Comprehensive Analysis of Object Detection using RGB Camera for Autonomous Vehicles under Adverse Weather Conditions
Bibliographic record
Abstract
In recent years, object detection and recognition algorithms have achieved acceptable performance in normal weather conditions. However, these algorithms fail to provide the same results in adverse weather conditions which hampers autonomous vehicles widespread utilization. As a result, analyzing the performance of autonomous vehicle sensors in severe environments is very important. This paper evaluates the performance of camera sensors under diverse weather scenarios using the BDD100k dataset. Specifically, we compare the detection accuracy of models trained and tested on individual weather conditions, such as rain, snow, and clear weather including different times like day and night. In addition, a general model is trained on aggregated data to compare the performance of models specifically trained for a weather condition with a generalized model. Metrics such as mean Average Precision, Recall, and Precision are calculated and assessed for 3 types of object groups categorized as small, medium, and large objects. The results demonstrate that weather-specific models perform better in their respective weather conditions, particularly for small and medium objects. However, the general model provides consistent and robust performance across all weather scenarios, making it suitable for general-purpose applications. Additionally, the paper highlights the trade-offs between model generalization and specialization, time of the day, and the impact of object size on detection accuracy. These findings contribute to the development of more reliable and adaptable object detection systems for adverse weather environments.
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 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 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 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".