Ship Wake Detection Based on Polarimetric Enhancement and Deep Learning via a Simulated Full-Polarized Dataset
Bibliographic record
Abstract
Ship wake detection provides a supplementary way for ship target detection; however, the detection performances of small and slow targets under high sea state are always unsatisfying. Also, the potential of polarimetric enhancement and deep learning (DL) techniques in ship wake detection still remains to be further discovered. In this article, first, the polarimetric enhancement methods based on the polarimetric whitening filter (PWF) and polarimetric detection optimization filter (PDOF) have been researched. Since the measured full-polarized synthetic aperture radar (SAR) images of ship wake are rather limited and inadequate for DL techniques, a full-polarized wake detection dataset (FPWDD) has been established for the first time. It was constructed based on 6000 simulated polarimetric SAR (PolSAR) images of a typical small surface vehicle, including four kinds of target numbers, five kinds of wind speeds, five kinds of wind directions, five kinds of ship velocities, and 36 kinds of ship heading angles. It is composed of two kinds of images: the full covariance matrixes stored in the TIF format and the images after the Pauli decomposition stored in the RGB format. Furthermore, the ship wake detection has been performed based on the FPWDD and three widely applied target detection DL networks for the HH, HV, VV, PWF, and PDOF channels. Finally, an analysis of the wake detection performances of the five channels has been carried out, which has proved that the polarimetric enhancement methods can obviously enhance the wake detection performance of small surface vehicles, with most evaluation metrics about 10% higher than that of HH, VV, and HV.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".