A method based on improved ant colony algorithm feature selection combined with GWO-SVR model for predicting chlorophyll-a concentration in Wuliangsu Lake
Chlorophyll-a (Chl-a) is an important parameter in water bodies. Due to the complexity of optics in water bodies, it is difficult to accurately predict Chl-a concentrations in water bodies by current traditional methods. In this paper, using Sentinel-2 remote sensing images as the data source combin...
| Publié dans: | Water science and technology : a journal of the International Association on Water Pollution Research. - 1986. - 89(2024), 1 vom: 07. Jan., Seite 20-37 |
|---|---|
| Auteur principal: | |
| Autres auteurs: | , , , , |
| Format: | Article en ligne |
| Langue: | English |
| Publié: |
2024
|
| Accès à la collection: | Water science and technology : a journal of the International Association on Water Pollution Research |
| Sujets: | Journal Article Chlorophyll A YF5Q9EJC8Y Chlorophyll 1406-65-1 Water 059QF0KO0R |
| Résumé: | Chlorophyll-a (Chl-a) is an important parameter in water bodies. Due to the complexity of optics in water bodies, it is difficult to accurately predict Chl-a concentrations in water bodies by current traditional methods. In this paper, using Sentinel-2 remote sensing images as the data source combined with measured data, taking Wuliangsu Lake as the study area, a new intelligent algorithm is proposed for prediction of Chl-a concentration, which uses the adaptive ant colony exhaustive optimization algorithm (A-ACEO) for feature selection and the gray wolf optimization algorithm (GWO) to optimize support vector regression (SVR) to achieve Chl-a concentration prediction. The ant colony optimization algorithm is improved to select remote sensing feature bands for Chl-a concentration by introducing relevant optimization strategies. The GWO-SVR model is built by optimizing SVR using GWO with the selected feature bands as input and comparing it with the traditional SVR model. The results show that the usage of feature bands selected by the presented A-ACEO algorithm as inputs can effectively reduce complexity and improve the prediction performance of the model, under the condition of the same model, which can provide valuable references for monitoring the Chl-a concentration in Wuliangsu Lake |
|---|---|
| Description: | Date Completed 15.01.2024 Date Revised 15.01.2024 published: Print Citation Status MEDLINE |
| ISSN: | 0273-1223 |
| DOI: | 10.2166/wst.2023.410 |