A new method for image segmentation
01 Nov 2009-Vol. 2, pp 123-125
TL;DR: A new segmentation method which is based on the morphology method, fuzzy K-means algorithm and some parts operator of the Canny algorithm, and the course of Canny operator that calculating the value and direction of grads, non-maxima suppression to the grad value and lag threshold process into the post-treatment process is introduced.
Abstract: On the basis of analyzing the blur images with noise, this paper presents a new segmentation method which is based on the morphology method, fuzzy K-means algorithm and some parts operator of the Canny algorithm. Because of the Canny's good performance on good detection, good localization and only one response to a single edge, we introduce the course of Canny operator that calculating the value and direction of grads, non-maxima suppression to the grad value and lag threshold process into our post-treatment process. Through experiments, it is demonstrated that the image segmentation method in this paper is very effective.
Citations
More filters
14 Mar 2012
TL;DR: This chapter deals with digital restoration, preservation, and data base storage of historical manuscripts images by focusing on restoration techniques and binarization methods combined with image processing applied on document images for text background enhancement and discrimination.
Abstract: This chapter deals with digital restoration, preservation, and data base storage of historical manuscripts images. It focuses on restoration techniques and binarization methods combined with image processing applied on document images for text background enhancement and discrimination. Sequential image processing procedures are applied for image refinement and enhancement on quality class categorized images. Research results on historical (i.e. Byzantine, old newspapers, etc) manuscripts are presented.
3 citations
Cites background from "A new method for image segmentation..."
...For textual images published on the web site, enlarged and printed to larger dimensions than 10″ x 8″ compression may provide inadequate quality....
[...]
Dissertation•
18 Jul 2016
TL;DR: L’echelle de l’observation est modifiee a un niveau inferieur afin d’affiner les resultats and selectionner uniquement ceux qui sont vraiment pertinents dans le document teste.
Abstract: Dans cette these, nous avons propose une approche analytique multi-echelle pour le word spotting dans les documents manuscrits Le modele propose fonctionne selon deux niveaux differents Un module de filtrage global permettant de definir plusieurs zones candidates de la requete dans le document teste Ensuite, l’echelle de l’observation est modifiee a un niveau inferieur afin d’affiner les resultats et selectionner uniquement ceux qui sont vraiment pertinents Cette approche de word spotting est basee sur des familles generalisees de filtres de Haar qui s’adaptent a chaque requete pour proceder au processus de spotting et aussi sur un principe de vote qui permet de choisir l’emplacement spatial ou les reponses generees par les filtres sont accumulees Nous avons en plus propose une autre approche pour l’extraction de texte du graphique dans les bandes dessinees Cette approche se base essentiellement sur les caracteristiques pseudo-Haar qui sont generees par l’application des filtres generalises de Haar sur l’image de bande dessinee Cette approche est une approche analytique et ne necessite aucun processus d’extraction ni des bulles ni d’autres composants
3 citations
TL;DR: In this paper, the authors propose a simple way to use an X-ray tomogram to infer local elastic properties by solving the mechanical equilibrium, and compare the up-scaled elastic moduli to indentation experiments performed at the same scale.
Abstract: Macroscopic mechanical properties of materials depend directly on their microstructure. Microscopy, and more specifically tomography, is a key method for studying microstructures. Here, we propose a simple way to use an X-ray tomogram to infer local elastic properties. We distinguish between two scenarios of microstructure images. In the first scenario, the material is composed by very apparent phases so the image can be easily segmented into a set of subspaces with homogenous properties. In the second scenario, the image, as that of sedimentary rocks, contains poorly contrasted phases, including strong intra-phase heterogeneities. For this case, we propose an alternative to segmentation techniques in order to factor in material heterogeneities. To do this, we use the local X-ray attenuation to define elastic moduli. Then, we compute up-scaled elastic moduli by solving the mechanical equilibrium. Finally, we confirm our method by comparing the up-scaled elastic moduli to indentation experiments performed at the same scale.
3 citations
Dissertation•
01 Nov 2011
TL;DR: This thesis designs a robust feature named connected component descriptor that is tailored for mosaicing camera-captured document images and addresses two critical issues often encountered in correspondence matching, the stability of features and robustness against false matches due to multiple instances of many characters in a document image.
Abstract: Text is no longer confined to scanned pages and often appears in camera-based images originating from text on real world objects. Unlike the images from conventional flatbed scanners, which have a controlled acquisition environment, camera-based images pose new challenges such as uneven illumination, blur, poor resolution, perspective distortion and 3D deformations that can severely affect the performance of any optical character recognition (OCR) system. Due to the variations in the imaging condition as well as the target document type, traditional OCR systems, designed for scanned images, cannot be directly applied to camera-captured images and a new level of processing needs to be addressed. In this thesis, we study some of the issues commonly encountered in camera-based image analysis and propose novel methods to overcome them. All the methods make use of color connected components.
1. Connected component descriptor for document image mosaicing
Document image analysis often requires mosaicing when it is not possible to capture a large document at a reasonable resolution in a single exposure. Such a document is captured in parts and mosaicing stitches them into a single image. Since connected components (CCs) in a document image can easily be extracted regardless of the image rotation, scale and perspective distortion, we design a robust feature named connected component descriptor that is tailored for mosaicing camera-captured document images. The method involves extraction of a circular measurement region around each CC and its description using the angular radial transform (ART). To ensure geometric consistency during feature matching, the ART coefficients of a CC are augmented with those of its 2 nearest neighbors. Our method addresses two critical issues often encountered in correspondence matching: (i) the stability of features and (ii) robustness against false matches due to multiple instances of many characters in a document image. We illustrate the effectiveness of the proposed method on camera-captured document images exhibiting large variations in viewpoint, illumination and scale.
2. Font and background color independent text binarization
The first step in an OCR system, after document acquisition, is binarization, which converts a gray-scale/color image into a two-level image -the foreground text and the background. We propose two methods for binarization of color documents whereby the foreground text is output as black and the background as white regardless of the polarity of foreground-background shades.
(a) Hierarchical CC Analysis: The method employs an edge-based connected component approach and automatically determines a threshold for each component. It overcomes several limitations of existing locally-adaptive thresholding techniques. Firstly, it can handle documents with multi-colored texts with different background shades. Secondly, the method is applicable to documents having text of widely varying sizes, usually not handled by local binarization methods. Thirdly, the method…
3 citations
Cites background from "A new method for image segmentation..."
...Yanowitz and Bruckstein [39] introduced a threshold that varies over different image regions so as to fit the spatially changing background and lighting conditions....
[...]
Journal Article•
TL;DR: In this paper local thresholding method Nilblack method with post processing was implemented and tested with a sample of ground tooth images selected from TOBACCO research database.
Abstract: Binarization is the preliminary process of Document Image Analysis and Processing. Image binarization is performed through Local and Global threshold methods. In this paper local thresholding method Nilblack method with post processing was implemented. The Nilblack algorithm was implemented using Matlab and tested with a sample of ground tooth images selected from TOBACCO research database.
3 citations
References
More filters
TL;DR: There is a natural uncertainty principle between detection and localization performance, which are the two main goals, and with this principle a single operator shape is derived which is optimal at any scale.
Abstract: This paper describes a computational approach to edge detection. The success of the approach depends on the definition of a comprehensive set of goals for the computation of edge points. These goals must be precise enough to delimit the desired behavior of the detector while making minimal assumptions about the form of the solution. We define detection and localization criteria for a class of edges, and present mathematical forms for these criteria as functionals on the operator impulse response. A third criterion is then added to ensure that the detector has only one response to a single edge. We use the criteria in numerical optimization to derive detectors for several common image features, including step edges. On specializing the analysis to step edges, we find that there is a natural uncertainty principle between detection and localization performance, which are the two main goals. With this principle we derive a single operator shape which is optimal at any scale. The optimal detector has a simple approximate implementation in which edges are marked at maxima in gradient magnitude of a Gaussian-smoothed image. We extend this simple detector using operators of several widths to cope with different signal-to-noise ratios in the image. We present a general method, called feature synthesis, for the fine-to-coarse integration of information from operators at different scales. Finally we show that step edge detector performance improves considerably as the operator point spread function is extended along the edge.
28,073 citations
"A new method for image segmentation..." refers methods in this paper
...Canny operator[2] transforms the edge detection problem into the problem of unit function maximum detection....
[...]
TL;DR: This work presents a simple and efficient implementation of Lloyd's k-means clustering algorithm, which it calls the filtering algorithm, and establishes the practical efficiency of the algorithm's running time.
Abstract: In k-means clustering, we are given a set of n data points in d-dimensional space R/sup d/ and an integer k and the problem is to determine a set of k points in Rd, called centers, so as to minimize the mean squared distance from each data point to its nearest center. A popular heuristic for k-means clustering is Lloyd's (1982) algorithm. We present a simple and efficient implementation of Lloyd's k-means clustering algorithm, which we call the filtering algorithm. This algorithm is easy to implement, requiring a kd-tree as the only major data structure. We establish the practical efficiency of the filtering algorithm in two ways. First, we present a data-sensitive analysis of the algorithm's running time, which shows that the algorithm runs faster as the separation between clusters increases. Second, we present a number of empirical studies both on synthetically generated data and on real data sets from applications in color quantization, data compression, and image segmentation.
5,288 citations
"A new method for image segmentation..." refers methods in this paper
...Fuzzy K-means algorithm[3] that divides the samples on various categories of membership according to the data is a clustering method in more common use....
[...]
Book•
15 Sep 1994
TL;DR: The fundamental principles of Digital Image Processing are explained, as well as practical suggestions for improving the quality and efficiency of image processing.
Abstract: What Is Image Processing?. Fundamentals of Digital Image Processing. The Digital Image. PROCESSING CONCEPTS. Image Enhancement and Restoration. Image Analysis. Image Compression. Image Synthesis. PROCESSING SYSTEMS. Image Origination and Display. Image Data Handling. Image Data Processing. PROCESSING IN ACTION. Image Operation Studies. Appendices. Glossary. Index.
457 citations
12 May 1998
TL;DR: A novel method for measuring the orientation of an edge is introduced and it is shown that it is without error in the noise-free case, and the wreath product transform edge detection performance is shown to be superior to many standard edge detectors.
Abstract: Wreath product group based spectral analysis has led to the development of the wreath product transform, a new multiresolution transform closely related to the wavelet transform. We derive the filter bank implementation of a simple wreath product transform and show that it is in fact, a multiresolution Roberts (1965) Cross edge detector. We also derive the relationship between this transform and the two-dimensional Haar wavelet transform. We prove that, using a non-traditional metric for measuring edge amplitude with the wreath product transform, yields a rotation and translation invariant edge detector. We introduce a novel method for measuring the orientation of an edge and show that it is without error in the noise-free case. The wreath product transform edge detection performance is shown to be superior to many standard edge detectors.
19 citations