Showing papers on "Object detection published in 1993"

Estimating friction using incipient slip sensing during a manipulation task

Abstract: A scheme is presented by which a manipulator can use dynamic tactile sensing to detect when it is about to lose hold of a grasped object. By detecting localized slips on the gripping surface which precede gross slip, the controller can modify the grasp force to prevent the object from slipping. Also, by monitoring normal and tangential forces at the contact when these incipient slip signals occur, the controller obtains an accurate estimate of the friction coefficient which can then be used during the manipulation task. Accurate knowledge of the friction coefficient is essential when grasping fragile objects or manipulating with sliding. >

Tracking non-rigid objects in complex scenes

Abstract: The authors describe a model-based method for tracking nonrigid objects moving in a complex scene. The method operates by extracting two-dimensional models of an object from a sequence of images. The basic idea underlying the technique is to decompose the image of a solid object moving in space into two components: a two-dimensional motion and a two-dimensional shape change. The motion component is factored out and the shape change is represented explicitly by a sequence of two-dimensional models, one corresponding to each image frame. The major assumption underlying the method is that the two-dimensional shape of an object will change slowly from one frame to the next. There is no assumption, however, that the two-dimensional image motion between successive frames will be small. >

A Pyramid Framework for Early Vision: Multiresolutional Computer Vision

Abstract: From the Publisher: Biological visual systems employ massively parallel processing to perform real-world visual tasks in real time. A key to this remarkable performance seems to be that biological systems construct representations of their visual image data at multiple scales. A Pyramid Framework for Early Vision describes a multiscale, or 'pyramid', approach to vision, including its theoretical foundations, a set of pyramid-based modules for image processing, object detection, texture discrimination, contour detection and processing, feature detection and description, and motion detection and tracking. It also shows how these modules can be implemented very efficiently on hypercube-connected processor networks. The volume is intended for both students of vision and vision system designers; it provides a general approach to vision systems design as well as a set of robust, efficient vision modules.

Tracking foveated corner clusters using affine structure

Abstract: The authors describe a novel method of obtaining a fixation point on a moving object for a real-time gaze control system. The method makes use of a real-time implementation of a corner detector and tracker and reconstructs the image position of the desired fixation point from a cluster of corners detected on the object using the affine structure available from two or three views. The method is fast, reliable, viewpoint invariant, and insensitive to occlusion and/or individual corner dropout or reappearance. Results are presented for the method used with a high performance head/eye platform. The results are compared with two naive fixation methods. >

Object contour extraction using color and motion

Abstract: A new approach to the extraction of the contour of a moving object is presented. The method is based on the integration of a motion segmentation technique using image substraction and a color segmentation technique based on the split-and-merge algorithm. The advantages of this method are: it can detect large moving objects and extract their boundaries; the background can be arbitrarily complicated and contain many non-moving objects occluded by the moving object; and it requires only three image frames that need not be consecutive, provided that the object is entirely contained in each of the three frames. The method is applied to a large number of color images of vehicles moving on a road and a highway ramp. The results are promising. The moving object boundaries are correctly extracted in 66 out of 73 test image sequences. The authors describe how this contour can be used as an input to a recognition system that classifies the vehicles into five generic categories. Of the 73 vehicles, 67 are correctly classified. >

Generalized tube model: recognizing 3D elongated objects from 2D intensity images

Abstract: The issue of recognizing 3D elongated objects from 2D intensity images is addressed. A tube model, locally similar to generalized cones, is developed for the class of elongated objects. A recognition strategy that combines 2D contour properties and surface shading information is used to exploit the power of the 3D model. Reliable contours provide constraints for localizing the objects of interest. The theory of optimal filters is adopted in verifying the shading of hypothesized objects. Object recognition is achieved through optimizing the signal-to-noise response with respect to model parameters. A sweeping operation is proposed as a further stage of identifying objects so that the overall performance of the system does not heavily rely on the quality of local feature detection. >

A connectionist model for category perception: theory and implementation

Abstract: A connectionist model for learning and recognizing objects (or object classes) is presented. The learning and recognition system uses confidence values for the presence of a feature. The network can recognize multiple objects simultaneously when the corresponding overlapped feature train is presented at the input. An error function is defined, and it is minimized for obtaining the optimal set of object classes. The model is capable of learning each individual object in the supervised mode. The theory of learning is developed based on some probabilistic measures. Experimental results are presented. The model can be applied for the detection of multiple objects occluding each other. >

Object recognition using ultrasonic sensors in robotic applications

Abstract: Presents the results of a computer controlled system using ultrasonic transducers in applications like detection and identification of simply shaped objects for their subsequent manipulation in an automatic process. The authors have employed two kinds of transducers, one commercial and an other which uses the authors' own electronic control circuit, on several structures. For a robotic arm, the system makes it capable to do simple tasks of elemental recognition and basic assembly of pieces. Thus, it has enough flexibility to operate in nonstatic environments. This alternative, using simple and economic proximity sensors, is very attractive for certain applications in manufacturing processes. >

Optimal estimation of object pose from a single perspective view

Abstract: The authors present a method for robustly and accurately estimating the rotation and translation between a camera and a 3-D object from point and line correspondences. First they devise an error function and then show how to minimize this error function. The quadratic nature of this function is made possible by representing rotation and translation with a dual number quaternion. A detailed account is provided of the computational aspects of a trust-region optimization method. This method compares favourably with Newton's method, which has extensively been used to solve the problem, and with Faugeras-Toscani's linear method (1986) for calibrating a camera. Some experimental results are presented which demonstrate the robustness of the method with respect to image noise and matching errors. >

The dynamic retina: contrast and motion detection for active vision

Abstract: The dynamic retina is an efficient, biologically-inspired early vision architecture that is well-suited to active vision platforms. It exploits normally undesirable camera motion as a necessary step in detecting image contrast, using dynamic receptive fields instead of traditional spatial-neighborhood operators. A receptor response function, based on a light-adaptation model for vertebrate receptors, works together with the camera movements to compute spatial image contrast. The dynamic retina also responds to moving objects, producing a clear signature from which motion parameters can be extracted. >

Motion detection in image sequences acquired from a moving platform

Abstract: A two-step motion segmentation algorithm for the detection of moving objects in image sequences acquired from a moving platform is presented. First, the image plane transformation induced by the moving platform is estimated using a subpixel accuracy image registration algorithm. The registration algorithm is fully automatic and performs well under significant camera rotation and translation. The input images are then transformed into a common coordinate system (for example, the coordinate system of the first image). One then segments the changed regions from the camera motion-compensated frame difference. Finally, the moving object is detected from the closure of changing segments, and the object motion parameters are estimated. Experimental results for real image sequences are presented. >

Visual detection of distant objects

Abstract: Taking an autonomous road vehicle as an example, approaches and methods for object detection (this being the first step of obstacle recognition) are discussed. Vision, using a single passive camera as the primary sensor and a real-time vision system for analyzing the resulting image sequence, is found to be the best approach. Results obtained in real-world experiments with such a system are reported: large obstacles have been detected in distances up to 716 m; all relevant objects that were encountered during a prolonged test on a highway were detected; the cycle time of the detection process was below 100 ms, and less than one false alarm occurred per minute.

Integration of photogrammetric cues into cartographic feature extraction

Abstract: Most systems for cartographic features extraction developed within the computer vision and image understanding community make little use of detailed camera information during object detection and delineation. For the most part the scale, size, and orientation of specific features are usually expressed in terms of image pixel size. Given the use of nadir and near-nadir mapping photography this has not severely impacted the development of several techniques at a variety of institutions for building detection, road network extraction, and other specific man-made objects. It is not too unfair to say that the inherent difficulties involved in achieving robust automated object detection and delineation have overshadowed any errors due to lack of rigor in the modeling of the image acquisition. In this paper we will develop several of these issues and discuss how the use of photogrammetric cues will play a major role in future systems for automated cartographic feature extraction.

Submerged object detection and classification system

Abstract: A detection and classification system for underwater objects uses a transting unit and a receiving unit The transmitting unit comprises a waveform generator, a power amplifier, and a transmitting antenna The receiving unit comprises a receiving antenna, a pre-amplifier, a first harmonic suppressor, a digitizer, and a computer The transmitting unit radiates an analog electromagnetic wave signal into a conductive medium such as seawater which the receiving unit detects and analyzes by a differential spectral analysis after conversion of the signal into binary code The system uses a signal-processing method which includes the steps of determining the size of the underwater object to be detected, transmitting an electromagnetic wave with a wavelength proportioned to the size of the object, performing a spectral analysis of the received signal, performing a spectral analysis at a different time or different location, comparing the two spectra performed, and analyzing the difference between the two spectra

Optimal polarizations for radar detection and recognition of targets in clutter

Abstract: The authors investigate the detection and recognition of stationary ground targets in high-resolution, fully polarimetric synthetic aperture radar (SAR) imagery. Using 1-ft*1-ft SAR imagery of targets and clutter gathered by a 33-GHz sensor, several techniques for improving target detection performance through optimal processing of the fully polarimetric SAR data. >

Method for semiautomated serial section reconstruction and visualization of neural tissue from TEM images

Abstract: A simple method to reconstruct details of neural tissue architectures from transmission electron microscope (TEM) images will help us to increase our knowledge of the functional organization of neural systems in general. To be useful, the reconstruction method should provide high resolution, quantitative measurement, and quick turnaround. In pursuit of these goals, we developed a modern, semiautomated system for reconstruction of neural tissue from TEM serial sections. Images are acquired by a video camera mounted on TEM (Zeiss 902) equipped with an automated stage control. The images are reassembled automatically as a mosaicked section using a crosscorrelation algorithm on a Connection Machine-2 (CM-2) parallel supercomputer. An object detection algorithm on a Silicon Graphics workstation is employed to aid contour extraction. An estimated registration between sections is computed and verified by the user. The contours are then tessellated into a triangle-based mesh. At this point the data can be visualized as a wireframe or solid object, volume rendered, or used as a basis for simulations of functional activity.© (1993) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

Chaotic detection of small target in sea clutter

Abstract: A new method based on chaos and neural network theory for small target detection in sea clutter is presented. The authors present the signal model. They then describe a neural network trained with the least-mean-square (LMS) algorithm for the dynamic reconstruction of sea clutter. A chaotic detection algorithm based on the neural network model is described. Some experimental results show that the new detection method operating with noncoherent radar data has a performance comparable with that of a standard Doppler CFAR processor. >

A self-tuning system for real-time optical flow detection

Abstract: The work presented in this paper is aimed towards the real-time analysis of image sequences acquired from a moving vehicle. The authors' goal is to extract the optical flow field from the image sequence. The information obtained is used to detect obstacles in front of a moving vehicle and to compute the time to impact for approaching objects. These requirements mean that special care must be taken with the response time of the algorithm. In order to obtain a fast response, a special purpose massively parallel architecture is used. Furthermore, the algorithm uses an heuristic approach rather than an analytical one. In the approach presented, the detection of optical flow is driven by a high level process, by tuning the algorithm's parameters. >

Off-board augmented fusion for improved target detection and track

Abstract: The development of a multiple platform distributed fusion capability offers the potential for effective data interchange linking multiple surveillance platforms and centers. Initial off-board derived data can provide the appropriate cues for sector prioritization and sensor resource management. The efficient correlation and fusion of surveillance data from multiple sources with sensors directly on the platform will provide increased situation awareness. This paper discusses a multiplatform distributed fusion architecture under development to provide enhanced cueing and sensor resource management. It shows the current status of ongoing testbed development activities that can be used to provide insight into the proposed architecture. An example is provided to correlate this research to the drug interdiction problem. >

Moving object detection and judgement device

Abstract: PURPOSE:To provide a moving object detection and judgement device capable of surely detecting an object moving in an area and picking up the image of the moving object in a state required for judgement without providing a dead angle in a wide monitoring area with few cameras. CONSTITUTION:This device is provided with a detection camera device 10 for moving object detection capable of detecting a wide range area, a judgement camera device for picking up the image in a state sufficient for judging whether or not the moving object is an object based on position information or the like outputted from the detection camera device 10 and a judgement processing part 30 for performing a judgement processing based on picture data image picked up by the judgement camera device. The detection camera device 10 is provided with a detection camera 11 facing upwards and an approximately conical reflection mirror 12 arranged above the detection camera 11 and can monitor the wide range of 360 degrees. A judgement camera 21 is rotated based on the position information and seizes the moving object and even at the time, the detection camera 10 performs a detection processing for the other moving object.

Fast alignment using probabilistic indexing

Abstract: The alignment method is a model-based object recognition technique that determines possible object transformations from three hypothesized matches of model and image points. For images and/or models with many features, the running time of the alignment method can be large. Methods of reducing the number of matches that must be examined are presented. The techniques described are the use of the probabilistic indexing system, and the elimination of groups of model points that produce large errors in the transformation determined by the alignment method. Results are presented which show that it is possible to achieve a speedup of over two orders of magnitude while still finding a correct alignment. >

Identification of object parameters with robot hand

Abstract: Generally, the grasp problem by the robot hand has two categories, one is stable grasp and the other is object parameter detection. This paper proposes to combine these two issues, and shows the concepts 'process of grasp' and 'cognition grasp'. Along these concepts, the way to detect mass center and to approximate the shape of the object using tactile sensor information from the dexterous robot hand system are discussed. Further a method to treat surface information to realize active sensing with rolling motion of the object around the fingertip. The robot hand system including a 6-axis force and torque sensor and controller to verify those algorithm is also shown. >

Spatial association analysis for radar image interpretation

Abstract: SAR image segmentation was attempted using the spatial association of elements based on pixel intensities. The associations (1) among very bright targets, (2) between very dark areas and the surrounding area, and (3) between the bright side and the dark side (ridge structures) were investigated. >

Spatio-temporal X-ray fluoroscopy filtering using object detection

Abstract: The authors describe an image processing technique to enhance X-ray fluoroscopy image sequences by reducing noise while minimizing motion blur. The method consists of object detection, followed by spatial and recursive temporal filtering. In many fluoroscopy applications, much of the motion occurs in isolated long thin objects (catheters, guide wires, etc.). The authors demonstrate that such structures can be detected and roughly segmented. Based upon the segmentation output they adjust spatial and temporal filtering parameters, such as to reduce noise without blurring the objects. As compared to conventional motion-detection. Filtering, this object-detection filtering method incorporates additional a priori knowledge of the image content. The authors demonstrate the segmentation and enhancement method on X-ray fluoroscopy sequences. >

Multiresolution target detection and tracking through a parallel coarse-to-fine search approach

Abstract: A new pyramidic-structure-based approach to moving object detection and tracking from image sequences is proposed. The approach uses a Gaussian low-pass filter to transform each frame in the image sequences into a multiresolution representation, or a pyramidic structure of the frame. At each level of the image pyramid, a correlative matching process is performed throughout a restricted area to detect and localize the moving object. The pyramidic structures of images used in current matching recurrence are formed in the previous recurrences, and the searching area is determined by the recurrence at the upper level. Theoretical analysis and experiment show that the proposed approach has raised the efficiency by 150 times or more as compared with a conventional matching method. Furthermore, the method for mapping the approach into a deliberately designed parallel processor structure named Elementary Pyramid is investigated for the purpose of speeding up the search performance further. >

Planning multiple observations for specular object recognition

Abstract: The most prominent features of specular objects are the specularities, which are highly variable and dependent on local object geometry. In order to unambiguously recognize specular objects, more information is required. An approach for specular object recognition that relies on the use of multiple observations from different viewpoints to resolve any ambiguity in scene interpretation is presented. The results show that the multiple observation strategy can be very accurate, and is in fact limited only by the accuracy with which decisions can be made about individual observations. >

Object detection device

Abstract: The invention makes use of fluctuations in the quality factor (Q) of an oscillation circuit (S) that occur when an object comes into the vicinity of an electrode (10) as a means of detecting the presence or absence of the object. The invention comprises an electrode (10), which is set on the edge of a detection field, a supplementary device (20) which, together with the electrode (10), forms an oscillation circuit (S), and a Q detection device (30), which detects fluctuations in the quality factor of the oscillation circuit (S). The object detection device of the invention dispenses with the need for a frequency detection circuit and this simplifies the overall configuration. The device is thus more compact, less expensive and easier to use than conventional object detection devices while at the same time ensuring detection of the presence or absence of objects in the detection field.

Multiple-camera/motion stereoscopy for range estimation in helicopter flight

Abstract: Pilot aiding to improve safety and reduce pilot workload to detect obstacles and plan obstacle-free flight paths during low-altitude helicopter flight is desirable. Computer vision techniques provide an attractive method of obstacle detection and range estimation for objects within a large field of view ahead of the helicopter. Previous research has met considerable success by using an image sequence from a single moving camera in solving this problem. The major limitations of single camera approaches are that no range information can be obtained near the instantaneous direction of motion or in the absence of motion. These limitations can be overcome through the use of multiple cameras. This paper presents a hybrid motion/stereo algorithm which allows range refinement through recursive range estimation while avoiding loss of range information in the direction of travel. A feature-based approach is used to track objects between image frames. An extended Kalman filter combines knowledge of the camera motion and measurements of a feature's image location to recursively estimate the feature's range and to predict its location in future images. Performance of the algorithm will be illustrated using an image sequence, motion information, and independent range measurements from a low-altitude helicopter flight experiment.