PRO-YOLOv4-tiny: towards more balance between accuracy and speed in the detection of small targets in remotely sensed images
Bibliographic record
Abstract
Aiming at the problem of low accuracy of small target detection in Unmanned Aerial Vehicle (UAV) aerial remote sensing images and limited computing resources of UAV platform, this paper proposes a novel real-time detection algorithm for aerial remote sensing images. Firstly, the Spatial Pyramid Pooling-Fast (SPPF) is used to fuse the global and local features in different receptive fields. Second, we propose the Drone-captured Path Aggregation Network (CPAN) to enrich the semantic features of small targets while keeping the model lightweight. CPAN adds a new detection layer and uses the fusion of deep and shallow feature information to enhance the detection of small targets. At the same time, it uses depthwise separable convolution (DSC) to reduce the number of parameters. Then, Coordinate Attention (CA) is used to capture the cross-channel information with direction-aware and position-aware information. Finally, Decoupled-Head is introduced to make the detection of classification and coordinate regression more robust. We evaluate our model based on the aerial remote sensing dataset. The experimental results show that the proposed method provides a better balance between accuracy and speed than other lightweight networks.
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 | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.001 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 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".