IMAGESEER: NASA IMAGEs for Science, Education, Experimentation and Research

NASA Earth and space images have traditionally been difficult to mine by non-remote sensing researchers; often only available in specialized, non-generic formats, they also do not always provide sufficient context for users unfamiliar with the NASA domain to understand their content and their challenges. This paper describes a new database and its associated website, called IMAGESEER (IMAGEs for Science, Education, Experimentation and Research), that seeks to address these issues. Through a graphical web site for browsing and downloading data, IMAGESEER provides a widely accessible database of NASA-centric, easy to read image data for teaching or validating new Image Processing algorithms. Although other NASA image databases exist, none is focused on the goal of providing validation of new and old image processing algorithms. The first IMAGESEER prototype includes a representative sampling of NASA multispectral and hyperspectral images from several Earth Science instruments, along with a few small tutorials. Image processing techniques are represented with techniques such as cloud detection, image registration, and classification. For each technique, corresponding data are selected from different geographic regions representing different image features (e.g., mountains, urban, water coastal, and agriculture areas). After geo-registration, these images are available in simple common formats such as GeoTIFF and raw formats, along with associated benchmark data.