Bridging the Fleet Distribution Data Gap with Satellite Imagery and Deep Learning for GHG Estimation
Bibliographic record
Abstract
Precise quantification of greenhouse gas (GHG) emissions is important for better urban sustainability. Transportation is one of the primary contributing sources of greenhouse gas emissions. To quantify better on-road GHG emissions, it is essential to decode fleet distribution. However, globally, many cities do not have the infrastructure to calculate a fleet distribution. Therefore, there will always be an uncertain error in the on-road GHG emissions estimation. However, very high-resolution satellite data can be helpful to overcome this gap due to its global temporal coverage. Hence, this study proposes a deep learning method, Faster Region-based Convolutional Neural Network (Faster R-CNN), and You Look Only Once (YOLO) based vehicle detection to identify the vehicles and vehicle categories from the very high-resolution satellite data and estimate the fleet distribution. The results show that our model can identify, Passenger Cars, Buses, Trucks, and Large Passenger Cars with the precision of 93.30%, 79.50%, 78.90%, and 81.15%, respectively. We applied this model to temporally available satellite images of Phoenix and calculated the fleet distribution and calculated the FFCO2 based on that fleet distribution and compared it with FFCO2 estimated using CURB dataset fleet distribution. Results show that CURB data-based FFOC2 is over-predicting by 22%, while using fleet distribution estimated by this method, FFCO2 over-predicting by 17% w.r.t VULCAN. These findings demonstrate the effectiveness of satellite-based fleet distribution estimation for improving FFCO₂ quantification in cities lacking robust data infrastructure. This approach provides a scalable and data-driven pathway to more accurate urban emissions modeling, enabling better-informed urban planning and sustainability efforts.
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 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".