Showing papers on "Range segmentation published in 2003"

Image segmentation with ratio cut

Abstract: This paper proposes a new cost function, cut ratio, for segmenting images using graph-based methods. The cut ratio is defined as the ratio of the corresponding sums of two different weights of edges along the cut boundary and models the mean affinity between the segments separated by the boundary per unit boundary length. This new cost function allows the image perimeter to be segmented, guarantees that the segments produced by bipartitioning are connected, and does not introduce a size, shape, smoothness, or boundary-length bias. The latter allows it to produce segmentations where boundaries are aligned with image edges. Furthermore, the cut-ratio cost function allows efficient iterated region-based segmentation as well as pixel-based segmentation. These properties may be useful for some image-segmentation applications. While the problem of finding a minimum ratio cut in an arbitrary graph is NP-hard, one can find a minimum ratio cut in the connected planar graphs that arise during image segmentation in polynomial time. While the cut ratio, alone, is not sufficient as a baseline method for image segmentation, it forms a good basis for an extended method of image segmentation when combined with a small number of standard techniques. We present an implemented algorithm for finding a minimum ratio cut, prove its correctness, discuss its application to image segmentation, and present the results of segmenting a number of medical and natural images using our techniques.

Active unsupervised texture segmentation on a diffusion based feature space

Abstract: We propose a novel and efficient approach for active unsupervised texture segmentation. First, we show how we can extract a small set of good features for texture segmentation based on the structure tensor and nonlinear diffusion. Then, we propose a variational framework that incorporates these features in a level set based unsupervised segmentation process that adaptively takes into account their estimated statistical information inside and outside the region to segment. The approach has been tested on various textured images, and its performance is favorably compared to recent studies.

Automated feature-based range registration of urban scenes of large scale

Abstract: We are building a system that can automatically acquire 3D range scans and 2D images to build geometrically and photometrically correct 3D models of urban environments. A major bottleneck in the process is the automated registration of a large number of geometrically complex 3D range scans in a common frame of reference. In this paper we provide a method for the accurate and efficient registration of a large number of complex range scans. The method utilizes range segmentation and feature extraction algorithms. Our algorithm automatically computes pairwise registrations between individual scans, builds a topological graph, and places the scans in the same frame of reference. We present results for building large scale 3D models of historic sites and urban structures.

Dynamic Texture Segmentation

Abstract: We address the problem of segmenting a sequence of imagesof natural scenes into disjoint regions that are characterizedby constant spatio-temporal statistics. We model thespatio-temporal dynamics in each region by Gauss-Markovmodels, and infer the model parameters as well as theboundary of the regions in a variational optimization framework.Numerical results demonstrate that - in contrast topurely texture-based segmentation schemes - our method iseffective in segmenting regions that differ in their dynamicseven when spatial statistics are identical.

A spatial constrained K-means approach to image segmentation

Abstract: General purposed color image segmentation is a challenging and important issue in image processing related applications. However, few systems successfully handle this issue for a broad diversity of images. In this paper, we are seeking a practical and generic solution to image segmentation. As a fast segmentation process, K-means based clustering is employed in feature space first. Then, in image plane, the spatial constraints are adopted into the hierarchical K-means clusters on each level. The two processes are carried out alternatively and iteratively. Also, an effective region merging method is proposed to handle the over segmentation. Extensive experiments show the proposed approach is fast and generic, thus practical in applications.

Robust background estimation for complex video sequences

Abstract: Knowing the background image of a video scene simplifies the general video-object segmentation problem and therefore it is required by several automatic segmentation algorithms. This paper presents a new background estimation algorithm which is applicable to complex video sequences where many objects are simultaneously visible and the background is visible for a short time period only. The algorithm applies a rough segmentation of the input images into foreground and background regions to exclude the foreground objects from background synthesis. This prevents a bias of the synthesized background image towards the color of foreground objects. Experiments show that the obtained background images differ significantly less from the real background than those obtained with previous algorithms.

Unsupervised image segmentation using local homogeneity analysis

Abstract: In this paper, a novel method is presented for unsupervised image segmentation based on local homogeneity analysis. First, a criterion for homogeneity of a certain pattern is proposed. Applying the criterion to local windows in the original image results in the "H-image". The high and low values of the H-image correspond to possible region boundaries and region interiors respectively. Then, a region growing method is used to segment the image based on the H-image. Finally, visually similar regions are merged together to avoid over-segmentation. Experimental results on real images show the effectiveness and robustness of the method.

Method and apparatus for enhancing an image using data optimization and segmentation

Abstract: A method and system are disclosed for generating enhanced images of multiple dimensional data using a depth-buffer segmentation process. The method and system operate in a computer system modify the image by generating a reduced-dimensionality image data set from a multidimensional image by formulating a set of projection paths through image points selected from the multidimensional image, selecting an image point along each projection path, analyzing each image point to determine spatial similarities with at least one other point adjacent to the selected image point in a given dimension, and grouping the image point with the adjacent point or spatial similarities between the points is found thereby defining the data set.

Tuning range image segmentation by genetic algorithm

Abstract: Several range image segmentation algorithms have been proposed, each one to be tuned by a number of parameters in order to provide accurate results on a given class of images. Segmentation parameters are generally affected by the type of surfaces (e.g., planar versus curved) and the nature of the acquisition system (e.g., laser range finders or structured light scanners). It is impossible to answer the question, which is the best set of parameters given a range image within a class and a range segmentation algorithm? Systems proposing such a parameter optimization are often based either on careful selection or on solution space-partitioning methods. Their main drawback is that they have to limit their search to a subset of the solution space to provide an answer in acceptable time. In order to provide a different automated method to search a larger solution space, and possibly to answer more effectively the above question, we propose a tuning system based on genetic algorithms. A complete set of tests was performed over a range of different images and with different segmentation algorithms. Our system provided a particularly high degree of effectiveness in terms of segmentation quality and search time.

Method and apparatus for low depth of field image segmentation

Abstract: A method for extracting an object of interest from an image is provided. The method initiates with defining an image feature space based upon frequency information. Then, the image feature space is filtered to smooth both focused regions and defocused regions while maintaining respective boundaries associated with the focused regions and the defocused regions. The filtered image feature space is manipulated by region merging and adaptive thresholding to extract an object-of-interest. A computer readable media, an image capture device and an image searching system are also provided.

Motion segmentation with accurate boundaries - a tensor voting approach

Abstract: Producing an accurate motion flow field is very difficult at motion boundaries. We present a noniterative approach for segmentation from image motion, based on two voting processes, in different dimensional spaces. By expressing the motion layers as surfaces in a 4D (four-dimensional) space, a voting process is first used to enforce the smoothness of motion and determine an estimation of pixel velocities, motion regions and boundaries. The boundary estimation is then combined with intensity information from the original images in order to locally define a boundary tensor field. The correct boundary is inferred by a 2D (two-dimensional) voting process within this field that enforces the smoothness of boundaries. Finally, correct velocities are computed for the pixels near boundaries, as they are reassigned to different regions. We demonstrate our contribution by analyzing several image sequences, containing multiple types of motion.

Pattern recognition in the automatic inspection of aluminium castings

Abstract: ∑ Image Formation: The images are obtained by X-ray irradiation of the test-piece. The X-rays are then converted to a visible image by means of an image amplifier or a flat-panel detector that are sensitive to X-rays. The sensor is bi-dimensional (or unidimensional in motion) in order to capture the two dimensions of the image. An A/D converter turns the electrical signal into binary code that can be interpreted by a computer to form a digital image of the study object. ∑ Pre-processing: This stage is devoted to improving the quality of the image in order to better recognise flaws. Some of the techniques used in this stage are elimination of noise by means of digital filters or integration, improvement of contrast, and restoration. ∑ Segmentation: The segmentation process divides the digital image into disjoint regions with the purpose of separating the parts of interest from the rest of the scene. Over the last few decades, diverse segmentation techniques have been developed. These can largely be divided into three groups: pixel, edge and region orientated techniques. The present investigation uses the segmentation process oriented towards the detection of edges by employing the LoG filter (Mery and Filbert, 2002b). As can be seen in Figure 2, this technique searches for changes in the grey values of the image (edges), thus identifying zones delimited by edges that indicate flaws. ∑ Feature extraction: In the inspection of cast pieces, segmentation detects regions that are denominated as ‘hypothetical defects’, which may be flaws or structural features of the object. Subsequently the feature extraction is centred principally around

Image processing apparatus, image processing method, program, recording medium, and image forming apparatus having the same

Abstract: An edge pixel is extracted from an input image by an edge determination section, an object image section is extracted by an object-image extraction section, and a first characteristic amount of the object image section in a first local pixel block containing a first target pixel is calculated by a first-characteristic-amount calculation section. A classification is made as to whether the first target pixel is a character edge pixel or a dot edge pixel by an edge-class determination section on the basis of the result of extraction for the edge pixel and the first characteristic amount. A second characteristic amount of a second local pixel block containing a second target pixel is calculated by a second-characteristic-amount calculation section on the basis of the edge classification result. The image of the second target pixel is classified by an image-class determination section on the basis of the second characteristic amount.

Stereo matching using segmentation of image columns

Abstract: In one embodiment of the present invention, a method includes grouping pixels of columns of a first image of an image pair into column segments; and creating a disparity map for the image pair using the column segments.

An improved image segmentation approach based on level set and mathematical morphology

Abstract: Level set methods offer a powerful approach for the medical image segmentation since it can handle any of the cavities, concavities, convolution, splitting or merging. However, this method requires specifying initial curves and can only provide good results if these curves are placed near symmetrically with respect to the object boundary. Another well known segmentation technique - morphological watershed transform can segment unique boundaries from an image, but it is very sensitive to small variations of the image magnitude and consequently the number of generated regions is undesirably large and the segmented boundaries is not smooth enough. In this paper, a hybrid 3D medical image segmentation algorithm, which combines the watershed transform and level set techniques, is proposed. This hybrid algorithm resolves the weaknesses of each method. An initial partitioning of the image into primitive regions is produced by applying the watershed transform on the image gradient magnitude, then this segmentation results is treated as the initial localization of the desired contour, and used in the following level set method, which provides closed, smoothed and accurately localized contours or surfaces. Experimental results are also presented and discussed.

Image segmentation using multi-region stability and edge strength

Abstract: A novel scheme for image segmentation is presented. An image segmentation criterion is proposed that groups similar pixels together to form regions. This criterion is formulated as a cost function. Using gradient-descent methods, which lead to a curve evolution equation that segments the image into multiple homogenous regions, minimizes this cost function. Homogeneity is specified through a pixel-to-pixel similarity measure, which is defined by the user and can be adaptive based on the current application. To improve the performance of the system, an edge function is also used to adjust the speed of the competing curves. The proposed method can be easily applied to vector valued images such as texture and color images without a significant addition to computational complexity.

Image segmentation using template prediction

Abstract: The invention relates to image segmentation using templates and spatial prediction. The templates of neighboring pixels are used for predicting the features of a current pixel. The pixel is assigned to the segments of neighboring pixels according to the deviation of its features from the templates.

Method for segmenting an image

Abstract: A method for isolating an element of an image made up of pixels comprising the steps of classifying the pixels into different groups based on the color value of the pixel, blurring the image, locating a pixel in the blurred image that has a predetermined color value corresponding to the element to be isolated, and growing a mask from the located pixel

Region-based active contours using geometrical and statistical features for image segmentation

Abstract: The problem of image segmentation through the minimization of an energy criterion involving both region and boundary functionals is considered. We study the derivation of these functionals using the notion of shape derivative. From the derivative, we deduce the evolution equation of an active contour that will make it evolve towards a minimum of the criterion introduced. We focus on geometric and statistical features globally attached to the boundary or to the region, and we take explicitly into account their evolution in the derivation. First, statistical region-based descriptors using the variance of a region or the distance to a reference region histogram are introduced. Then a geometric prior term is combined with statistical features for homogeneous region segmentation. This geometric prior is introduced to provide a free form deformation from a reference shape. Some experimental results on real images and video sequences show the benefit of combining geometrical and statistical features for segmentation.

Natural color image segmentation

Abstract: A new method for natural color image segmentation using integrated features is proposed in this paper. Edges are first detected in term of the high phase congruency in the gray-level image. K-means cluster is used to label long edge lines based on the global color information to estimate roughly the distribution of objects in the image, while short ones are merged based on their positions and local color differences to eliminate the negative affection caused by texture or other trivial features in image. Region growing technique is employed to achieve the final segmentation results. The proposed method unifies edges, both the whole and local color distributions, as well as the spatial information to solve the natural image segmentation problem. The feasibility and effectiveness of this method have been demonstrated by various experiments.

Two-Level Image Segmentation Based on Region and Edge Integration

Abstract: This paper introduces a two-level approach for image seg- mentation based on region and edge integration. Edges are first detected in the original image using a combination of operators for intensity gra- dient and texture discontinuities. To preserve the spatial coherence of the edges and their surrounding image regions, the detected edges are vectorized into connected line segments which serve as the basis for a constrained Delaunay triangulation. Segmentation is first performed on the triangulation using graph cuts. Our method favors segmentations that pass through more vectorized line segments. Finally, the obtained segmentation on the triangulation is projected onto the original image and region boundaries are refined to achieve pixel accuracy. Experimen- tal results show that the two-level approach can achieve accurate edge localization, better spatial coherence and improved efficiency.

Segmentation tool for identifying flow regions in an imaging system

Abstract: An ultrasound system and method that identify flow regions within a volume. The system comprises: a survey system for collecting motion data from a target image; a segmentation system for mapping a region of flow within the image based on the motion data; and a flow acquisition system that automatically limits the collection of flow image data within the image to the region of flow.

New discrepancy measures for segmentation evaluation

Abstract: In this paper, we propose new evaluation measures for scene segmentation results, which are based on computing the difference between a region extracted from a segmentation map and the corresponding one on an ideal segmentation. The proposed measures take into account separately both under and over detected pixels. It also associates in its computation the compactness of the region under investigation.

A segmentation algorithm for contrast-enhanced images

Abstract: Medical imaging often involves the injection of contrast agents and the subsequent analysis of tissue enhancement patterns. Many important types of tissue have characteristic enhancement patterns; for example, in magnetic resonance (MR) mammography, malignancies exhibit a characteristic "wash out" temporal pattern, while in MR angiography, arteries, veins and parenchyma each have their own distinctive temporal signature. In such image sequences, there are substantial changes in intensities; however, this change is due primarily to the contrast agent rather than the motion of scene elements. As a result, the task of segmenting contrast-enhanced images poses interesting new challenges for computer vision. We propose a new image segmentation algorithm for image sequences with contrast enhancement, using a model-based time series analysis of individual pixels. We use energy minimization via graph cuts to efficiently ensure spatial coherence. The energy is minimized in an expectation-maximization fashion that alternates between segmenting the image into a number of nonoverlapping regions and finding the temporal profile parameters which best describe the behavior of each region. Preliminary experiments on MR mammography and MR angiography studies show the algorithm's ability to find an accurate segmentation.

Fast SAR image restoration, segmentation, and detection of high-reflectance regions

Abstract: An iterative filter that can be used for speckle reduction and restoration of synthetic aperture radar (SAR) images is presented here. This method can be considered as a first step in the extraction of other important information. The second step is the detection of high-reflectance regions and continues with the segmentation of the total image. We have worked in three-look simulated and real European Remote Sensing 1 satellite amplitude images. The iterative filter is based on a membrane model Markov random field approximation optimized by a synchronous local iterative method. The final form of restoration gives a total sum-preserving regularization for the pixel values of our image. The high-reflectance regions are defined as the brightest regions of the restored image. After the separation of this extreme class, we give a fast segmentation method using the histogram of the restored image.

Method of processing an image to form an image pyramid

Abstract: A method of processing an image to form an image pyramid having multiple image levels includes receiving a base level image comprising pixel values at pixel locations arranged in rows and columns; determining sample locations for a next level image in the pyramid such that the sample locations are arranged in a regular pattern and the sample locations exceed the range of the pixel locations of the base level image; determining the pixel values of the next level image by interpolating the pixel values of the base level image using an interpolation filter at the sample locations; and treating the next level image as the base level image and repeating steps of determining sample locations and pixel values until a predetermined number of pyramid image levels are generated, or until a predetermined condition is met.

Fingerprint Images Segmentation Using Two Stages Coarse to Fine Discrimination Technique

Abstract: Segmentation of fingerprint image is necessary to reduce the size of the input data, eliminating undesired background, which is the noisy and smudged area in favor of the central part of the fingerprint. In this paper, an algorithm for the segmentation which uses two stages coarse to fine approach is presented. The coarse segmentation will be performed at first using the orientation certainty values that derived from the blockwise directional field of the fingerprint image. The coarse segmented image will be carry on to the second stage which consist Fourier based enhancement and adaptive thresholding. Orientation certainty values of the resultant binarized image are calculated once again to perform the fine segmentation. Finally, binary image processing is applied as postprocessing to further reduce the segmentation error. Visual inspection shows that the proposed method produce accurate segmentations result. The algorithm is also evaluated by counting the number of false and missed detected center points and compare with the fingerprint image which have no segmentation and with the proposed method without postprocessing. Experiments show that the proposed segmentation method perform well than others.