Navigation and imagery in the orbit, descent, and landing phases during an interplanetary mission require methods that are able to derive the elevation map of a planetary body using remote sensing tools. The authors propose stereovision techniques for this task. An algorithm for correspondence matching, which is one of the crucial steps in automatic terrain modeling, is introduced. It uses well known pyramid-based data structures, but is novel in its direct application of methods from statistical pattern recognition. Feature vectors for correspondence matching and feature selection techniques are used to find optimal features. These include grey-level statistics (mean variance) as well as more sophisticated features derived from operators like local frequency edge gradient or, as an extension, Moravec-, Gabor- or Fourier-features. The applicability of the algorithm in the remote sensing scenario of interplanetary missions is verified using a mockup simulation of the Martian surface.>

The autonomous navigation of a spacecraft during descent and landing using a computer vision system has been recognized as a key to the feasibility of interplanetary missions. One possible approach for navigation is to compute an elevation model and high level description of the planetary body during orbit, and use this information jointly with real-time dynamic range estimation and tracking to navigate the spacecraft. The authors describe a vision system that is used to navigate the spacecraft during descent. The final goal is to position a roving vehicle on a selected landing site and position it on its trajectory. The methods studied should be applicable to the landing phase and later on also in the autonomous motion of the rover on the surface.>

Data and model-driven processes are basic to image understanding systems and correspond preattentive and focal attentive processes, respectively. The authors describe how to develop preattentive and focal attentive processes primed by known object models complementary to efforts on data driven preattentive processes. The techniques used to implement attentional mechanisms are shown to be robust with respect to noise given as changes in illumination, able to reduce crosstalk and able to count outliers.>

The authors describe the major aspects in their transputer-based automatic vision system (TAVS) aiming to implement a scaleable and easily reconfigurable system, in which the mapping of image processing and recognition algorithms to the hardware is facilitated by automatic code generation schemes, separating methodic design and implementation details. The paper presents the system design and underlying hardware architecture first. The authors describe the modules available for iconic image processing and feature extraction and selection. The code generation for the hierarchical statistical decision network is then described, followed by the implementation of the language pi for process allocation. The various system parts were tested in an implementation of real-time wooden board inspection. For this example the authors present details on typical algorithms and how they were implemented on a maintainable and scaleable industrial system.>

A method of 2-D object recognition based on the Moore-Penrose distributed associative memory is presented. Using known relationships between DAMs and regression analysis, the selectivity of the association weights is improved in an iterative way be discarding from further consideration response vectors deemed to be insignificant. Such selectivity allow the system to focus on the significant associations and to reduce crosstalk effects. The same formalism that allows the significance of the association weights to be computed also provides for a reject option. Experiments incorporating the proposed method onto an invariant recognition system prove the feasibility and benefits of the recognition scheme.>