Synthetic Aperture Radar-Based Ship Classification Using CNN and Traditional Handcrafted Features
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Ship classification for maritime surveillance is done using satellite-borne synthetic aperture radar (SAR) images that consist of only a small group of bright pixels with noisy background and no proper gradient. Abstract (AB) features obtained using deep learning techniques such as a convolutional neural networks (CNN) alone are insufficient to provide an accurate ship classification. The abstract features (AB) extracted from CNN and common meaningful handcrafted (HC) features such as histogram of oriented gradients (HOG), local binary features (LBF), KAZE features (KF), binary robust invariant scalable keypoints features (BF), and scale-invariant feature transform (SIFT), are combined for ship classification using SAR images. HC features of the two polarizations of SAR images are dimensionally reduced and concatinated. AB and HC features are derived individually from each polarization of the SAR image and the classifier outputs are combined through soft and weighted voting (late fusion). Consolidated AB features are obtained through early fusion of the two polarization images or through mid fusion and then combined with HC features for classification. Experimental results on OpenSARShip dataset demonstrates the effectiveness of fusing HC features with abstract features as the combined feature set outperforms individual (both HC and AB) feature sets. Additionally, it has been observed that both early fusion and late fusion (using weighted voting) yield superior results compared to mid-fusion. The highest accuracy is achieved while combining AB features from early fusion and LBF features, while almost similar accuracy is obtained with weighted voting in late fusion using the same features.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it