By Topic

Image Processing And Its Applications, 1999. Seventh International Conference on (Conf. Publ. No. 465)

Date 1999

Go

Filter Results

Displaying Results 1 - 25 of 83
  • A handwriting understanding environment (HUE) for research in document and handwriting processing

    Page(s): 740 - 744 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (272 KB)  

    Researchers in the fields of document processing and handwriting analysis and recognition increasingly need to construct elaborate software systems to investigate and evaluate new research components such as cursive handwriting algorithms, contextual recognition systems and document parsing tools. Although a number of commercial and public-domain image processing frameworks already exist, for example Khoros, KBVision and the Image Understanding Environment (IUE) none of these systems fully supports the cycle of research, software development, informal and formal evaluation which is typically required by the research community. TABS (The Almost a Blackboard System) was originally developed to address this problem by providing a library of domain-specific image processing and computer vision components within a small, efficient, and easily accessible Tcl Tk script-based software framework. TABS was evaluated over a period of more than a year, and has now been rewritten and extended to produce HUE, a handwriting understanding environment which provides a rapid prototyping environment for document and handwriting research. We review the features of HUE, and illustrate its use with a realistic research evaluation example View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An algorithm for face and facial-feature location based on grey-scale information and facial geometry

    Page(s): 625 - 629 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (348 KB)  

    We describe a two-stage algorithm for detecting human faces based on the grey-scale information of the input image and facial geometry. The first stage involves locating approximately the head boundaries and the facial features, the exact locations of which are determined in the second stage. The algorithm is evaluated on data sets and some characteristics of its performance are identified. As well as being of value for the feature location algorithm in particular, the approach is intended to illustrate how any vision technique may be characterized in an objective way View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Modelling of facial features and its applications

    Page(s): 634 - 637 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (252 KB)  

    It has been demonstrated by linear regression that contrary to current thinking the relationship between the mouth width and lower lip depth is one of negative association. Early results indicate a strong case for a linear model. Random generation of mouth templates from this model show a good fit to typical image data. An application has been presented which is able to locate and model the lower lip and decide whether a mouth image is open or closed. The parameters of the model are used to classify the located mouth shape into one of a predetermined set. Thus this subsystem/component can be incorporated into existing lip sync systems, where given both streams of data and the knowledge gained from the above system, it would be possible to obtain a better match between the speech and image data in multi-dimensional vector space View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A nonrigid-body approach to matching mammograms

    Page(s): 484 - 488 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (360 KB)  

    This paper has explored the use of a nonrigid-body approach to mammogram registration. Mammogram registration is inherently complicated by factors such as the inhomogeneous composition of the breasts and their nonrigid behavior. We have developed a technique which accounts for such nonrigid-body idiosyncrasies and allows corresponding mammograms to be registered for subsequent fusion. Future work in this area involves the development of techniques to extract control-points from the inner portion of the breast and integration into a completely autonomous comparative analysis schema View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Wavelet based filtering in nuclear medicine

    Page(s): 851 - 855 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (240 KB)  

    Nuclear medicine imaging utilises the nuclear properties of radioactive nuclides to make diagnostic evaluation of the anatomic and physiologic condition of the body. In contrast to other imaging techniques which are based upon anatomy, it permits one to gather medical information that may be otherwise unavailable, require surgery or necessitate more expensive diagnosis tests. One of the major sources of error in nuclear medicine imaging is Poisson noise which hinders image analysis and interpretation. This paper introduces a wavelet based Poisson noise filtering procedure and assesses its performance View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Learning structural change for identification and tracking of vehicles moving in open world scenes

    Page(s): 706 - 710 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (432 KB)  

    This paper describes a system, which is able to track multiple man made objects, typically vehicles moving in a natural open world scene. The system interprets changes in structural features in the scene across consecutive image frames and uses the change in structure to identify and track man made objects (target). No a priori knowledge of any structure within the image is assumed. Differences in statistics between a reference image and the current image generate motion cues that are used to identify regions of interest (ROI). Intensity and edgel information are extracted for the ROI in both the current and reference images. Correlation between the extracted data for each ROI yields an initial recognition of the ROI as either a target or object. This process is repeated on a frame by frame basis generating sets of object and target dynamics. The extracted dynamics are used in conjunction with a high-level reasoning process to solve the frame to frame correspondence process. Identified targets are labelled and tracked, objects that fail both the recognition and correspondence process are removed and no longer processed View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Remote bar-code localisation using mathematical morphology

    Page(s): 642 - 646 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (364 KB)  

    Bar-code labels are used in industry to identify and classify a wide variety of products, tools and equipment, their reading generally being undertaken via a laser scanning system. In most cases the number, position and orientation of the bar-codes are well known. However, a particular application under consideration here involves the identification of radioactive material containers via their bar-codes, in which general safety requirements demand remote identification (at least five metre distant) and where laser scanning cannot be used. The images acquired can contain more than one bar-code and these can be found in varying positions within the field of view. Furthermore the container bar code labels are known to occur in random position and orientation. This work focuses on a computer vision based solution to this problem in which the bar-codes are remotely located from within a general grey scale image using a morphological based processing and algorithm. The final solution utilises a pipelined architecture to obtain near real-time operation View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Hierarchical fast decoding of fractal image representation using quadtree partitioning

    Page(s): 581 - 585 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (396 KB)  

    Fractal image coding using either fixed-size blocks or variable-size blocks in a quadtree structure is by now common. However, a fast decoding algorithm based on a hierarchical representation, which requires only a finite number of iterations to exactly reach the fixed point of the fractal transformation, has been shown only for fixed-size blocks. In this work, a generalization of the fast decoding algorithm is made, enabling its use with a quadtree image partitioning. A theorem extending the hierarchical representation of the fractal transformation fixed point to include quadtree partitioning is given and proved View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Tongue tracking in medical X-ray sequences

    Page(s): 494 - 497 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (348 KB)  

    This payer presents a system for the automated tracking of non-rigid anatomic structures in two-dimensional image sequences. It is applied to X-ray image sequences of the vocal tract. In this application articulatory organs have to be measured to investigate the complex dynamic characteristics of human speech production, with particular interest in the robust boundary detection of non-rigid organs such as the tongue View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A registration algorithm of eye fundus images using a Bayesian Hough transform

    Page(s): 479 - 483 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (420 KB)  

    Image registration is the real challenge of retinal image analysis. Temporal registration is necessary in order to follow the various steps of a disease and to measure the evolution of some lesions. This is particularly true for patients developing diabetes, the first cause of legal blindness in most occidental countries, where the number and the turnover rate of lesions such as micro-aneurysms is related to the gravity of the illness. This paper presents an algorithm for temporal registration of retinal images based on point correspondence. The algorithm has been applied to the registration of fluorescein images (obtained after a fluorescein dye injection). The vascular tree is first detected in each image and bifurcation points are labelled with surrounding vessel orientations. An angle-based invariant is then computed in order to give a probability for two points to match. Then a Bayesian Hough transform is used to sort the possible matchings with their respective likelihood. A precise affine estimate and a score are computed for most likely transformations and the best transformation is chosen View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A minimum classification error method for face recognition

    Page(s): 630 - 633 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (216 KB)  

    In this paper, we propose a minimum classification error rate based face recognition system. In our work, the minimum classification error formulation is incorporated into a neural network classifier called a multilayer perceptron. Experimental results show that our system is robust to noisy images and complex background View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Reduced FOV imaging with motion adaptation

    Page(s): 502 - 506 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (316 KB)  

    This paper addresses the use of reduced FOV (field-of-view) imaging with spatially variable resolution for 3-D MR coronary angiography. Details involved in using the technique for reducing imaging time and facilitating the management of respiratory motion are discussed. The technique was evaluated with phantom experiments and right coronary artery images of 11 asymptomatic volunteers using a 0.5 T MR system View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Region-based image coding for remote sensing

    Page(s): 808 - 812 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (316 KB)  

    This paper argues that texture regions in remotely sensed images of the Earth are often of no interest to projects for example, concerned with agricultural applications. These regions require a large number of bits to be encoded. It is proposed that they can be identified using a generic algorithm that identifies the boundaries of textured regions irrespective of their class, and removed from the encoding process. The rest of the regions which may be of interest to the specific application, may be encoded using the 1D wavelet transform applied to the string of pixels created by raster scanning the region. This approach can help remove the bottleneck of image down-loading from microsatellites in low Earth orbit, because these satellites can obtain hundreds of images in an orbit but they can only download a few of them during each pass over the tracking station. The proposed approach can be fully implemented for on-board image preprocessing before the down-loading, for cases that urban and forest regions (textured regions) in the images are of no interest View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Geometric 2D image analysis for automatic visual inspection

    Page(s): 656 - 660 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (280 KB)  

    We propose two novel algorithms to analyse the motion parameters and structural data of filter components for quality inspection in a manufacturing line. The algorithms are based on geometrical properties of image correspondence vectors providing a closed form solution to all parameters of interest making full use of the distance between feature points and angle information. For comparison purposes, we also implemented Tsai and Huang's (1984) calibration algorithm based on epipolar geometry. Experimental results show that the proposed algorithms are in general more accurate than Tsai and Huang's and that they perform well in the presence of noise View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Page similarity and the Hausdorff distance

    Page(s): 755 - 759 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (276 KB)  

    The VideoWriter is a real-time system that uses digitized images from a computer-controlled video camera to find the location of pages on a desktop. Once an image of a page is extracted, the system must determine if the page has been previously stored. We investigate the use of the Hausdorff distance metric to compare two-level versions of page images. The Hausdorff distance metric does not require exact correspondence between pels in images. We compare the accuracy of the original metric with several modifications View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A one pass parallel hexagonal thinning algorithm

    Page(s): 841 - 845 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (360 KB)  

    The development of a fully parallel hexagonal thinning algorithm is reported and compared with a similar algorithm for use with rectangularly sampled images. Both algorithms produced good skeletons, but the hexagonal could be implemented with only 50% of the logical operations required by the rectangular View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computerized classification of breast lesions: shape and texture analysis using an artificial neural network

    Page(s): 517 - 521 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (244 KB)  

    In this study we have investigated two groups of features: shape features, extracted from microcalcifications and texture features, extracted from the original regions of interest (ROI), in order to classify early breast cancers which have microcalcifications associated. These features were analyzed using different topologies of the artificial neural network (ANN) multi-layer perceptron (MLP). The performance of the ANN was analyzed with receiver operating characteristic (ROC) methodology View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Temporal knowledge for cooperative distributed vision

    Page(s): 726 - 730 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (300 KB)  

    The development of a distributed vision system (DVS) performing continuous surveillance is an interesting area of investigation. We are interested in the interpretation of wide areas, this constraint means that numerous sensors have to be distributed in space and have to cooperate in order to obtain a global interpretation. In the European community the SMART project, close to the video surveillance and monitoring (VSAM) field has been defined. The main area of research concerns: 2/3D object and event recognition, sensor fusion and active perception. The final objective of this work is the ability to track multiple objects in a wide outdoor area. Also in general, it is not possible to cover the whole of a scene, so the sensors are generally separated into blind zones, for which we do not have any observation. One of the principal difficulties is to ensure a robust recognition of the mobile objects perceived by the different sensors from different points of view at different moments. We decided to use fuzzy temporal curves of events (DOP: domain occurrence possibility) described by Dubois and Prade (1989) in order to predict object motion in blind zones View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A common image processing framework for 2D barcode reading

    Page(s): 652 - 655 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (208 KB)  

    The present work describes an image processing system able to locate, segment and decode the most common 2D symbol used in bar code applications. The different symbol are treated by exploiting their similarities, in order to achieve an unified computational structure View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • List of authors

    Page(s): xxviii - xxxii
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (168 KB)  

    First Page of the Article
    View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new modal matching algorithm based on dynamically located sub-regions for point feature correspondence

    Page(s): 716 - 720 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (284 KB)  

    The correspondence problem is often encountered in computer vision. In particular, point feature matching of two images is a critical aspect in: stereo vision, structure-from-motion problems, and moving object tracking in image sequences. Besides, it is one possible approach to define similarity metrics among shapes in content-based retrieval for pictorial databases. This paper proposes an algorithm for reliably matching a large number of point features between two images by means of modal matching, and introduces a new method for the point-matching phase that is the most critical step when dealing with large deformations. The proposed algorithm performs a reliable and exhaustive point-feature association between shapes. The robustness is achieved by limiting risky situations using the principle of locality of associations. The approach does not use any distance based control, therefore it avoids any ad hoc threshold and copes effectively even with very large deformations. Practical experiments have been done for content based image retrieval, and for tracking clouds movements in remote sensing image sequences. In both these applications very promising results have been obtained View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Diagnostic quality test for wavelet-compressed digital angiogram images

    Page(s): 512 - 516 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (216 KB)  

    In this paper we discuss the requirements of medical image compression methods and present results from diagnostic quality tests of lossily wavelet-compressed digital angiogram images View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Analysis of gamma-ray nuclear resonant absorption (NRA) images for automatic explosives detection

    Page(s): 789 - 793 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (492 KB)  

    A method developed for explosives detection via NRA image analysis is discussed. The results of processing simulated images, as well as experimental NRA-images are presented. The experimental images were obtained by scanning aviation cargo along with mock explosives provided by the US Federal Aviation Administration (FAA) using the NRA system developed at Soreq NRC. NRA is a novel transmission radiography technique that was originally developed at Soreq NRC for airline-baggage and cargo-container inspection View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automatic line extraction by the square scan algorithm

    Page(s): 832 - 835 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (184 KB)  

    This paper describes a new image processing algorithm for the automatic extraction of line features from raster images with width variation. The features are transferred into vector format. The algorithm has capabilities for extracting crossing and branching lines. In addition it has the ability to connect disconnected lines. The algorithm is implemented and also tested on several image cases View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Hierarchical image segmentation using a watershed scale-space tree

    Page(s): 522 - 526 vol.2
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (336 KB)  

    The watershed transformation is a useful morphological segmentation tool which has been used in a variety of grey-scale image processing applications. However, a major problem with the watershed transformation is that it produces a severe over-segmentation due to the great number of minima embedded in the image or its gradient, and therefore it is rarely applied directly to images. In this paper we discuss modifications to the basic watershed transformation that enable watershed scale trees to be produced and we illustrate the approach with some example segmentations taken from a medical image processing application. The mapping between image and scale tree allows the user to overcome the over-segmentation by either setting a global threshold or interactively editing the tree description of the image View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.