| Work Detail |
The American company Maxar Technologies used satellite images with a resolution of 31 cm to 15.5 cm to identify photovoltaic systems on rooftops in southern Germany. The company claims that its new method allowed 97.8% of solar installations to be identified with a high degree of reliability.
Researchers at US space technology company Maxar Technologies have created a new data set for the detection of residential photovoltaic systems using high-resolution satellite images.
“This dataset can be used independently or in conjunction with larger non-satellite image datasets to produce robust detection models capable of generalizing across image types,” the scientists explained. “It can better support the use of satellite imagery to rapidly detect and monitor residential-scale solar panel installations, allowing researchers and policymakers to address the needs of various applications.”
The data set contains 2,542 photovoltaic systems located in southern Germany. The researchers explained that they chose this region because of the high concentration of residential and commercial photovoltaic systems. After acquiring images of the area, they randomly chose three areas within the selected region and used a computer program to identify the photovoltaic installations.
They manually annotated the individual solar panel objects and, to verify their identification, compared the results with images from Google Earth, which offers higher resolution images. Objects that potentially only had one or two solar modules, adjacent or separate, were considered non-panel objects, such as skylights, vent covers, and chimneys.
“In total, 2,487 objects with solar panels were identified, that is, 97.8%, with a high level of confidence,” the academics stated. “Less than 3.0% of solar panel objects were identified with moderate or low confidence.”
The scientists said the new data set can be used to develop detection models applicable exclusively to satellite images or in conjunction with existing solar panel aerial image data sets to support generalized detection models.
The data set was presented in the study “ A solar panel dataset of very high-resolution satellite imagery to support the Sustainable Development Goals ”, published in scientific data .
Maxar Technologies made both the imaging chips and object tags available online. |
