A short wave infrared hyperspectral imager for landmine detection
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
DRDC Suffield and Itres Research have jointly investigated the use of visible and infrared hyperspectral imaging for landmine detection since 1988. There has been considerable success detecting surface-laid landmines by classification of their visible/near infrared (VNIR - 400 to 1000 nm wavelength) spectral signatures, but it has not been possible to find VNIR spectral characteristics that would generically distinguish anthropogenic objects from natural features such as rocks, vegetation, soil, etc. Preliminary studies in 1998 suggested that it might be possible to develop such a generic classifier in the short wave infrared (SWIR) and that detection performance might improve. Because of a lack of available SWIR hyperspectral imagers with adequate performance for mine detection, a prototype pushbroom SWIR hyperspectral imager was developed and completed in summer 2002. The now commercially available instrument, <i>sasi</i>, has 160 bands over a spectral range of 850 to 2450 nm, signal to noise ratio of 400:1 with f/1.8 fore-optics, and 600 pixels over a 37.7° field of view. A number of mission flights have been carried out and excellent imagery obtained. In October 2003, Itres and DRDC Suffield personnel obtained field SWIR hyperspectral imagery in the DRDC Suffield Mine Pen of numerous surface-laid mines, one buried mine, other surface-laid human-made items, background materials and people from a horizontally scanning personnel-lift at an altitude of roughly 5 m. Preliminary indications are that a simple generic classification decision boundary should be able to distinguish surface-laid landmines from many human-made artifacts and natural materials. The buried mine was not detected, but the mine had been buried for several years and hence there would be no residual surface disturbance. Furthermore, the small sample size and limited observation time make it difficult to generalize about SWIR performance for buried mines. The instrument is described and the preliminary results of the trial, planned improvements and future research are discussed.
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 | 0.001 |
| 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