Person:
Howard,
Ayanna M.
Howard,
Ayanna M.
Permanent Link
Associated Organization(s)
Organizational Unit
Organizational Unit
ORCID
ArchiveSpace Name Record
Publication Search Results
Now showing
1 - 5 of 5
-
ItemA generalized approach to real-time pattern recognition in sensed data(Georgia Institute of Technology, 1999-12) Howard, Ayanna M. ; Padgett, CurtisMany applications that focus on target detection in an image scene develop algorithms specific to the task at hand. These algorithms tend to be dependent on the type of input data used in the application and thus generally fail when transplanted to other detection spaces. We wish to address this data dependency issue and develop a novel technique which autonomously detects, in real time, all target objects embedded in an image scene irrespective of the imagery representation. We accomplish this task using a heirarchical approach in which we use an optimal set of linear filters to reduce the data dimensionality of an image scene and then spatially locate objects in the scene with a neural network classifier. We prove the generality of this approach by applying it to two distinctly separate applications. In the first application, we use our algorithm to detect a specified set of targets for an Automatic Target Recognition (ATR) task. The data for this application is retrieved from two-dimensional camera imagery. In the second task, we address the problem of sub-pixel target detection in a hyperspectral image scene. This data set is represented by hyperspectral pixel bands in which target objects occupy a portion of a hyperspectral pixel. A summarized description of our algorithm is given in the following section.
-
ItemIntelligent learning for deformable object manipulation(Georgia Institute of Technology, 1999-11) Howard, Ayanna M. ; Bekey, George A.The majority of manipulation systems are designed with the assumption that the objects’being handled are rigid and do not deform when grasped. This paper addresses the problem of robotic grasping and manipulation of 3-D deformable objects, such as rubber balls or bags filled with sand.‘ Specifically, we have developed a generalized learning algorithm for handling of 3-D deformable objects in which prior knowledge of object attributes is not required and thus it can be applied to a large class of object types. Our methodology relies on the implementation of two main tasks. Our first task is to calculate deformation characteristics for a non-rigid object represented by a physically-based model. Using nonlinear partial differential equations, we model the particle motion of the deformable object in order to calculate the deformation characteristics. For our second task, we must calculate the minimum force required to successfully lift the deformable object. This minimum lifting force can be learned using a technique called ‘iterative lifting’. Once the deformation characteristics and the associated lifting force term are determined, they are used to train a neural network for extracting the minimum force required for subsequent deformable object manipulation tasks. Our developed algorithm is validated with two sets of experiments. The first experimental results are derived from the implementation of the algorithm in a simulated environment. The second set involves a physical implementation of the technique whose outcome is compared with the simulation results to test the real world validity of the developed methodology.
-
ItemIntelligent Target Detection in Hyperspectral Imagery(Georgia Institute of Technology, 1999-03) Howard, Ayanna M. ; Padgett, Curtis ; Brown, Kenneth R.Many applications that use hyperspectral imagery focus on detection and recognition of targets that occupy a portion of a hyperspectral pixel. We address the problem of sub-pixel target detection by evaluating individual pixels belonging to a hyperspectral image scene. We begin by clustering each pixel into one of n classes based on the minimum distance to a set of n cluster prototypes. These cluster prototypes have previously been identified using a modified clustering algorithm based on prior sensed data. Associated with each cluster is a set of linear filters specifically designed to separate signatures derived from a target embedded in a background pixel from other typical signatures belonging to that cluster. The filters are found using directed principal component analysis which maximally separates the two groups. Each pixel is projected on this set of filters and the result is fed into a trained neural network for classification. A detailed description of our algorithm will be given in this paper. We outline our methodology for generating training and testing data, describe our modified clustering algorithm, explain how the linear filters are designed, and provide details on the neural network classifier. Evaluation of the overall algorithm demonstrates that for pixels with embedded targets taking up no more than 10% of the area, our detection rates approach 99.9% with a false positive rate of less than 10 ⁻⁴.
-
ItemA Multi-Stage Neural Network for Automatic Target Detection(Georgia Institute of Technology, 1998-05) Howard, Ayanna M. ; Padgett, Curtis ; Liebe, Carl ChristianAutomatic Target Recognition (ATR) involves processing two-dimensional images for detecting, classifying, and tracking targets. The first stage in ATR is the detection process. This involves discrimination between target and non-target objects in a scene. In this paper, we shall discuss a novel approach which addresses the target detection process. This method extracts relevant object features utilizing principal component analysis. These extracted features are then presented to a multi-stage neural network which allows an overall increase in detection rate, while decreasing the false positive alarm rate. We shall discuss the techniques involved and present some detection results that have been implemented on the multi-stage neural network.
-
ItemRecursive Learning for Deformable Object Manipulation(Georgia Institute of Technology, 1997-07) Howard, Ayanna M. ; Bekey, George A.This paper presents a generalized approach to handling of 3D deformable objects. Our task is to learn robotic grasping characteristics for a non-rigid object represented by a physically-based model. The model is derived from discretizing the object into a network of interconnected particles and springs. Using Newtonian equations, we model the particle motion of a deformable object and thus calculate the deformation characteristics of the object. These deformation characteristics allow us to learn the required minimum forces necessary to successfully grasp the object and by linking these parameters into a learning table, we can subsequently retrieve the forces necessary to grasp an object presented to the system during run time. This new method of learning is presented and the results of a virtual simulation are shown.