Cloud Removal from Multi-Spectral Satellite Images Using Multi-Frequency SAR Data

Cloud cover is recognized as a source of significant loss of information and data quality by many scientific studies. The existence of clouds hinders the extraction of meaningful information because they are a considerable source of uncertainty with regard to the application of any algorithm aiming at the retrieval of land surface parameters. In order to overcome the problem of cloud cover in Earth observation data, the reconstruction of areas beneath clouds can be regarded as a fundamental research topic, especially concerning the visible and infrared regions of the electromagnetic spectrum.

[et_pb_section bb_built=”1″ _builder_version=”3.0.72″ background_color=”#eeeeee” custom_padding_last_edited=”off|phone” custom_padding_phone=”50px||50px|” custom_padding=”50px|50px|50px|50px”][et_pb_row _builder_version=”3.0.72″ background_repeat_1=”no-repeat” background_repeat_2=”no-repeat” custom_padding=”50px||50px|” background_color=”#0c71c3″][et_pb_column type=”1_2″][et_pb_button admin_label=”Downlaod Paper” _builder_version=”3.0.72″ button_text=”Download Paper” button_url=”http://www.mdpi.com/2072-4292/5/6/2973″ url_new_window=”on” button_alignment=”center” background_layout=”light” custom_button=”on” button_letter_spacing=”0″ button_icon_placement=”right” button_letter_spacing_hover=”0″ button_border_width=”4″ /][/et_pb_column][et_pb_column type=”1_2″][et_pb_button admin_label=”Downlaod Source Code” _builder_version=”3.0.72″ button_text=”Download Source Code” button_url=”https://eo-college.org/wp-content/uploads/2017/09/ImageReconstruction-1.0.zip” url_new_window=”on” button_alignment=”center” background_layout=”light” custom_button=”on” button_letter_spacing=”0″ button_icon_placement=”right” button_letter_spacing_hover=”0″ button_border_width=”4″ /][/et_pb_column][/et_pb_row][/et_pb_section][et_pb_section bb_built=”1″ fullwidth=”off” specialty=”off” _builder_version=”3.0.72″ custom_padding=”||20px|”][et_pb_row _builder_version=”3.0.72″ background_repeat_1=”no-repeat” background_repeat_2=”no-repeat” custom_margin=”20px|||”][et_pb_column type=”3_4″][et_pb_text _builder_version=”3.0.72″ background_layout=”light” text_orientation=”justified” border_style=”solid”]

Recent advances in remote sensing are providing numerous opportunities in almost every branch of environmental research. The civil fields of application for multi-spectral remote sensing instruments in Earth observation are diverse, spanning from the monitoring of forests, oceans or urban areas over agricultural applications to the exploration of natural resources, etc. A major prerequisite for a sufficient analysis of Earth observation data is the provision of information that is free from external influences and disturbances.

[/et_pb_text][/et_pb_column][et_pb_column type=”1_4″][et_pb_image _builder_version=”3.0.72″ src=”https://eo-college.org/wp-content/uploads/2017/09/test-area.jpg” show_in_lightbox=”on” url_new_window=”off” use_overlay=”off” sticky=”on” align=”right” always_center_on_mobile=”on” border_style=”solid” force_fullwidth=”off” /][et_pb_text _builder_version=”3.0.72″ background_layout=”light” text_orientation=”left” text_font_size=”12″ border_style=”solid”]

Study Area in central Germany.

[/et_pb_text][/et_pb_column][/et_pb_row][et_pb_row][et_pb_column type=”4_4″][et_pb_text _builder_version=”3.0.72″ background_layout=”light” text_orientation=”justified” border_style=”solid”]

One possible cause of data gaps is cloud cover. Cloud cover is recognized as a source of significant loss of information and data quality by many scientific studies. The existence of clouds hinders the extraction of meaningful information because they are a considerable source of uncertainty with regard to the application of any algorithm aiming at the retrieval of land surface parameters. In order to overcome the problem of cloud cover in Earth observation data, the reconstruction of areas beneath clouds can be regarded as a fundamental research topic, especially concerning the visible and infrared regions of the electromagnetic spectrum. With the growing availability of multi-spectral, hyperspectral and microwave satellite sensors, opportunities for the analysis of complementary data sources arise. Therefore the task of reconstruction of cloud-obscured pixels in multi-spectral images may be regarded as an issue of synergy between these complementary sources.

[/et_pb_text][/et_pb_column][/et_pb_row][et_pb_row][et_pb_column type=”4_4″][et_pb_image _builder_version=”3.0.72″ src=”https://eo-college.org/wp-content/uploads/2017/09/workflow.jpg” show_in_lightbox=”on” url_new_window=”off” use_overlay=”off” sticky=”off” align=”left” always_center_on_mobile=”on” border_style=”solid” force_fullwidth=”off” /][et_pb_text _builder_version=”3.0.72″ background_layout=”light” text_orientation=”center” text_font_size=”12″ border_style=”solid”]

Study Workflow.

[/et_pb_text][/et_pb_column][/et_pb_row][/et_pb_section][et_pb_section bb_built=”1″ fullwidth=”off” specialty=”off” _builder_version=”3.0.72″ custom_padding=”20px|||”][et_pb_row][et_pb_column type=”1_2″][et_pb_text _builder_version=”3.0.72″ background_layout=”light” text_orientation=”justified” border_style=”solid”]

This study presents a method for the reconstruction of pixels contaminated by optical thick clouds in multi-spectral Landsat images using multi-frequency SAR data. A number of reconstruction techniques have already been proposed in the scientific literature. However, all of the existing techniques have certain limitations. In order to overcome these limitations, we expose the Closest Spectral Fit (CSF) method proposed by Meng et al. to a new, synergistic approach using optical and SAR data. Therefore, the term Closest Feature Vector (CFV) is introduced. The technique facilitates an elegant way to avoid radiometric distortions in the course of image reconstruction. Furthermore the cloud cover removal is independent from underlying land cover types and assumptions on seasonality, etc.

[/et_pb_text][/et_pb_column][et_pb_column type=”1_2″][et_pb_image _builder_version=”3.0.72″ src=”https://eo-college.org/wp-content/uploads/2017/09/ed_workflow.png” show_in_lightbox=”on” url_new_window=”off” use_overlay=”off” sticky=”off” align=”left” always_center_on_mobile=”on” border_style=”solid” force_fullwidth=”off” /][/et_pb_column][/et_pb_row][et_pb_row][et_pb_column type=”4_4″][et_pb_text _builder_version=”3.0.72″ background_layout=”light” text_orientation=”justified” border_style=”solid”]

The methodology is applied to mono-temporal, multi-frequency SAR data from TerraSAR-X (X-Band), ERS (C-Band) and ALOS Palsar (L-Band). This represents a way of thinking about Radar data not as foreign, but as additional data source in multi-spectral remote sensing. For the assessment of the image restoration performance, an experimental framework is established and a statistical evaluation protocol is designed. The results show the potential of a synergistic usage of multi-spectral and SAR data to overcome the loss of data due to cloud cover.

[/et_pb_text][/et_pb_column][/et_pb_row][et_pb_row][et_pb_column type=”4_4″][et_pb_image _builder_version=”3.0.72″ src=”https://eo-college.org/wp-content/uploads/2017/09/result.jpg” show_in_lightbox=”on” url_new_window=”off” use_overlay=”off” sticky=”off” align=”left” always_center_on_mobile=”on” border_style=”solid” force_fullwidth=”off” /][et_pb_text _builder_version=”3.0.72″ background_layout=”light” text_orientation=”center” text_font_size=”12″ border_style=”solid”]

Study Results.

[/et_pb_text][/et_pb_column][/et_pb_row][/et_pb_section]