The impact of training dataset on a vision-based smart road sensor to measure the level of flood on the streets
Bibliographic record
Abstract
With recent advancements in Artificial Intelligence (AI) and decreases in computing costs, early warning systems have become affordable and popular in many different applications such as smart cities, and disaster management. AI and the Internet of Things (IoT) enable the automatic detection of different incidences, such as flash flooding. The IoT solutions at the scene of an emergency, enable capturing various data and sharing in real-time with first responders. The accuracy of flood level estimation is critical for proper emergency response. This is while the accuracy depends on multiple factors, such as camera properties, the brightness of the environment, training data size, etc. In this research, we focus on the impact of training dataset size, the number of classes, and the labelling method of training data on the accuracy of the proposed Machine Learning (ML) method. A vision-based smart road sensor was developed to detect flood occurrences in urban areas, which can be used as a sub-module of an early warning system. We used the portion of cars’ tires that is visible to the camera as an indicator to detect and classify the depth of standing water into four classes: “no-water”, “low-level”, “high-level”, and “water-splash”. A Convolutional Neural Network (CNN) called You Only Look Once (YOLO) was trained for this matter. The results showed that the accuracy of the trained CNN depends on multiple factors. In this study, we evaluated and studied the impact of the training data size, the number of classes, and the labelling method of training data on the accuracy of the early warning system.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.002 | 0.001 |
| 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".