|
Lightweight agile remote sensors have become one type of the most important payloads and were
widely utilized in space reconnaissance and resource survey. These imaging sensors are designed to obtain the high
spatial, temporary and spectral resolution imageries. Key techniques in instrumentation include flexible
maneuvering, advanced imaging control algorithms and integrative measuring techniques, which are closely
correlative or even acting as the bottle-necks for each other. Therefore, mutual restrictive problems must be solved
and optimized. Optical flow is the critical model which to be fully represented in the information transferring as well
as radiation energy flowing in dynamic imaging. For agile sensors, especially with wide-field-of view, imaging
optical flows may distort and deviate seriously when they perform large angle attitude maneuvering imaging. The
phenomena are mainly attributed to the geometrical characteristics of the three-dimensional earth surface as well as
the coupled effects due to the complicated relative motion between the sensor and scene. Under this circumstance,
velocity fields distribute nonlinearly, the imageries may badly be smeared or probably the geometrical structures are
changed since the image velocity matching errors are not having been eliminated perfectly. In this paper, precise
imaging optical flow model is established for agile remote sensors, for which optical flows evolving is factorized by
two forms, which respectively due to translational movement and image shape changing. Moreover, base on that,
agile remote sensors instrumentation was investigated. The main techniques which concern optical flow modeling
include integrative design with lightweight star sensors along with micro inertial measurement units and
corresponding data fusion, the assemblies of focal plane layout and control, imageries post processing for agile
remote sensors etc. Some experiments show that the optical analyzing method is effective to eliminate the
limitations for the performance indexes, and succeeded to be applied for integrative system design. Finally, a
principle prototype of agile remote sensor designed by the method is discussed.
|
