No-reference quality metrics for satellite weather images and sky images
| Autor: | Karen Panetta, Sos S. Agaian, Arash Samani |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
business.industry
Computer science Image quality media_common.quotation_subject ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Gaussian blur 020206 networking & telecommunications 02 engineering and technology Sharpening symbols.namesake Gaussian noise Sky Computer Science::Computer Vision and Pattern Recognition Metric (mathematics) 0202 electrical engineering electronic engineering information engineering symbols 020201 artificial intelligence & image processing Computer vision Satellite Artificial intelligence Noise (video) business Physics::Atmospheric and Oceanic Physics media_common Remote sensing |
| Zdroj: | 2017 IEEE International Symposium on Technologies for Homeland Security (HST). |
| DOI: | 10.1109/ths.2017.7943499 |
| Popis: | Satellite weather images and sky images obtained from a satellite or an airborne aircraft are often affected by similar distortions. These images commonly suffer from low contrast, Gaussian noise, blurring artifacts similar to Gaussian smoothing, and sharpening artifacts due to quantization noise. These anomalies are often caused by instability of flying apparatus, compression noise, transmission errors, hardware limitations of imaging sensors, heating noise, or other electronic or physical limitation of moving cameras. To properly analyze these images, they usually need to be enhanced and the enhancement processes require a metric to evaluate the image quality in order to adjust the enhancement parameters. Currently, there is no measure of image quality suitable for evaluating sky images during the image enhancement process. Furthermore, there have been no studies performed to identify the proper quality metric for satellite weather images. In this article, we introduce a novel metric to measure the quality of sky images, which could assist the parameter selection in image enhancement for aviation applications. We also investigate the appropriate configurations of the current image quality measures to be used with the satellite weather images to help parameter selection and optimization during the image enhancement process. Beside weather forecast, these images are sometimes used to assist in recovery and rescue during hurricanes and severe weather. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|