Point-like Features Research Articles

Line segment based axial motion stereo

Axial motion stereo possesses some significant advantages over the more popular lateral stereo. While all earlier works involving the axial motion stereo model have used point-like features as matching primitives, we propose in this paper a line segment based matching technique. We discuss various matching constraints applicable to line segments under the axial stereo model. In particular, we prove that the orientations of corresponding two dimensional (2D) line segments stemming from the same 3D line segment tend to the equal. Our correspondence finding algorithm consists of a local matching procedure based on a number of constraints and a subsequent global disambiguation process. The method has been implemented, and the results are very promising.

Alignment using an uncalibrated camera system

We describe a method for the visual control of a robotic system which does not require the formulation of an explicit calibration between image coordinates and the world coordinates. By extracting control information directly from the image, we free our technique from the errors normally associated with a fixed calibration. We attach a camera system to a robot such that the camera system and the robot's gripper rotate simultaneously. As the camera system rotates about the gripper's rotational axis, the circular path traced out by a point-like feature projects to an elliptical path in image space. We gather the projected feature points over part of a rotation and fit the gathered data to an ellipse. The distance from the rotational axis to the feature point in world space is proportional to the size of the generated ellipse. As the rotational axis gets closer to the feature, the feature's projected path will form smaller and smaller ellipses. When the rotational axis is directly above the object, the trajectory degenerates from an ellipse to a single point. We demonstrate the efficacy of the algorithm on the peg-in-hole problem.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Motion stereo for mobile robots

A method allowing a mobile robot to measure accurately its distance from external objects is introduced. It only requires a single uncalibrated camera and a dynamic vision system. The method was evaluated in an optical laboratory and also in outdoor-experiments. Even in outdoor-experiments with a vehicle moving at a speed of 30 km/h, errors of less than 1% of the true distance were achieved in real time. The accuracy achievable with the method depends on the nature of the image features used as a basis for the measurement. Laboratory experiments have been performed to investigate quantitatively the characteristics of the feature detectors under conditions of reduced lighting and, consequentially, reduced SNR, and with image features not matching the feature detectors perfectly. The best results can be achieved by using edge-like features. Point-like features do have certain desirable characteristics, but point detectors are more sensitive to noise than edge detectors. Hence, the accuracy achieved with corner detectors was less than the one achieved with edge detectors. Nevertheless, even with corner detectors the resulting error was less than 1%, despite the fact that the image of the target object did not contain good corners.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Linear feature compatibility for pattern-matching relaxation

This paper extends point-pattern matching relaxation techniques to allow matching of both point-like and linear features. We define a compatibility function that relies on relative orientation information, which is translation and rotation invariant and which can be more reliably extracted from noisy images than can positional information. The function is used to generalize the matching technique of Ranade and Rosenfeld ( Pattern Recognition 12, 269–275 (1980)), but it can also be incorporated into other relaxation algorithms. Examples are presented from the domain of object recognition in synthetic aperture radar (SAR) imagery.

Registration of MR and CT images for skull base surgery using point-like anatomical features

We have developed a registration technique for combining magnetic resonance imaging (MRI) and computed tomography (CT) images of the skull base for use in surgical planning. The technique is based on user identification of point-like landmarks visible in both modalities. The combination of images involves a small amount of expert interaction, is relatively quick and preliminary evaluation indicates that it is accurate to within 1.5 mm. Registered or fused images can be viewed either on an image processing workstation, or fused images can be printed onto conventional film for convenience in clinical use. We present one patient in order to demonstrate the technique's indications and advantages.

Structure-from-motion under orthographic projection

Structure-from-motion algorithms based on matched point-like features under orthographic projection are explored, for use in analysing image motion from small rigid moving objects. For two-frame analysis, closed-form n- point algorithms are devised that minimize image-plane positional errors. The bas-relief ambiguity is shown to exist for arbitrary object rotations. The algorithm is applied to real images, and good estimates of the projection of the axis of rotation on to the imageplane are obtained.

Identification and analysis of the alignments of point-like features in remotely-sensed imagery:Volcanic cones in the Pinacate Volcanic Field, Mexico

Abstract Abstract. Some fractures in the crust are expressed at the surface as a field of point-like features that have developed in response to usually fluid processes within those fractures. We describe two complementary quantitative methods for analysing such point-like features from remotely-sensed imagery in terms of structural alignment. The two-point azimuth method is probabilistic and tests for the existence of regional and field-wide structural anisotropies. By contrast, the Hough transform method is deterministic and locates specific alignments. These methods have been applied to a Landsat Multispectral Scanner (MSS) image of the Pinacate Volcanic Field in north-west Mexico. The point-like features of interest are cinder cones and maars. The results reveal a northerly structural anisotropy and specific alignments, trending mainly northerly and north-northwesterly, have been mapped. This application was limited by the relatively coarse spatial resolution of the MSS image used. However, although o...

Three-Dimensional Ultrasound C-Scan Imaging Using Holographic Reconstruction

A three-dimensional ultrasound C-scan imaging method based on holographic reconstruction has been developed. Using wide angle coherent illumination of the object and holographic principles, high lateral resolution is achieved throughout the imaging space. For this method the object is insonified by monochromatic ultrasound bursts, and the wavefronts reflected from the different object levels are recorded into a stack of holograms, each of which corresponds to a different object level. After recording each plane of the three-dimen- sional hologram is first Fourier transformed. Then each plane wave spectrum is transferred to the object space by a transfer function, and the wavefront of the object plane is calculated using an inverse Fourier transform. Finally all the object planes are combined to form a three- dimensional image, which is displayed as front, side, and top projec- tions and as a stereopair. Both simulated and real measurements of test objects that contained point-like features were accomplished. An ultrasonic frequency of 4 MHz was used in all measurements. The sim- ulated measurements showed that the resolution of the three-dimen- sional imaging method is better than 2 mm in all three directions. Using effective coding methods and fast Fourier transform algorithms, the imaging time of a 64 X 64 X 32-point image could be reduced to 2 h with an ordinary 256-kword minicomputer.

An analysis of the orientations of large‐scale crustal structures: A statistical approach based on areal distributions of pointlike features

The spatial distributions of agmatic complexes and other features thought to be associated with major crustal structures may be important sources of information about large‐scale structural patterns. However, attempts to incorporate these features into quantitative analyses of linear features have used arbitrary or inaccurate criteria to judge hypothetical geological relationships. In this paper, features of limited spatial extent are considered pointlike, and the concept of a probabilistic lattice point distribution is used to formulate a statistical method that leads to a quantitative and reproducible analysis of directional patterns based solely on the locations of the points. Thus this analysis is independent of linear patterns and provides a measure of the directional information intrinsic to point patterns. The procedure determines the most likely trends of structural anisotropies; Monte Carlo simulations of random point patterns provide a reference distribution from which confidence levels can be determined. Applications to published data for magmatic complexes, magnetic contour closures, and structural change points are used as examples. The results suggest that there has been a tendency to overestimate the amount of information available from point patterns.

The X-Ray Corona and the Photospheric Magnetic Field

Soft X-ray (3–60 Å) photographs of the solar corona have been obtained on four flights of a rocket-borne grazing incidence telescope having a resolution of a few arc seconds. The configuration of the X-ray emitting structures in the corona has been compared to the magnetic field distribution measured by photospheric longitudinal magnetograms. The X-ray structures trace the three dimensional configuration of the magnetic field through the lower corona.Active regions in the corona take the form of tubular structures connecting regions of opposite magnetic polarity within the same or adjacent chromospheric active regions.Higher, larger structures link widely separated active regions into complexes of activity covering substantial fractions of the disk. The complexes are separated by areas of low average field in the photosphere. Interconnections across the solar equator appear to originate over areas of preceding polarity.Enhanced X-ray emission is observed, outside the active belt, over areas of enhanced general magnetic field. Bright point-like X-ray features are observed above bipolar areas in the general coronal field.Two, probably related, classes of X-ray features are associated with points of high field gradient along the longitudinal magnetic, ‘neutral line’. Both in the non-flaring active region and in flares the brightest emission is observed from one or more small line sources (at our present resolution). In flares this can account for a substantial fraction of the total soft X-ray emission of the flare.