OSIRIS-REx AIVA Images

Skip to content | Skip to navigation

Site Actions

Site Sections

Breadcrumbs

Home → OSIRIS-REx → Displaying OSIRIS-REx AIVA Images

OSIRIS-REX ARCHIVAL DOCUMENTATION IMAGING

These image data are research-grade, high-resolution precision photography for archival documentation of the sample returned to Earth from the asteroid Bennu by the OSIRIS-REx Mission. The images are produced by the Advanced Imaging & Visualization of Astromaterials (AIVA) team members Erika Blumenfeld (Creative Lead), Joseph Aebersold (Project Lead), and Christopher Snead (Civil Servant Lead). Blumenfeld is a Transdisciplinary Artist, Photographer, and Researcher as well as a Conservation & Imaging Specialist. Aebersold is an Image Scientist and 3D Imagery Analyst. Snead is the Deputy Sample Curator of OSIRIS-REx, as well as the Hayabusa2 Sample Curator and the Small Particle Lead.

Archival documentation is the first step in any conservation practice with imagery being a means to observe and understand the sample before any scientific analysis is performed. The images can be used in preliminary scientific analysis through image science techniques. Every effort is made to produce images that provide the highest possible image quality we can achieve with a methodical approach to maintain scientific consistency and accuracy. However, due to the highly unusual techniques necessary for imaging within NASA's stringent cleanroom protocols, which have been established to keep the sample pristine for research and posterity, these image data have certain limitations that are important to understand in order to properly analyze and interpret the images.

Description of Imaging Procedure:

AIVA imagery is created using manual high-resolution precision photography (HRPP) and a semi-automated focus stacking procedure. Our process involves the use of a medium format camera system with a full-frame 102-Megapixel sensor and an array of lenses. In all our imaging procedures, the sample must remain inside of a cleanroom nitrogen glovebox cabinet that is made of highly reflective materials and where all our camera and lighting equipment must always be outside of the cabinet. Our goal is to achieve extreme detail in the sample regions, yet the constraints we work within mean we are essentially doing macro photography at telephoto distances, with limited range of motion around the sample, through small windows of glass, and imaging into a mirrored box.

To achieve heightened focus detail across the entire image-scene, we use a focus stacking procedure, where our final products are made from hundreds of images, each with extremely small areas of focus in the field of view. In our post-imaging processing procedures, we work with the full stacks simultaneously in RAW format and manually apply brightness, density, color, and optimization to balance the images in equal measure. Then, we use an automated blending algorithm to process the image stack with additional manual adjustments to reduce processing artifacts and to reveal data that may be otherwise obscured by the software.

The goal of these image data is to provide as much detail in the sample regions as possible to benefit science and research objectives for current and future investigators and the general public. The sample – the rocks and particles from the asteroid Bennu – is overall extremely dark, almost blackish in tone, with areas of very high luster and reflective particles. We process the data to pull out extreme detail in the rocks and particles and therefore the sample will appear lighter in the imagery than it does in real life.

Limitations of Data:

Extremely restrictive requirements necessary to maintain the pristine condition of the sample result in conditions unconducive to imaging. Every effort is taken to mitigate lighting, reflection, shadow, depth of field, vibration, color casting, and other environmental conditions working against perfect images. However, it is not always possible to eliminate the occurrence of these issues within a cleanroom environment. The resulting data is our best attempt to balance imaging artefacts and maintain scientific accuracy while showing as much detail of the sample as reasonably possible.