UNSUPERVISED LEARNING FOR ANOMALY DETECTION IN REMOTE SENSING IMAGERY
Bibliographic record
Abstract
Landfill fire is a potential hazard of waste mismanagement, and could occur both on and below the surface of active and closed sites. Timely identification of temperature anomalies is critical in monitoring and detecting landfill fires, to issue warnings that can help extinguish fires at early stages. The overarching objective of this research is to demonstrate the applicability and advantages of remote sensing data, coupled with machine learning techniques, to identify landfill thermal states that can lead to fire, in the absence of onsite observations. This dissertation proposed unsupervised learning techniques, notably variational auto-encoders (VAEs), to identify temperature anomalies from aerial landfill imagery. Twenty years of Landsat satellite observations at a number of landfills were examined for hotspots that may be associated with or leading to subsurface fires. The main contribution of this dissertation is to detect temperature anomalies in landfills using the state-of-the-art unsupervised deep learning technique of VAE based on both model reconstruction error and encoder module feature extraction. Additionally, a simple framework for assessing the health state of the landfill at any given time was established by using the clustering findings to generate a past behavior for each location in the landfill and eventually assigning it to one of four risk categories (No Risk, Low Risk, Moderate Risk, and High Risk). This framework can function as a monitoring system, inferring information such as past landfill temperature profiles, predicting possible heat elevation or smoldering events as new observations are added, and identifying the percentage of each of the four risk categories and how they increase or decrease over the lifetime of the landfill.
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 | 0.002 | 0.006 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| 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".