An algorithm for glare detection via photometric, colorimetric, and global positioning 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
Biological and automated vision systems which use digital video for navigation depend on the video to be of sufficient quality in order to extract reliable information that can inform the guidance and/or other decision-making processes. Although systems are available for detection and mitigation of digital distortions (e.g., compression, packet loss), detection and mitigation of natural distortions such as glare, rain, and fog have received much less attention. In this paper, we address the issue of glare detection in a single captured frame. We propose an algorithm which uses a combination of simple and efficient photometric, colorimetric, and GPS features to detect the location and spatial extent of glare within captured images. Specifically, feature maps using lightness, saturation, contrast, and color distance are computed, combined, and then, refined based on the sun's predicted location from the GPS information. In addition, we present a new ground-truth database for glare detection, in which the location, extent, and severity of glare was rated by human subjects for a collection of images. Testing on our ground-truth database revealed that the proposed algorithm can reliably detect the locations and spatial extents of glare sources in a variety of images based on subjective ratings and well-known quantitative measures.
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 | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.004 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 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