Automated motion imagery exploitation for surveillance and reconnaissance
Bibliographic record
Abstract
Airborne surveillance and reconnaissance are essential for many military missions. Such capabilities are critical for troop protection, situational awareness, mission planning and others, such as post-operation analysis / damage assessment. Motion imagery gathered from both manned and unmanned platforms provides surveillance and reconnaissance information that can be used for pre- and post-operation analysis, but these sensors can gather large amounts of video data. It is extremely labour-intensive for operators to analyse hours of collected data without the aid of automated tools. At MDA Systems Ltd. (MDA), we have previously developed a suite of automated video exploitation tools that can process airborne video, including mosaicking, change detection and 3D reconstruction, within a GIS framework. The mosaicking tool produces a geo-referenced 2D map from the sequence of video frames. The change detection tool identifies differences between two repeat-pass videos taken of the same terrain. The 3D reconstruction tool creates calibrated geo-referenced photo-realistic 3D models. The key objectives of the on-going project are to improve the robustness, accuracy and speed of these tools, and make them more user-friendly to operational users. Robustness and accuracy are essential to provide actionable intelligence, surveillance and reconnaissance information. Speed is important to reduce operator time on data analysis. We are porting some processor-intensive algorithms to run on a Graphics Processing Unit (GPU) in order to improve throughput. Many aspects of video processing are highly parallel and well-suited for optimization on GPUs, which are now commonly available on computers. Moreover, we are extending the tools to handle video data from various airborne platforms and developing the interface to the Coalition Shared Database (CSD). The CSD server enables the dissemination and storage of data from different sensors among NATO countries. The CSD interface allows operational users to search and retrieve relevant video data for exploitation.
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 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 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) | 0.006 | 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".