Interpretation and Indexing of Multi-modal orbital Images of the Planet Mars – I2-Mars

WP1. Radiometric enhancement and atmospheric/photometric correction of the Martian images to enhance surface features and to improve the compatibility of the different images WP2. Fusion of extremely heterogeneous multi modal images in order to create an augmented view of the Martian surface WP3. Feature extraction from the multi-modal remotely sensed images to extract chemical, mineralogical, physical and geomorphologic information. WP4. Joint interpretation, cross-validation, and indexing.

Expected results
1. better radiometric and atmospheric correction for the datasets available for Mars performed in a coherent manner between the different types of images
2. a better co-registration of the datasets
3. a new generation of high resolution digital elevation models of the martian surfaces from stereoscopic pairs (bayesian approach to estimate the errors)
4. a system for the simulation of Visible and IR planetary hyperspectral images, field irradiances and radiances by radiative transfer methods and ray tracing
5. an algorithm for the inversion of the forward model on radiance hyperspectral images to map Bidirectional Reflectance Factors (BRF)
6 algorithms for the extraction and fusion of structural, thermal and mineralogical information

New Martian investigations, geological as well as climatic, will be made possible by the previous developments.

After 6 months
Rubén Marrero, Sylvain Douté, Antonio Plaza and Jocelyn Chanussot. VALIDATION OF SPECTRAL UNMIXING METHODS USING PHOTOMETRY AND TOPOGRAPHY INFORMATION. IEEE Fifth Workshop on Hyperspectral Image and Signal Processing : Evolution in Remote Sensing, 2013. WHISPERS ’13. , 4 pages.

Information technologies to answer effectively the current challenges in the exploitation of large collections of planetary data.
Successful missions launched by the United States and Europe in recent years have allowed the acquisition of large datasets of images of Mars. These remotely sensed data cover different spectral domains and have huge differences in their spatial and spectral resolutions. Furthermore their volume and complexity have dramatically increased thanks to constant technological improvements. Thus the multi-modal processing, analysis and interpretation of the collections is a challenging task. We propose a multidisciplinary project organised around 5 main work-packages WP1. Radiometric enhancement and atmospheric/photometric correction of the Martian images to enhance surface features and to improve the compatibility of the different images WP2. Fusion of extremely heterogeneous multi modal images in order to create an augmented view of the Martian surface WP3. Feature extraction from the multi-modal remotely sensed images to extract chemical, mineralogical, physical and geomorphologic information. WP4. Joint interpretation, cross-validation, and indexing WP5. geological as well as climatic investigations will be made possible by the previous developments. The four partners share common interests for the analysis of hyperspectral and very high resolution remote sensing data using advanced image and signal processing techniques as well as modeling. They have a unique experience in applying these techniques to address planetary issues in the framework of the exploration of Planet Mars.