A Multi-Stage Approach to UAV Detection, Identification, and Tracking Using Region-of-Interest Management and Rate-Adaptive Video Coding
Bibliographic record
Abstract
The drone industry has opened its market to ordinary people, making drones prevalent in daily life. However, safety and security issues have been raised as the number of accidents rises (e.g., losing control and colliding with people or invading secured properties). For safety and security purposes, observers and surveillance systems must be aware of UAVs invading aerial spaces. This paper introduces a UAV tracking system with ROI-based video coding capabilities that can efficiently encode videos with a dynamic coding rate. The proposed system initially uses deep learning-based UAV detection to locate the UAV and determine the ROI surrounding the detected UAVs. Afterward, the ROI is tracked using optical flow, which is relatively light in computational load. Furthermore, our devised module for effective compression, XROI-DCT, is applied to non-ROI regions, so a different coding rate is applied depending on the region during encoding. The proposed UAV tracking system is implemented and evaluated by utilizing videos from YouTube, Kaggle, and a video of 3DR Solo2 taken by the authors. The evaluation verifies that the proposed system can detect and track UAVs significantly faster than YOLOv7 and efficiently encode a video, compressing 70% of the video based on the ROI. Additionally, it can successfully identify the UAV model with a high accuracy of 0.9869 ROC–AUC score.
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 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 | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".