On-board processing for an infrared observatory

During the past two decades, image compression has developed from a mostly academic Rate-Distortion (R-D) field, into a highly commercial business. Various lossless and lossy image coding techniques have been developed.
This thesis represents an interdisciplinary work between the field of astronomy and digital image processing and brings new aspects into both of the fields. In fact, image compression had its beginning in an American space program for efficient data storage. The goal of this research work is to recognize and develop new methods for space observatories and software tools to incorporate compression in space astronomy standards. While the astronomers benefit from new objective processing and analysis methods and improved efficiency and quality, for technicians a new field of application and research is opened. For validation of the processing results, the case of InfraRed (IR) astronomy has been specifically analyzed.
This work presents a solution for infrared astronomy, where the concept of On-Board Processing (OBP) is introduced for efficient exploitation of the telemetry bandwidth and the budget-limited space observatories.
Indeed, IR astronomy, most commonly, requires space observatories because the Universe cannot be accessed from ground in the full IR range as Earth's atmosphere blocks most IR wavelengths. Thus, IR astronomy is a good candidate to support our investigation. Furthermore, IR imaging with dedicated observations requires specific techniques with a complex semiconductors technology.
Thus, the resulting data is very sensitive to noise, which make the feasibility of our approach challenging.
IR detectors consist, as a rule, of fewer pixels than those for the visual range, but the design of multi-sensor instruments for space applications with special technologies and a harsh radiation environment require high readout rates leading again to larger data volumes.
Therefore, although many applications exist, which generate or manipulate astronomical data (including wavelet-based methods), transmitting image information still faces a bottleneck such that the proposed techniques are often ad-hoc and sometimes inconsistent. One intuitional solution can be the JPEG 2000 standard to achieve the telemetry requirements. Indeed, a large scientific and commercial community is contributing for the development and the improvement of the JPEG 2000 compression codec. We demonstrate with a simple example the limitation of this compression method (JPEG 2000), concerning this astronomical application while OBP outperforms this generic compression codec.
Indeed, thermal IR detector raw data (at wavelengths > 5µm) consist of two constituent contributions: the source signal, and the unwanted background. The background is generally higher than the source signal in the order of several thousands. Therefore, generic quantization (e.g.
case of JPEG 2000) may lead to drop away the relevant information, while a dedicated compression technique using infrared detector knowledge is the only way to optimal performance.
The performance of this solution (OBP) is being measured by considering the compression ratio, result quality and algorithmic complexity. A new complexity analysis and measure is developed for Digital Signal Processor (DSP) architecture. The OBP complexity is evaluated for the Analog Device processor ADSP 21020. The impact of this research on the future of information technology is to develop data delivery systems where communication bandwidth and quality are at a premium and archival storage is a costly endeavor. This new framework has an improved compression ratio and result quality over the best-known pre-existing compression algorithms, which will lead to a reduction of the data traffic for infrared observatories.