Fatma Abdelkefi | Searcharxiv

arXiv2026

MariSat: A Maritime Dataset for Instance Segmentation of Objects in Satellite and Aerial Images

Automated maritime surveillance from satellite and aerial imagery requires large, precisely annotated datasets, which remain scarce for the instance-segmentation task, particularly for small vessels in cluttered port environments. We present MariSat, a new benchmark dataset of 1260 aerial and satellite images covering diverse port and coastal scenes, annotated at the pixel level for eight maritime object classes (sailboat, yacht, jet-ski, fishing boat, cruise ship, military vessel, tugboat and cargo ship). The dataset was produced through a semi-automatic annotation pipeline combining the textpromptable segmentation model SAM 3 with a cascade of geometric and colorimetric post-processing filters, followed by a manual correction and quality-control pass performed with the CVAT annotation platform. We describe the image-collection methodology, the annotation and correction process, and the resulting data organization. We also report class-wise statistics for the training, validation, and test splits. MariSat has already been used to fine-tune and benchmark segmentation and detection models (SAM 3 and YOLO11) for real-time maritime monitoring. We report detailed quantitative and per-class results for both tasks. The MariSat dataset is publicly available on GitHub : https://github.com/amirabbes/P2M-Maritime-Segmentation

cs.CV↗

arXiv2014

PAPR Reduction Scheme In MIMO-OFDM Systems With Efficient Embedded Signaling

Multiple Input Multiple Output (MIMO) Orthogonal Frequency Division Multiplexing (OFDM) is a promising transmission scheme for high performance broadband wireless communications. However, this technique suffers from a major drawback which is the high Peak to Average Power Ratio (PAPR) of the output signals. In order to overcome this issue, several methods that require the transmission of explicit Side Information (SI) bits have been proposed. In fact, the transmitted bits must be channel-encoded as they are particularly critical to the performance of the considered OFDM system. This channel-encoding highly increases the system complexity and also decreases the transmission data rate. For these reasons, we propose in this paper, two robust blind techniques that embed the (SI) implicitly into the OFDM frame. First, we investigate a new technique referred as Blind Space Time Bloc Codes (BSTBC) that is inspired from the conventional Selected Mapping (SLM) approach. This technique banks on an adequate embedded signaling that mainly consist on a specific Space Time Bloc Codes (STBC) patterns and a precoding sequences codebook. Second, in order to improve the signal detection process and the PAPR gain, we propose a new efficient combined Blind SLM-STBC (BSLM-STBC) method. Both methods have the benefit of resulting in an optimized scheme during the signal estimation process that is based on the Max-Log-Maximum A Posteriori (MAP) algorithm. Finally, the obtained performance evaluation results show that our proposed methods result in a spectacular PAPR reduction and furthermore lead to a perfect signal recovery at the receiver side.

cs.IT↗