From YOLO to VLMs: advancing zero-shot and few-shot detection of wastewater treatment plants using satellite imagery in MENA Region

In regions of the Middle East and North Africa (MENA), there is a high demand for wastewater treatment plants (WWTPs), crucial for sustainable water management. Precise identification of WWTPs from satellite images enables environmental monitoring. Traditional methods like YOLOv8 segmentation requir...

Descripción completa

Detalles Bibliográficos
Autores principales: Premarathna, Akila, Hewageegana, Kanishka, Garcia Andarcia, Mariangel
Formato: Preprint
Lenguaje:Inglés
Publicado: 2025
Materias:
Acceso en línea:https://hdl.handle.net/10568/179492
_version_ 1855516915737296896
author Premarathna, Akila
Hewageegana, Kanishka
Garcia Andarcia, Mariangel
author_browse Garcia Andarcia, Mariangel
Hewageegana, Kanishka
Premarathna, Akila
author_facet Premarathna, Akila
Hewageegana, Kanishka
Garcia Andarcia, Mariangel
author_sort Premarathna, Akila
collection Repository of Agricultural Research Outputs (CGSpace)
description In regions of the Middle East and North Africa (MENA), there is a high demand for wastewater treatment plants (WWTPs), crucial for sustainable water management. Precise identification of WWTPs from satellite images enables environmental monitoring. Traditional methods like YOLOv8 segmentation require extensive manual labeling. But studies indicate that vision-language models (VLMs) are an efficient alternative to achieving equivalent or superior results through inherent reasoning and annotation. This study presents a structured methodology for VLM comparison, divided into zero-shot and few-shot streams specifically to identify WWTPs. The YOLOv8 was trained on a governmental dataset of 83,566 high-resolution satellite images from Egypt, Saudi Arabia, and UAE: ~85% WWTPs (positives), 15% non-WWTPs (negatives). Evaluated VLMs include LLaMA 3.2 Vision, Qwen 2.5 VL, DeepSeek-VL2, Gemma 3, Gemini, and Pixtral 12B (Mistral), used to identify WWTP components such as circular/rectangular tanks, aeration basins and distinguish confounders via expert prompts producing JSON outputs with confidence and descriptions. The dataset comprises 1,207 validated WWTP locations (198 UAE, 354 KSA, 655 Egypt) and equal non-WWTP sites from field/AI data, as 600m×600m GeoTIFF images (Zoom 18, EPSG:4326). Zero-shot evaluations on WWTP images showed several VLMs out-performing YOLOv8's true positive rate, with Gemma-3 highest. Results confirm that VLMs, particularly with zero-shot, can replace YOLOv8 for efficient, annotation-free WWTP classification, enabling scalable remote sensing.
format Preprint
id CGSpace179492
institution CGIAR Consortium
language Inglés
publishDate 2025
publishDateRange 2025
publishDateSort 2025
record_format dspace
spelling CGSpace1794922026-01-08T05:38:55Z From YOLO to VLMs: advancing zero-shot and few-shot detection of wastewater treatment plants using satellite imagery in MENA Region Premarathna, Akila Hewageegana, Kanishka Garcia Andarcia, Mariangel wastewater treatment plants satellite imagery artificial intelligence models remote sensing datasets In regions of the Middle East and North Africa (MENA), there is a high demand for wastewater treatment plants (WWTPs), crucial for sustainable water management. Precise identification of WWTPs from satellite images enables environmental monitoring. Traditional methods like YOLOv8 segmentation require extensive manual labeling. But studies indicate that vision-language models (VLMs) are an efficient alternative to achieving equivalent or superior results through inherent reasoning and annotation. This study presents a structured methodology for VLM comparison, divided into zero-shot and few-shot streams specifically to identify WWTPs. The YOLOv8 was trained on a governmental dataset of 83,566 high-resolution satellite images from Egypt, Saudi Arabia, and UAE: ~85% WWTPs (positives), 15% non-WWTPs (negatives). Evaluated VLMs include LLaMA 3.2 Vision, Qwen 2.5 VL, DeepSeek-VL2, Gemma 3, Gemini, and Pixtral 12B (Mistral), used to identify WWTP components such as circular/rectangular tanks, aeration basins and distinguish confounders via expert prompts producing JSON outputs with confidence and descriptions. The dataset comprises 1,207 validated WWTP locations (198 UAE, 354 KSA, 655 Egypt) and equal non-WWTP sites from field/AI data, as 600m×600m GeoTIFF images (Zoom 18, EPSG:4326). Zero-shot evaluations on WWTP images showed several VLMs out-performing YOLOv8's true positive rate, with Gemma-3 highest. Results confirm that VLMs, particularly with zero-shot, can replace YOLOv8 for efficient, annotation-free WWTP classification, enabling scalable remote sensing. 2025-12-16 2026-01-08T05:35:19Z 2026-01-08T05:35:19Z Preprint https://hdl.handle.net/10568/179492 en Open Access Premarathna, A.; Hewageegana, K.; Garcia Andarcia, M. 2025. From YOLO to VLMs: advancing zero-shot and few-shot detection of wastewater treatment plants using satellite imagery in MENA Region. arXiv. 9p. doi: https://doi.org/10.48550/arXiv.2512.14312
spellingShingle wastewater treatment plants
satellite imagery
artificial intelligence
models
remote sensing
datasets
Premarathna, Akila
Hewageegana, Kanishka
Garcia Andarcia, Mariangel
From YOLO to VLMs: advancing zero-shot and few-shot detection of wastewater treatment plants using satellite imagery in MENA Region
title From YOLO to VLMs: advancing zero-shot and few-shot detection of wastewater treatment plants using satellite imagery in MENA Region
title_full From YOLO to VLMs: advancing zero-shot and few-shot detection of wastewater treatment plants using satellite imagery in MENA Region
title_fullStr From YOLO to VLMs: advancing zero-shot and few-shot detection of wastewater treatment plants using satellite imagery in MENA Region
title_full_unstemmed From YOLO to VLMs: advancing zero-shot and few-shot detection of wastewater treatment plants using satellite imagery in MENA Region
title_short From YOLO to VLMs: advancing zero-shot and few-shot detection of wastewater treatment plants using satellite imagery in MENA Region
title_sort from yolo to vlms advancing zero shot and few shot detection of wastewater treatment plants using satellite imagery in mena region
topic wastewater treatment plants
satellite imagery
artificial intelligence
models
remote sensing
datasets
url https://hdl.handle.net/10568/179492
work_keys_str_mv AT premarathnaakila fromyolotovlmsadvancingzeroshotandfewshotdetectionofwastewatertreatmentplantsusingsatelliteimageryinmenaregion
AT hewageeganakanishka fromyolotovlmsadvancingzeroshotandfewshotdetectionofwastewatertreatmentplantsusingsatelliteimageryinmenaregion
AT garciaandarciamariangel fromyolotovlmsadvancingzeroshotandfewshotdetectionofwastewatertreatmentplantsusingsatelliteimageryinmenaregion