This thesis investigates the application of GIST features [13] to the problem of object detection in images. Object detection refers to locating instances of a given object category in an image. It is contrasted with object recognition, which simply decides whether an image contains an object, regardless of the object's location in the image. In much of computer vision literature, object detection uses a "sliding window" approach to finding objects in an image. This requires moving various sizes of windows across an image and running a trained classifier on the visual features of each window. This brute force method can be time consuming. I investigate whether global, easily computed GIST features can be used to classify the size and location of objects in the image to help reduce the number of windows searched before the object is found. Using K–means clustering and Support Vector Machines to classify GIST feature vectors, I find that object size and vertical location can be classified with 73–80% accuracy. These classifications can be used to constrain the search location and window sizes explored by object detection methods.

A long standing debate in computer vision community concerns the link between segmentation and recognition. The question I am trying to answer here is, Does image segmentation as a preprocessing step help image recognition? In spite of a plethora of the literature to the contrary, some authors have suggested that recognition driven by high quality segmentation is the most promising approach in image recognition because the recognition system will see only the relevant features on the object and not see redundant features outside the object (Malisiewicz and Efros 2007; Rabinovich, Vedaldi, and Belongie 2007). This thesis explores the following question: If segmentation precedes recognition, and segments are directly fed to the recognition engine, will it help the recognition machinery? Another question I am trying to address in this thesis is of scalability of recognition systems. Any computer vision system, concept or an algorithm, without exception, if it is to stand the test of time, will have to address the issue of scalability.

Thesis (M.S.)--Massachusetts Institute of Technology, Dept. of Interdisciplinary Science, 1979. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Includes bibliographical references. / by Richard Eugene Hamilton. / M.S.

This thesis describes an approach for object recognition using the chordiogram shape-based descriptor. Global shape representations are highly susceptible to clutter generated due to the background or other irrelevant objects in real-world images. To overcome the problem, we aim to extract precise object shape using superpixel segmentation, perceptual grouping, and connected components. The employed shape descriptor chordiogram is based on geometric relationships of chords generated from the pairs of boundary points of an object. The chordiogram descriptor applies holistic properties of the shape and also proven suitable for object detection and digit recognition mechanisms. Additionally, it is translation invariant and robust to shape deformations. In spite of such excellent properties, chordiogram is not scale-invariant. To this end, we propose scale invariant chordiogram descriptors and intend to achieve a similar performance before and after applying scale invariance. Our experiments show that we achieve similar performance with and without scale invariance for silhouettes and real world object images. We also show experiments at different scales to confirm that we obtain scale invariance for chordiogram.

No description available.

No description available.

No description available.

Thesis. 1978. Ph.D.--Massachusetts Institute of Technology. Dept. of Earth and Planetary Science. / Microfiche copy available in Archives and Science. / Bibliography: leaves 230-233. / Ph.D.

An acoustic pattern recognition system has been developed to identify aerospace acoustic sources. The system is capable of classifying five different types of air and ground sources: jets, propeller planes, helicopters, trains, and wind turbines. The system consists of one microphone for data acquisition, a preprocessor, a feature selector, and a classifier. This thesis presents two new classifiers, one based on an associative memory and one on artificial neural networks, and compares their performance to that of the original classifier developed at VPI&SU (1,2). The acoustic patterns are classified using features that have been calculated from the time and frequency domains. Each of the classifiers undergoes a training period during which a set of known patterns is used to teach the classifier to classify unknown patterns correctly. Once training was completed each classifier is tested using a new set of unknown data. Two different classifier structures were tested, a single level structure and a tree structure. Results show that the single level associative memory and artificial neural network classifiers each identified 90.6 percent of the acoustic sources correctly. The original linear discriminant function single level classifier (1,2) identified 86.7 percent of the sources. The tree structure classifiers classified respectively 90.6 percent, 91.8 percent, and 90.1 percent of the sources correctly. / Master of Science

This thesis describes a machine vision solution to an industrial classification task. The specific problem is the identification of kitchen cabinet frames that travel on a conveyor belt. For these components, it is possible to acquire silhouette images and perform two dimensional image analysis. Corner point locations are used as features which are compared with a database of known frame styles. This thesis describes a novel image acquisition system that utilizes backlighting and a line scan camera. Several corner detection methods are compared and a mathematical formulation of frame comparison is given. A fast database search technique is also presented. The laboratory system has been successfully tested with a set of 27 cabinet frames. / Master of Science