Showing papers on "Feature (computer vision) published in 1997"

Feature selection: evaluation, application, and small sample performance

Abstract: A large number of algorithms have been proposed for feature subset selection. Our experimental results show that the sequential forward floating selection algorithm, proposed by Pudil et al. (1994), dominates the other algorithms tested. We study the problem of choosing an optimal feature set for land use classification based on SAR satellite images using four different texture models. Pooling features derived from different texture models, followed by a feature selection results in a substantial improvement in the classification accuracy. We also illustrate the dangers of using feature selection in small sample size situations.

Image indexing using color correlograms

Abstract: We define a new image feature called the color correlogram and use it for image indexing and comparison. This feature distills the spatial correlation of colors, and is both effective and inexpensive for content-based image retrieval. The correlogram robustly tolerates large changes in appearance and shape caused by changes in viewing positions, camera zooms, etc. Experimental evidence suggests that this new feature outperforms not only the traditional color histogram method but also the recently proposed histogram refinement methods for image indexing/retrieval.

A geometric snake model for segmentation of medical imagery

Abstract: We employ the new geometric active contour models, previously formulated, for edge detection and segmentation of magnetic resonance imaging (MRI), computed tomography (CT), and ultrasound medical imagery. Our method is based on defining feature-based metrics on a given image which in turn leads to a novel snake paradigm in which the feature of interest may be considered to lie at the bottom of a potential well. Thus, the snake is attracted very quickly and efficiently to the desired feature.

Geometric hashing: an overview

Abstract: Geometric hashing, a technique originally developed in computer vision for matching geometric features against a database of such features, finds use in a number of other areas. Matching is possible even when the recognizable database objects have undergone transformations or when only partial information is present. The technique is highly efficient and of low polynomial complexity.

A paraperspective factorization method for shape and motion recovery

Abstract: The factorization method, first developed by Tomasi and Kanade (1992), recovers both the shape of an object and its motion from a sequence of images, using many images and tracking many feature points to obtain highly redundant feature position information. The method robustly processes the feature trajectory information using singular value decomposition (SVD), taking advantage of the linear algebraic properties of orthographic projection. However, an orthographic formulation limits the range of motions the method can accommodate. Paraperspective projection, first introduced by Ohta et al. (1981), is a projection model that closely approximates perspective projection by modeling several effects not modeled under orthographic projection, while retaining linear algebraic properties. Our paraperspective factorization method can be applied to a much wider range of motion scenarios, including image sequences containing motion toward the camera and aerial image sequences of terrain taken from a low-altitude airplane.

Feature-oriented programming : A fresh look at objects

Abstract: We propose a new model for flexible composition of objects from a set of features. Features are similar to (abstract) subclasses, but only provide the core functionality of a (sub)class. Overwriting other methods is viewed as resolving feature interactions and is specified separately for two features at a time. This programming model allows to compose features (almost) freely in a way which generalizes inheritance and aggregation. For a set of n features, an exponential number of different feature combinations is possible, assuming a quadratic number of interaction resolutions. We present the feature model as an extension of Java and give two translations to Java, one via inheritance and the other via aggregation. We further discuss parameterized features, which work nicely with our feature model and can be translated into Pizza, an extension of Java.

Feature-oriented programming: A fresh look at objects

Abstract: We propose a new model for flexible composition of objects from a set of features Features are similar to (abstract) subclasses, but only provide the core functionality of a (sub)class Overwriting other methods is viewed as resolving feature interactions and is specified separately for two features at a time This programming model allows to compose features (almost) freely in a way which generalizes inheritance and aggregation For a set of n features, an exponential number of different feature combinations is possible, assuming a quadratic number of interaction resolutions We present the feature model as an extension of Java and give two translations to Java, one via inheritance and the other via aggregation We further discuss parameterized features, which work nicely with our feature model and can be translated into Pizza, an extension of Java

Acoustic signature recognition and identification

Abstract: An acoustic signature recognition and identification system receives signals from a sensor placed on a designated piece of equipment. The acoustic data is digitized and processed, via a Fast Fourier Transform routine, to create a spectrogram image of frequency versus time. The spectrogram image is then normalized to permit acoustic pattern recognition regardless of the surrounding environment or magnitude of the acoustic signal. A feature extractor then detects, tracks and characterizes the lines which form the spectrogram. Specifically, the lines are detected via a KY process that is applied to each pixel in the line. A blob coloring process then groups spatially connected pixels into a single signal object. The harmonic content of the lines is then determined and compared with stored templates of known acoustic signatures to ascertain the type of machinery. An alert is then generated in response to the recognized and identified machinery.

Synchronization of two Lorenz systems using active control

Abstract: Using techniques from active control theory, we demonstrate that a coupled Lorenz system can be synchronized The synchronization is verified using the Simulink feature in MATLAB

Pose determination and tracking by matching 3D objects to a 2D sensor

Abstract: An improved method of pose determination and tracking does away with conventional segmentation while taking advantage of multi-degree-of-freedom numerical fitting or match filtering as opposed to a syntactic segment or feature oriented combinatorial match. The technique may be used to improve image database query based on object shape descriptors by allowing the user to request images from a database or video sequence which contain a key object described by a geometric description that the user designates or supplies. The approach is also applicable to target or object acquisition and tracking based on the matching of one or a set of object shape data structures.

Method for segmenting chest CT image data using an anatomical model: preliminary results

Abstract: Presents an automated, knowledge-based method for segmenting chest computed tomography (CT) datasets. Anatomical knowledge including expected volume, shape, relative position, and X-ray attenuation of organs provides feature constraints that guide the segmentation process. Knowledge is represented at a high level using an explicit anatomical model. The model is stored in a frame-based semantic network and anatomical variability is incorporated using fuzzy sets. A blackboard architecture permits the data representation and processing algorithms in the model domain to be independent of those in the image domain. Knowledge-constrained segmentation routines extract contiguous three-dimensional (3-D) sets of voxels, and their feature-space representations are posted on the blackboard. An inference engine uses fuzzy logic to match image to model objects based on the feature constraints. Strict separation of model and image domains allows for systematic extension of the knowledge base. In preliminary experiments, the method has been applied to a small number of thoracic CT datasets. Based on subjective visual assessment by experienced thoracic radiologists, basic anatomic structures such as the lungs, central tracheobronchial tree, chest wall, and mediastinum were successfully segmented. To demonstrate the extensibility of the system, knowledge was added to represent the more complex anatomy of lung lesions in contact with vessels or the chest wall. Visual inspection of these segmented lesions was also favorable. These preliminary results suggest that use of expert knowledge provides an increased level of automation compared with low-level segmentation techniques. Moreover, the knowledge-based approach may better discriminate between structures of similar attenuation and anatomic contiguity. Further validation is required.

Parametrized structure from motion for 3D adaptive feedback tracking of faces

Abstract: A real-time system is described for automatically detecting, modeling and tracking faces in 3D. A closed loop approach is proposed which utilizes structure from motion to generate a 3D model of a face and then feed back the estimated structure to constrain feature tracking in the next frame. The system initializes by using skin classification, symmetry operations, 3D warping and eigenfaces to find a face. Feature trajectories are then computed by SSD or correlation-based tracking. The trajectories are simultaneously processed by an extended Kalman filter to stably recover 3D structure, camera geometry and facial pose. Adaptively weighted estimation is used in this filter by modeling the noise characteristics of the 2D image patch tracking technique. In addition, the structural estimate is constrained by using parametrized models of facial structure (eigen-heads). The Kalman filter's estimate of the 3D state and motion of the face predicts the trajectory of the features which constrains the search space for the next frame in the video sequence. The feature tracking and Kalman filtering closed loop system operates at 25 Hz.

Feature subset selection: a correlation based filter approach

Abstract: Recent work has shown that feature subset selection can have a positive affect on the performance of machine learning algorithms. Some algorithms can be slowed or their performance irrelevant or redundant to the learning task. Feature subset selection, then, is a method for enhancing the performance of learning algorithms, reducing the hypothesis search space, and, in some cases, reducing the storage requirement. This paper describes a feature subset selector that uses a correlation based evaluates its effectiveness with three common ML algorithms: a decision tree inducer (C4.5), a naive Bayes classifier, and an instance based learner (IB1). Experiments using a number of standard data sets drawn from real and artificial domains are presented. Feature subset selection gave significant improvement for all three algorithms; C4.5 generated smaller decision trees.

Feature Article—Genetic Algorithms for the Operations Researcher

Abstract: Genetic algorithms have become increasingly popular as a means of solving hard combinatorial optimization problems of the type familiar in operations research. This feature article will consider what genetic algorithms have achieved in this area, discuss some of the factors that influence their success or failure, and offer a guide for operations researchers who want to get the best out of them.

Semi-Automatic Linear Feature Extraction by Dynamic Programming and LSB-Snakes

Abstract: This paper deals with semi-automatic linear feature extraction from digital images for GIS data capture, where the identification task is pe$ormed manually on a single image, while a special automatic digital module performs the high precision feature tracking in two-dimensional (2-0) image space or even three-dimensional (3-0) object space. A human operator identifies the object from an on-screen display of a digital image, selects the particular class this object belongs to, and provides a very few coarsely distributed seed points. subseq;ently, with th;?sk seed as an approximation of the ~osition and sham the linear feature will be extracted automatically by either a dynamic programming approach or by LSB-S~~~~S [Least-Squares E-spline Snakes). With dynamic programming, the optimization problem is set up as a discrete multistage decision process and is solved by a "timedelayed" algorithm. It ensures global optimality, is numerically stable, and allows for hard constraints to be enforced on the solution. In the least-squares approach, we combine three types of observation equations, one radiometric, formulating the matching of a generic object model with image data, and two that express the internal geometric constraints of a curve and the location of operator-given seed points. The solution is obtained by solving a pair of independent normal equations to estimate the parameters of the spline curve. Both techniques can be used in a monoplotting mode, which combines one image with its underlying DTM. The LSB-S~~~~S approach is also implemented in a multi-image mode, which uses multiple images simultaneously and provides for a robust and mathematically sound full 3D approach. These techniques are not restricted to aerial images. They can be applied to satellite and close-range images as well. The issues related to the mathematical modeling of the proposed methods are discussed and experimental results are shown in this paper too.

Automatic feature detection and age classification of human faces in digital images

Abstract: The invention includes a four step process for automatically finding facial images of a human face in an electronically digitized image (for example, taken by a video-camera), and classifying the age of the person (associated with the face) into an age category. For example three age categories: a baby(up to approximately age 3), a junior person(above age 3 to approximately age forty), and a senior adult (over forty years old). Categories can be further subdivided whereas every three years could be a further age category. Step 1 of the process is to find facial features of the digital image encompassing the chin, sides of the face, virtual top of the head, eyes, mouth and nose of the image. Step 2 is to compute the facial feature ratios of the facial features ratios of the facial features found in Step 1. Step 3 is to compute a wrinkle analysis of the image. Step 4 is to combine the previous two steps to categorize age of the facial image. The invention can locate and detect facial images for age classification from digital camera images and computerized generated images. The invention can be practiced in areas such as population statistic gathering for patrons at entertainment/amusement parks, television viewer ratings. Furthermore, the invention has utility in automated security/surveillance systems, demographic studies, safety monitoring systems, computer human-interface operations and automated photography. The latter to allow for point and shoot focus on specific individuals as a function of their age classification.

The interaction of spatial and object pathways: Evidence from balint's syndrome

Abstract: An earlier report described a patient (RM) with bilateral parietal damage who showed severe binding problems between shape and color and shape and size (Friedman-Hill, Robertson, & Treisman, 1995). When shown two different-colored letters, RM reported a large number of illusory conjunctions (ICs) combining the shape of one letter with the color of the other, even when he was looking directly at one of them and had as long as 10 sec to respond. The lesions also produced severe deficits in locating and reaching for objects, and difficulty in seeing more than one object at a time, resulting in a neuropsychological diagnosis of Balint's syndrome or dorsal simultanagnosia. The pattern of deficits supported predictions of Treisman's Feature Integration Theory (FIT) that the loss of spatial information would lead to binding errors. They further suggested that the spatial information used in binding depends on intact parietal function. In the present paper we extend these findings and examine other deficits in RM that would be predicted by FIT. We show that: (1) Object individuation is impaired, making it impossible for him correctly to count more than one or two objects, even when he is aware that more are present. (2) Visual search for a target defined by a conjunction of features (requiring binding) is impaired, while the detection of a target defined by a unique feature is not. Search for the absence of a feature (0 among Qs) is also severely impaired, while search for the presence (Q among 0s) is not. Feature absence can only be detected when all the present features are bound to the nontarget items. (3) RM's deficits cannot be attributed to a general binding problem: binding errors were far more likely with simultaneous presentation where spatial information was required than with sequential presentation where time could be used as the medium for binding. (4) Selection for attention was severely impaired, whether it was based on the position of a marker or on some other feature (color). (5) Spatial information seems to exist that RM cannot access, suggesting that feature binding relies on a relatively late stage where implicit spatial information is made explicitly accessible. The data converge to support our conclusions that explicit spatial knowledge is necessary for the perception of accurately bound features, for accurate attentional selection, and for accurate and rapid search for a conjunction of features in a multiitem display. It is obviously necessary for directing attention to spatial locations, but the consequences of impairments in this ability seem also to affect object selection, object individuation, and feature integration. Thus, the functional effects of parietal damage are not limited to the spatial and attentional problems that have long been described in patients with Balint's syndrome. Damage to parietal areas also affects object perception through damage to spatial representations that are fundamental for spatial awareness.

Understanding cooperative advertising participation rates in conventional channels

Abstract: Cooperative advertising plans feature prominently in marketing programs in conventional channels and make up the majority of marketing funds in some product categories. Available data show that coo...

Image insertion in video streams using a combination of physical sensors and pattern recognition

Abstract: A live video insertion system in which one or more event cameras (110) include sensors (113) for sensing camera zoom, focus, pan, and tilt. Sensor data from each camera is provided to a live video insertion system to give a coarse indication of where an insertion should occur in video scenes. The sensor and tally data essentially replaces the search mode of conventional pattern recognition live video insertion systems. An accurate final determination of an insertion location is determined by using feature and/or texture analysis in the actual video image. This analysis compares the position of the features and/or texture within the video frame to their corresponding location in a common reference image or previous image of the insertion location and surroundings.

Dimensionality reduction of unsupervised data

Abstract: Dimensionality reduction is an important problem for efficient handling of large databases. Many feature selection methods exist for supervised data having class information. Little work has been done for dimensionality reduction of unsupervised data in which class information is not available. Principal component analysis (PCA) is often used. However, PCA creates new features. It is difficult to obtain intuitive understanding of the data using the new features only. We are concerned with the problem of determining and choosing the important original features for unsupervised data. Our method is based on the observation that removing an irrelevant feature from the feature set may not change the underlying concept of the data, but not so otherwise. We propose an entropy measure for ranking features, and conduct extensive experiments to show that our method is able to find the important features. Also it compares well with a similar feature ranking method (Relief) that requires class information unlike our method.

Portable information-providing apparatus

Abstract: A portable information-providing apparatus comprising a position detector, a direction detector, a processor, a memory, and information output device. The position detector provides data of a use's position. The direction detector provides data of a direction as determined by the user. The processor operates to correlate said position and direction data with information of features in the user's surroundings as stored in the memory. The information output device provides the information to the user, the information being related to a feature in the user's surroundings generally along the direction determined by the user.

Methods of playing poker games

Abstract: The present invention relates to methods for playing seven card stud and five card draw poker variants including an optional progressive jackpot feature, and more particularly relates to particular variations of the CARIBBEAN STUD (registered trademark) poker variant disclosed in U.S. Pat. No. 4,836,553 and including an optional progressive jackpot component as disclosed in U.S. Pat. Nos. 4,861,041 and 5,078,405.

Configuration based scene classification and image indexing

Abstract: Scene classification is a major open challenge in machine vision. Most solutions proposed so far such as those based on color histograms and local texture statistics cannot capture a scene's global configuration, which is critical in perceptual judgments of scene similarity. We present a novel approach, "configural recognition", for encoding scene class structure. The approach's main feature is its use of qualitative spatial and photometric relationships within and across regions in low resolution images. The emphasis on qualitative measures leads to enhanced generalization abilities and the use of low-resolution images renders the scheme computationally efficient. We present results on a large database of natural scenes. We also describe how qualitative scene concepts may be learned from examples.

Classification by Voting Feature Intervals

Abstract: A new classification algorithm called VFI (for Voting Feature Intervals) is proposed. A concept is represented by a set of feature intervals on each feature dimension separately. Each feature participates in the classification by distributing real-valued votes among classes. The class receiving the highest vote is declared to be the predicted class. VFI is compared with the Naive Bayesian Classifier, which also considers each feature separately. Experiments on real-world datasets show that VFI achieves comparably and even better than NBC in terms of classification accuracy. Moreover, VFI is faster than NBC on all datasets. This project is supported by TUBITAK (Scientific and Technical Research Council of Turkey) under Grant EEEAG-153.