Machine learning approach for spatio-temporal prediction of normalized difference vegetation index towards early estimation of leaf-fall risk on transportation routes
Bibliographic record
Abstract
During the autumn season, deciduous trees shed their leaves due to the process of abscission. Due to different rates and uncommon starting points of abscission, the fall of leaves happens throughout the fall season. Leaf fall is one of the phenology events that can be tracked by monitoring the changes in the values of normalized difference vegetation index (NDVI). This thesis proposes to use deep learning model to predict the NDVI of a vegetation area for an interval of few days in future. The model takes series of images, having multiple channels as the training data. Each image has a channel to hold data for NDVI, which was calculated from satellite images. Other channels are occupied by parameters that are critical to change in NDVI, such as elevation and weather data like temperature, precipitation, etc. These images cover the period of autumn (fall) season, which is from end of summer until the start of winter. It is the period that tracks trees when they have full foliage until the point when no or very less leaves are present on the trees. The approach of using neural network for prediction of vegetation index is very novel. There are numerous applications where the predicted NDVI values could be used for further investigation and analysis. Railway authorities in many countries like USA, Canada, UK, Germany, etc. face the problem of tree leaves falling on railway tracks and roads during the autumn season. When trains and vehicles travel over these leaves, they get pressed into a slippery layer on the surface of tracks and roads respectively. This layer causes road accidents and overshooting of trains, causing the train to cover longer distance to halt. In post-prediction phase, predicted NDVI image with demarked railway track on it can be used to identify track segments impacted with possible leaf fall. Based on it, priority (low, medium and high) of railway track segments can be set and the necessary leaf removal resources can be planned accordingly. Another promising application for NDVI based leaf-fall estimation can be with self-driving vehicles where the autonomous systems could use it as meta-data for choosing paths to drive on. These estimates can help the autonomous system to avoid the routes with low visibility of road markings due to fallen leaves.
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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| 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".