V2C: Venμs Vessel Classification

This dataset comprises raw multispectral imagery acquired from the VENµS (Vegetation and Environment monitoring on a New Micro Satellite)  satellite, featuring a single sensor, the VSSC (VENµS SuperSpectral Camera), which captures high-resolution, multispectral optical images of the Earth's surface. VENµS is an innovative Earth observation mission, resulting from a collaborative effort between France and Israel. The mission serves dual purposes: it conducts both scientific and technological exploration. On the scientific front, VENµS focuses on capturing detailed imagery of the Earth's surface to bolster our comprehension and simulation of terrestrial, vegetative, and ecological dynamics. It aims to propel advancements in various environmental applications, including assessments of water balance, agricultural productivity, and carbon exchange processes

Purpose: This dataset is specifically curated for the development and validation of deep learning models aimed at classifying maritime vessels directly onboard satellites from raw multispectral data. By providing a range of spectral information (12 Bands), the dataset is instrumental in distinguishing between different types of vessels, such as cargo ships and tankers, based on their spectral signatures.

Data Characteristics:

- Data Type: Raw multispectral imagery. The VENµS satellite's VSSC instrument has 12 spectral bands with central wavelengths ranging from 420 nm to 910 nm. These bands have bandwidths ranging from 16-20 nm. The selection of these bands is strategic for characterizing vegetation status and for estimating atmospheric constituents like aerosol optical depth and water vapor content for accurate atmospheric corrections. The sensor's capabilities are optimized for a variety of applications, including vegetation monitoring and studies of coastal areas and inland waters.
- Annotations: The dataset for detection is annotated in the COCO (Common Objects in Context) format, supplemented with AIS (Automatic Identification System) data to provide accurate vessel identification and classification. In this dataset the annotations are just categories.
- Categories: Annotated categories include various vessel types, for example, Cargo, Tanker, Fishing, Passenger, and others, as discernible from the AIS data.

Use Case: The primary use case of this dataset is to facilitate research and development in the field of vessel classification using cutting-edge AI techniques in remote sensing. The dataset is poised to contribute to the enhancement of maritime surveillance, traffic monitoring, and environmental studies related to maritime activities.

Note: For further details on the sensor's technical specifications and configuration, refer to the detailed documentation available in [10.1109/IGARSS.2010.5652087]