Deep Learning Model for Unmanned Aerial Vehicle-based Object Detection on Thermal Images
Bibliographic record
Abstract
This research is related to the use of deep learning models on thermal images for object detection using Unmanned Aerial Vehicles (UAVs). Thermal imaging proves to be efficient in environments with minimal light and during nighttime, as it operates based on emitted heat rather than visible light. The ability to detect objects in thermal images enhances surveillance and security measures, particularly in dark conditions. During search and rescue missions, especially in settings with restricted visibility, thermal imaging aids in the identification of individuals or animals by detecting their heat signatures. This facilitates the process of locating them in diverse conditions. Detecting objects at night is challenging due to the lack of illumination. Experiments were conducted using the publicly available HIT-UAV dataset, consisting of 2898 images. This dataset includes several classes such as Person, Car, Bicycle, Other Vehicle, and DontCare. This study proposes the use of both YOLOv5 and YOLOv8 for object detection on this dataset. The YOLOv8 model is the latest model currently available. Both YOLOv5 and YOLOv8 variants were developed by Ultralytics. The experiment used five Yolo models: nano (n), small (s), medium (m), large (l), and extra-large (x). By using YOLOv8m, we achieved a mean Average Precision at IoU threshold 0.5 (mAP@0.5) of 0.855. The performance exceeds that of several previously proposed models, such as YOLOv4-tiny, Faster-RCNN, and YOLOv4, which yielded mAP scores of 0.504, 0.768, and 0.847, respectively.
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) | 0.001 | 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 | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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".