Sensor Transforms for Invariant Image Enhancement
Bibliographic record
Abstract
The invariant image [1, 2] formed from an RGB image taken under light that can be approximated as Planckian solves the colour constancy problem at a single pixel. The invariant is a very useful tool for possible use in a large number of computer vision problems, such as removal of shadows from images [3]. This image is formed by projecting log-log chromaticity coordinates into a 1D direction determined by a calibration of the imaging camera. The invariant can be formed whether or not gammacorrection is applied to images and thus can work for ordinary webcam images, for example, once a self-calibration is carried out [3]. As such, the invariant image is an important new mechanism for image understanding. Since the resulting greyscale image is approximately independent of illumination, it is impervious to lighting change and hence to the presence of shadows. However, in forming the invariant image, it can sometimes happen that shadows are not completely removed. Here, we consider the problem of simple matrixing of sensor values so that the resulting invariant image is improved. To do so, we consider the calibration images and apply an optimization routine for establishing a 3 × 3 matrix to apply to the sensors, prior to forming the invariant, with an eye to improving lighting invariance. We find that an optimization does indeed improve the invariant. The resulting image generally has smaller entropy value because the invariant value is smoothed out across former shadow boundaries; thus the new invariant more smoothly captures the underlying intrinsic reflectance properties in the scene.
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 | 0.000 | 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 | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| Insufficient payload (model declined to judge) | 0.007 | 0.002 |
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".