Ikhlass Ammar

Ikhlass Ammar

Auto-generated profile, based on publicly available information

Ikhlass Ammar

Computer Science, Faculty of Sciences of Tunis (FST), University of Tunis El Manar, Tunisia | OASIS Laboratory, National Engineering School of Tunis, University of Tunis El Manar, Tunisia

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
1 September 2021 in Journal of Information Security and Applications
Verified
Journal Article
Closed access
IAIkhlass Ammar
Ikhlass Ammar
YTYamen El Touati
Yamen El Touati
MYMoez Yeddes
Moez Yeddes
JMJohn Mullins
John Mullins

3 April 2025 in Engineering, Technology & Applied Science Research
Verified
Journal Article
Open access
PDF file
HZHafedh Mahmoud Zayani
Hafedh Mahmoud Zayani
AKAmani Kachoukh
Amani Kachoukh
RGRefka Ghodhbani
Refka Ghodhbani
NKNouha Khediri
Nouha Khediri
EAEman H. Abd-Elkawy
Eman H. Abd-Elkawy
Taoufik SaidaniTaoufik Saidani
Taoufik Saidani
Accurate identification of cat breeds poses a significant challenge due to subtle inter-breed differences and intra-breed variability. This study leverages YOLOv11, the latest version of the YOLO family, to address these challenges through advanced deep-learning techniques. By training on a dataset consisting of five distinct cat breeds (Persian, Maine Coon, Siamese, Pallas's Cat, and Bengal), the model demonstrates exceptional capability in identifying nuanced breed-specific features. Data augmentation techniques were employed to enhance the dataset's diversity, while various optimization algorithms (Adam, Adamax, NAdam, AdamW, RAdam, RMSProp, and SGD) were evaluated to optimize the performance of the model. Experimental results showed that RAdam and SGD emerged as the top-performing optimizers, achieving an average recall of 96.8%, precision of 97.2%, and mAP50 of 98.1%, significantly outperforming other optimization methods. In contrast, RMSProp exhibited the lowest performance, particularly in terms of precision and mean Average Precision (mAP50). Additionally, data augmentation techniques were applied to enhance the diversity of the dataset, improving the robustness of the model. These findings highlight the effectiveness of YOLOv11 in cat breed classification, with potential applications in pet identification, animal conservation, and veterinary diagnostics.

20 July 2022
Verified
Conference Paper
Closed access
IAIkhlass Ammar
Ikhlass Ammar
YTYamen El Touati
Yamen El Touati
MYMoez Yeddes
Moez Yeddes
The verification of the inclusion problem between two timed languages accepted by timed automata is undecidable in the general case. In the literature, there are several studies proposing decidable solutions for this problem under some restrictions. An interesting work, [4], proposes a theory of timed bounded verification applied to a non-Zeno timed automaton with an upper bound. The solution framework is based on a new discretization technique for timed automata without clocks reset and simple constraints restrictions. The first contribution in this paper aims at proposing an extension of this discretization technique to solve the non-diagonal constraints restriction applied to the timed automata without clocks reset. The proposed solution consists to integrate the Difference Bound Matrices DBM data structure to update the old discretized language. The second contribution focuses on designing all required algorithms relative to the generation of the timed bounded discretized language and the verification of the inclusion problem. In addition, we calculate the time complexity of each proposed algorithm.

6 January 2021 in International Journal of Foundations of Computer Science
Verified
Journal Article
Closed access
PDF file
IAIkhlass Ammar
Ikhlass Ammar
YTYamen El Touati
Yamen El Touati
JMJohn Mullins
John Mullins
MYMoez Yeddes
Moez Yeddes