Title :
Precise matching of 3-D target models to multisensor data
Author :
Stevens, Mark R. ; Beveridge, J. Ross
Author_Institution :
Dept. of Comput. Sci., Colorado State Univ., Fort Collins, CO, USA
fDate :
1/1/1997 12:00:00 AM
Abstract :
This paper presents a three-dimensional (3-D) model-based ATR algorithm that operates simultaneously on imagery from three heterogeneous, approximately boresight aligned sensors. An iterative search matches models to range and optical imagery by repeatedly predicting detectable features, measuring support for these features in the imagery, and adjusting the transformations relating the target to the sensors in order to improve the match. The result is a locally optimal and globally consistent set of 3-D transformations that precisely relate the best matching target features to combined range, IR, and color images. Results show the multisensor algorithm recovers 3-D target pose more accurately than does a traditional single-sensor algorithm. Errors in registration between images are also corrected during matching. The intended application is imaging from semiautonomous military scout vehicles
Keywords :
error correction; image colour analysis; image matching; image reconstruction; image registration; iterative methods; military systems; search problems; sensor fusion; 3-D target models; 3-D target pose; 3-D transformations; IR images; best matching target features; color images; detectable features; heterogeneous approximately boresight aligned sensors; iterative search; multisensor algorithm; multisensor data; optical imagery; range imagery; registration errors correction; semiautonomous military scout vehicles; three-dimensional model-based ATR algorithm; Color; Computer vision; Error correction; Image sensors; Iterative algorithms; Optical imaging; Optical sensors; Predictive models; Sensor phenomena and characterization; Vehicles;
Journal_Title :
Image Processing, IEEE Transactions on
