functional and design-focused use cases.