Abstract
The automatic detection of Center Pivot Irrigation Systems (CPIS) is fundamental for establishing public policies, especially in countries with a growth perspective in this technology, like Brazil. Previous studies to detect CPIS using deep learning used single-date optical images, containing limitations due to seasonal changes and cloud cover. Therefore, this research aimed to detect CPIS using Sentinel-2 multitemporal images (containing six dates) and instance segmentation, considering seasonal variations and different proportions of cloudy images, generalizing the models to detect CPIS in diverse situations. We used a novel augmentation strategy, in which, for each iteration, six images were randomly selected from the time series (from a total of 11 dates) in random order. We evaluated the Mask-RCNN model with the ResNext-101 backbone considering the COCO metrics on six testing sets with different ratios of cloudless ( $ ) and cloudy images ( $>75\%$ ), from six cloudless images and zero cloudy images (6:0) up to one cloudless image and five cloudy images (1:5). We found that using six cloudless images provided the best metrics [80% average precision (AP), 93% AP with a 0.5 intersection over union threshold (AP50)]. However, results were similar (74% AP, 88% AP50) even in extreme scenarios with abundant cloud presence (1:5 ratio). Our method provides a more adaptive and automatic way to map CPIS from time series, significantly reducing interference such as cloud cover, atmospheric effects, shadow, missing data, and lack of contrast with the surrounding vegetation.
Highlights
S TRATEGIES for technological advances in agricultural production are essential to feed the world’s growing population [1], [2]
The model evaluation considered the COCO metrics [67] average precision (AP), AP50, AP75, APs, APm, and APl. These metrics are the most widely used in instance segmentation problems and have proven to be satisfactory to evaluate different models, including the original Mask-RCNN paper [61] and other influential papers on the subject [70]–[73]
The box and mask results are similar, mainly because of the Center Pivot Irrigation Systems (CPIS) round shape, which yields similar Intersection over Union thresholds (IoU) results for the boxes and segmentation masks. This result demonstrates the ability of the deep learning (DL) method to detect features even under conditions of little information in the time series
Summary
S TRATEGIES for technological advances in agricultural production are essential to feed the world’s growing population [1], [2]. Irrigation plays a fundamental role in increasing agricultural productivity and decreasing costs and manual labor, being essential for crops in arid and semi-arid regions. Conflicts over water use increase, requiring governmental agencies to balance the diverse demands from hydroelectric production, irrigation, domestic, and industrial use. In the context of fastgrowing water demand in the agriculture, constant irrigated area monitoring is crucial to predict and minimize current and potential conflicts. The main alternative for assessing the spatial distribution and estimating irrigated areas is the remote sensing monitoring because of its speed, periodicity, cost-effectiveness, and reliable data acquisition. Consistent remote sensing information on irrigation areas contributes to water management, anticipating necessary changes and negative impacts
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
