Michael A. Sipe

Feature space trajectory methods for active computer vision

We advance new active object recognition algorithms that classify rigid objects and estimate their pose from intensity images. Our algorithms automatically detect if the class or pose of an object is ambiguous in a given image, reposition the sensor as needed, and incorporate data from multiple object views in determining the final object class and pose estimate. A probabilistic feature space trajectory (FST) in a global eigenspace is used to represent 3D distorted views of an object and to estimate the class and pose of an input object. Confidence measures for the class and pose estimates, derived using the probabilistic FST object representation, determine when additional observations are required as well as where the sensor should be positioned to provide the most useful information. We demonstrate the ability to use FSTs constructed from images rendered from computer-aided design models to recognize real objects in real images and present test results for a set of metal machined parts.

Global feature space neural network for active object recognition

We present new test results for our active object recognition algorithms which are based on the feature space trajectory (FST) representation of objects and a neural network processor for computation of distances in global feature space. The algorithms are used to classify, and estimate the pose of objects in different stable rest positions and automatically re-position the camera if the class or pose of an object is ambiguous in a given image. Multiple object views are used in determining both the final object class and pose estimate. An FST in eigenspace is used to represent 3D distorted views of an object. FSTs are constructed using images rendered from solid models. The FSTs are analyzed to determine the camera positions that best resolve ambiguities in class or pose. Real objects are then recognized from intensity images using the FST representations derived from rendered imagery.

Feature space trajectory distorted object representation for classification and pose estimation

The feature space trajectory (FST) neural net is used for clas- sification and pose estimation of the contents of regions of interest. The FST provides an attractive representation of distorted objects that over- comes problems present in other classifiers. We discuss its use in re- jecting clutter inputs, selecting the number and identity of the aspect views most necessary to represent an object, and to distinguish between two objects, temporal image processing, automatic target recognition, and active vision.

Global feature space neural network for active computer vision

We advance new active computer vision algorithms based on the Feature space Trajectory (FST) representations of objects and a neural network processor for computation of distances in global feature space. Our algorithms classify rigid objects and estimate their pose from intensity images. They also indicate how to automatically reposition the sensor if the class or pose of an object is ambiguous from a given viewpoint and they incorporate data from multiple object views in the final object classification. An FST in a global eigenfeature space is used to represent 3D distorted views of an object. Assuming that an observed feature vector consists of Gaussian noise added to a point on the FST, we derive a probability density function for the observation conditioned on the class and pose of the object. Bayesian estimation and hypothesis testing theory are then used to derive approximations to the maximum a posterioriprobability pose estimate and the minimum probability of error classifier. Confidence measures for the class and pose estimates, derived using Bayes theory, determine when additional observations are required, as well as where the sensor should be positioned to provide the most useful information.

Neural net classification of x-ray pistachio nut data

Classification results for agricultural products are presented using a new neural network. This neural network inherently produces higher-order decision surfaces. It achieves this with fewer hidden layer neurons than other classifiers require. This gives better generalization. It uses new techniques to select the number of hidden layer neurons and adaptive algorithms that avoid other such ad hoc parameter selection problems; it allows selection of the best classifier parameters without the need to analyze the test set results. The agriculture case study considered is the inspection and classification of pistachio nuts using x- ray imagery. Present inspection techniques cannot provide good rejection of worm damaged nuts without rejecting too many good nuts. X-ray imagery has the potential to provide 100% inspection of such agricultural products in real time. Only preliminary results are presented, but these indicate the potential to reduce major defects to 2% of the crop with 1% of good nuts rejected. Future image processing techniques that should provide better features to improve performance and allow inspection of a larger variety of nuts are noted. These techniques and variations of them have uses in a number of other agricultural product inspection problems.

Active object recognition using appearance-based representations derived from solid geometric models

We present new test results for our active object recognition algorithms. The algorithms are used to classify and estimate the pose of objects in different stable rest positions and automatically re-position the camera if the class or pose of an object is ambiguous in a given image. Multiple object views are now used in determining both the final object class and pose estimate; previously, multiple views were used for classification only. A feature space trajectory (FST) in eigenspace is used to represent 3D distorted views of an object. FSTs are constructed using images rendered from solid models. We discuss lighting and material settings for photorealism in the rendering process. The FSTs are analyzed to determine the camera positions that best resolve ambiguities. Real objects are recognized from intensity images using the FST representation derived from rendered imagery.

FST-based active object recognition for automated assembly

We demonstrate the use of our active object recognition algorithms in a mechanical assembly task. The algorithms are used to classify and estimate the pose of parts of the assembly in different stable rest positions and automatically re-position the camera if the class or pose of an object is ambiguous in a given image. Multiple object views are used in determining both the final object class and pose estimate. The FSTs are analyzed off-line to determine the camera positions that best resolve ambiguities. We also describe methods for rejecting untrained objects and adding new parts to an existing set of FSTs using a new feature update method.

Nonlinear fusion of Gabor wavelet filters for locating objects, edges, and clutter

We note the use of Gabor wavelet filters to locate objects and edges (for detections of regions of interest in scenes), to locate clutter regions (to reduce false alarms) and to produce distortion invariant features (for object classification). We describe new ways to select Gabor filter parameters, new nonlinear ways to combine Gabor function outputs for improved performance and to reduce the number of filters necessary.

Distortion-invariant object representation and discrimination using an FST neural net

A feature space trajectory (FST) neural net is used to represent distorted versions of an object. Its use in determining the class and pose of an object is addressed with attention to: which aspect views and how many are needed to represent an object, which viewpoint gives the best pose animate and the best classification confidence. New eigen features are also advanced to provide improved results.

Best viewpoints for active vision classification and pose estimation

We advance new active computer vision algorithms that classify objects and estimate their pose from intensity images. Our algorithms automatically reposition the sensor if the class or pose of an object is ambiguous in a given image and incorporate data from multiple object views in determining the final object classification. A feature space trajectory (FST) in a global eigenfeature space is used to represent 3-D distorted views of an object. Assuming that an observed feature vector consists of Gaussian noise added to a point on the FST, we derive a probability density function (PDF) for the observation conditioned on the class and pose of the object. Bayesian estimation and hypothesis testing theory are then used to derive approximations to the maximum a posteriori probability pose estimate and the minimum probability of error classifier. New confidence measures for the class and pose estimates, derived using Bayes theory, determine when additional observations are required as well as where the sensor should be positioned to provide the most useful information.