We describe a novel application of support vector machines and multiscale texture and color invariants to a problem in biological oceanography: the identification of 6 species of bivalve larvae. Our data consists of polarized color images of scallop and other bivalve larvae (between 2 and 17 days old) collected from the ocean by a shipboard optical imaging system of our design. Larvae of scallops, clams, and oysters are small (100 microns) with few distinguishing features when observed under standard light microscopy. However, the use of polarized light with a full wave retardation plate produces a vivid color, birefringence pattern. The patterns display very subtle differences between species, often not discernable to human observers. We show that a soft-margin support vector machine with Gaussian RBF kernel is a good discriminator on a feature set extracted from Gabor wavelet transforms and color distribution angles of each image. By constraining the Gabor center frequencies to be low, the resulting system can attain classification accuracy in excess of 90% for vertically oriented images, and in excess of 80% for randomly oriented images.

This paper describes a novel application of multiscale texture invariants and statistical learning theory to the identification of 6 species of bivalve larvae in biological oceanography. Our data consists of polarized color images of scallop and other bivalve larvae (between 2 and 17 days old) collected from the ocean by a shipboard optical imaging system of our design. Larvae of scallops, clams, and oysters are small (100 microns) with few distinguishing features when observed under standard light microscopy. However, the use of polarized light with a full wave retardation plate produces a vivid color, bi-refringence pattern. The patterns display very subtle differences between species, often not discernable to human observers. Our texture invariants are extracted from Gabor wavelet transforms of each image. We show that by constraining the Gabor center frequencies to be low, the invariants can be optimized to capture subtle inter-species colored-texture differences. When used as input to a support vector machine classifier, the invariants provide classification of larvae to species with accuracy of 85%.

The performance of systems for sensor-based robot motion planning is known to depend heavily on the basic search algorithms used. Typically, these are provable graph search schemes modified into maze search algorithms by taking into account the topology of the physical objects in the workspace-for example, the fact that the boundary of a 2-dimensional object is a simple closed curve. Although several maze search algorithms are already available, more are needed to improve the performance and fill gaps in required properties. This work presents a new algorithm that possesses properties not covered so far, and then assesses its performance. The algorithm allows one to use directional input information (such as azimuth angles) rather than the usual Cartesian coordinates; it also allows the robot to "cut corners" on its way to the target, thus shortening the path.>

This paper is concerned with motion planning for a mobile robot operating in a planar environment with multiple arbitrarily moving obstacles. The robot and obstacles are rigid bodies of arbitrary shapes. All rigid motion-rotation and translation-is considered for the obstacles; the robot can move only by translation. The robot has no advance information about the obstacles' shapes or motion, and obtains information about the surrounding environment from its sensors, in real time. Necessary and sufficient conditions are derived under which collision-free motion of the robot to its target position can be guaranteed. Namely, it is shown that if the maximum velocity of the robot is bounded below by a multiple of the maximum allowable velocity of the obstacles, then a provable motion planning algorithm can be devised. The constant multiple in this bound takes into account the nonconvexity of the obstacles and a bound on the (minimal) distance between any obstacle and the robot's target position.>