A Comparative Analysis of Machine Learning Methods for Algal Bloom Detection Using Remote Sensing Images
Bibliographic record
Abstract
Algal blooms are a major environmental challenge for lakes and reservoirs and pose severe threats to water on both aquatic and human health. Conventional algorithms used for al-gal bloom detection based on remote sensing reflectance proved to be effective in some lakes. However, it is still difficult to obtain high accuracy for multiple lakes using single-threshold-based de-tection. Currently, machine learning (ML) algorithms have been applied to pinpoint algal bloom locations with excellent results, but the ability of different ML models to be applied in different lakes is still unknown. This paper presents the performance of al-gal bloom detection with commonly used ML algorithms in Chi-nese eutrophic inland lakes based on Sentinel-2 images. A series of comprehensive tests for accuracy, stability, and robustness were designed for four ML models, including random forest (RF), extreme gradient boosting, artificial neural network, and support vector machine, which were tested in Lake Taihu, Lake Chaohu, and Lake Dianchi. In addition, the index-based methods, includ-ing floating algae index and adjusted floating algae index, were also calculated for comparison with ML methods. The results showed that RF model outperformed other ML models. The com-parison results between the RF model and algal indices revealed that the overall accuracy of RF remained above 0.90. Even with a single lake dataset used as training samples, the RF still main-tained a fairly high accuracy of 0.88 for other lakes. To summa-rize, four ML models demonstrate promising potential for algal bloom detection across different lakes and provide a practical ref-erence for further applications.
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 | 0.007 | 0.011 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.002 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| 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".