By Topic

Digital Image Processing, 2009 International Conference on

Date 7-9 March 2009

Filter Results

Displaying Results 1 - 25 of 95
  • [Front cover]

    Page(s): C1
    Save to Project icon | Request Permissions | PDF file iconPDF (785 KB)  
    Freely Available from IEEE
  • [Title page i]

    Page(s): i
    Save to Project icon | Request Permissions | PDF file iconPDF (61 KB)  
    Freely Available from IEEE
  • [Title page iii]

    Page(s): iii
    Save to Project icon | Request Permissions | PDF file iconPDF (141 KB)  
    Freely Available from IEEE
  • [Copyright notice]

    Page(s): iv
    Save to Project icon | Request Permissions | PDF file iconPDF (110 KB)  
    Freely Available from IEEE
  • Table of contents

    Page(s): v - xi
    Save to Project icon | Request Permissions | PDF file iconPDF (165 KB)  
    Freely Available from IEEE
  • Preface

    Page(s): xii
    Save to Project icon | Request Permissions | PDF file iconPDF (71 KB)  
    Freely Available from IEEE
  • Organizing Committee

    Page(s): xiii
    Save to Project icon | Request Permissions | PDF file iconPDF (87 KB)  
    Freely Available from IEEE
  • Technical Committee

    Page(s): xiv
    Save to Project icon | Request Permissions | PDF file iconPDF (89 KB)  
    Freely Available from IEEE
  • Reviewers

    Page(s): xv
    Save to Project icon | Request Permissions | PDF file iconPDF (88 KB)  
    Freely Available from IEEE
  • A Combination Scheme for Fuzzy Partitions Based on Fuzzy Weighted Majority Voting Rule

    Page(s): 3 - 7
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (283 KB) |  | HTML iconHTML  

    This paper devotes to the combination of fuzzy partitions with the same number of clusters by means of generalizing the weighted majority voting rule to fuzzy weighted majority voting rule. The difficulties of this generalization are to establish the correspondences among the classes and determine the weight coefficients of component fuzzy partitions. We propose a class-matching algorithm based on Hungarian method and generalize pattern recognition rate to fuzzy pattern recognition rate to overcome the difficulties. Employing the proposed class-matching algorithm and the fuzzy weighted majority voting rule, a combining scheme for fuzzy partitions is developed. Experimental results on real datasets show that the proposed ensemble of fuzzy partitions outperforms or is comparable to other two existed ensembles of fuzzy partitions in terms of most evaluation indexes for fuzzy partition. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • On Line Wavelets Transform on a Xilinx FPGA Circuit to Medical Images Compression

    Page(s): 8 - 12
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (339 KB) |  | HTML iconHTML  

    Knowing that, the computing process of the S.Mallat Transform algorithm is characterized by a purely sequential structure, and from the fact, the on line mode arithmetic is more suitable for the computation of this kind of operations. We propose in this paper, a new wavelet Transform algorithm and a suitable architecture implemented on a Xilinx FPGA circuit. In this study, we will show how on line arithmetic is used to implement a pipelined architecture of the S.Mallat Transform and we will demonstrate through different implementations under different medical image and different computation mode that it might be used successfully for medical image compression. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automatic Counting of Leukocytes in Giemsa-Stained Images of Peripheral Blood Smear

    Page(s): 13 - 16
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (686 KB) |  | HTML iconHTML  

    There are many different classes of leukocyte in peripheral blood image. Leukocyte count is used to determine the presence of an infection in the human body. To be able to observe and recognize the different kinds of leukocyte, you must stain them. For this purpose, normally Giemsa stain is used. There are two difficult issues in image segmentation which common segmentation algorithms can not overcome them. Nucleus which is laid inside white cell is the darkest part of image which can be used to count cells. Since Giemsa staining is done by humans, intensity of images is slightly different from each others. Neutrophils are kinds of leukocytes which have segmented and distinctive nucleus. These reasons cause a considerable error in counting. In this paper, we have proposed to use histogram of images and intensity of red cells which are major objects in images to select appropriate point for thresholding. And then the distances among centers of the extracted nuclei have been calculated, according to the specified size of leukocytes, we merge the nuclei which those distances are less than the diameter of one leukocyte. Experimental results show that our approach is very efficient. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Novel Color Space Creating Method Applied to Skin Color Detection

    Page(s): 17 - 21
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (153 KB) |  | HTML iconHTML  

    This each three-component collection such as {red, green, blue} (RGB), and {luminance Y, chrominance Cr, chrominance Cb} (YCbCr) is termed as a color space. Many color spaces are related to each other by linear transformations that are captured by 3 times 3 matrices. Hence a given color, and thereby any color image, can be represented in terms of another color space by transforming its 3-d vector representation using the 3 times 3 matrix. The Main target of this paper is introduce new color transform from viewpoint of convex constraint programming. Skin detection is used as benchmark problem for the proposed algorithm. In the New color space, the skin and non-skin classes are separated as well. This problem is converted to a convex constraint programming which Lagrange multipliers method is used for solving this problem. Founded converting matrix is tested in skin detection in simple to complex scene. Obtained results over many databases are compared with existing methods which show superiority of the proposed method. Skin and non-skin clusters in the new space color have clustering criteria better than RGB and YCbCr color space. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An Image Segmentation Algorithm Based on Fuzzy C-Means Clustering

    Page(s): 22 - 26
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (343 KB) |  | HTML iconHTML  

    Image segmentation algorithm based on fuzzy c-means clustering is an important algorithm in the image segmentation field. It has been used widely. However, it is not successfully to segment the noise image because the algorithm disregards of special constraint information. It only considers the gray information. Therefore, we proposed a weighed FCM algorithm based on Gaussian kernel function for image segmentation. The original Euclidean distance is replaced by a kernel-induced distance in the algorithm. Then, a bound term is added to the objective function to compensate the influence of the spatial information. The experimental results illustrate that the proposed method is more effective to image segmentation. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Multilevel Image Reconstruction by Interpolating Wavelet Coefficients

    Page(s): 27 - 33
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (254 KB) |  | HTML iconHTML  

    High resolution (HR) image reconstruction produces one high-resolved image from a set of low-resolution (LR) images which may be blurred, degraded and shifted. In this paper, a wavelet-based multilevel Heritage reconstruction method is proposed to recover the HR image from the estimated middle-resolution (MR) data which are obtained from the observed LR images. The relationship between the wavelet subbands and the LR images is extensively investigated. Based on this relationship, an iterative balanced reconstruction with error correction approach is developed in the low-level reconstruction process to estimate the MR data from the LR images. The numerical experiments show that the proposed method outperforms the state of the art methods such as Tikhonov least-squares approach and Chan et al. Algorithm 3. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Blood Vessel Enhancement and Segmentation Using Wavelet Transform

    Page(s): 34 - 38
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1389 KB) |  | HTML iconHTML  

    Retinal vessel segmentation is an essential step for the diagnoses of various eye diseases. An automated tool for blood vessel segmentation is useful to eye specialists for purpose of patient screening and clinical study. Vascular pattern is normally not visible in retinal images. In this paper, we present a method for enhancing, locating and segmenting blood vessels in images of retina. We present a method that uses 2-D Gabor wavelet and sharpening filter to enhance and sharpen the vascular pattern respectively. This technique locates and segments the blood vessels using edge detection algorithm and morphological operations. This technique is tested on publicly available STARE database of manually labeled images which has been established to facilitate comparative studies on segmentation of blood vessels in retinal images. The validation of our retinal image vessel segmentation technique is supported by experimental results. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Image Retrieval Using Fuzzy and Neuro-fuzzy Approaches with Fuzzy Color Semantics

    Page(s): 39 - 44
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (875 KB) |  | HTML iconHTML  

    In this paper, two different approaches to color binning and subsequent JNS (Just Not the Same) color histogram computation are discussed. The first approach is based on a neural network color classifier trained using error back propagation training algorithm. The second approach is based on heuristically designed fuzzy classifier using fuzzy if-then rules for classifying color pixels into one of the eleven JNS colors. Color signatures for images in the database are obtained using both the methods. Further a fuzzy set theoretic approach is proposed to describe and extract the fuzzy color semantics that attempt to reduce the semantic gap between the low-level visual features and the high-level semantic features. Five linguistic variables are used to represent the image color semantics providing a flexible query scheme that is able to effectively represent vagueness in human color perception. The calibration images are inserted in the database for verifying correctness of the two approaches. The comparison of the image retrieval results obtained using fuzzy and fuzzy-neural approaches are presented at the end. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Motion-Based Blind Super-Resolution Approach to Overcome Spatial Limitation of Camera Sensor Network

    Page(s): 45 - 49
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (292 KB) |  | HTML iconHTML  

    Camera sensor network (CSN) is widely used in many new applications. But low quality images or videos obtained by cheap surveillance cameras are a bottleneck for the development of CSN. A new motion-based blind super-resolution algorithm is proposed to improve the resolution of acquired image. In this method, improved projection transformation is adopted to estimate relative motion of the low-resolution (LR) images and do motion compensation. Then a universal projection function is created and iterates to reconstruct high-resolution (HR) images from these shifted LR images with blur and noise under the condition that any parameters of the degradation model are unknown. The experiment shows that whether evaluate from subjective or objective way the method is robust and effective. The drawback is that noise are still exists and slightly enlarged in somewhere. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Novel Region Based Multifocus Image Fusion Method

    Page(s): 50 - 54
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (874 KB) |  | HTML iconHTML  

    Image fusion is a process of combining multiple input images of the same scene into a single fused image, which preserves relevant information and also retains the important features from each of the original images and makes it more suitable for human and machine perception. In this paper, a novel region based image fusion method is proposed. In literature shows that region based image fusion algorithm performs better than pixel based fusion method. Proposed algorithm is applied on large number of registered images and results are compared using standard reference and no reference based fusion parameters. The proposed method is also compared with different methods reported in the recent literature. The simulation results show that our method performs better than other methods. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Quantitative Analysis of Brain Tissues from Magnetic Resonance Images

    Page(s): 57 - 61
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (319 KB) |  | HTML iconHTML  

    This paper deals with an automatic and relatively efficient method for estimating intracranial volume from MR brain images. The proposed method consists of mainly three steps, namely skull removal, image segmentation and volume calculation. The present method uses morphological operations followed by 3D connected component labeling and image subtraction for extracting the brain mask from the original brain slices. The skull stripped images are then segmented into the three tissues namely gray matter, white matter and cerebrospinal fluid using an efficient clustering technique namely, weighted k-means clustering algorithm followed by expectation maximization algorithm. Finally the volume of the segmented tissues is calculated using Cavalier's estimator of morphometric volume method and some sample results are presented. The proposed method gives reliable results for making quantitative analysis and diagnosis of tissues from magnetic resonance brain image slices. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Research on the Video Segmentation Method with Integrated Multi-features Based on GMM

    Page(s): 62 - 66
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (358 KB) |  | HTML iconHTML  

    Video segmentation is a hot issue in the image research field. In the current video segmentation method, the pixel color feature in a frame is only considered. The pertinent problem between adjacent pixels is not taken into account. This paper proposes a video segmentation method based on GMM (Gaussian Mixture Model) modeling, meanwhile a method integrating the neighborhood characteristic of a pixel, such as pixel color and brightness characteristic is considered. The neighbor characteristic of a pixel can be a good solution for the bad segmentation result because of the tiny change in the background. The characteristic of brightness and chromaticity can solve the problem arising from the light and shadow change. In this method, the Gaussian mixture models for each pixel are built firstly. Then the relevant parameters are trained and identified. Combining the neighbor characteristic of pixel, brightness and chromaticity, the video can be segmented. Experiment results show that this method compared with other methods improves the video segmentation results. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A Blind Image Adaptive Watermarking Scheme for Audio Using Wavelet Transform

    Page(s): 67 - 71
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1100 KB) |  | HTML iconHTML  

    In this paper, we present a novel algorithm for robust audio watermarking in image using wavelet transform based on an image texture. The algorithm is based on decomposition of images using Daubechies wavelet basis. The technique proposed in this paper resolves the problem of severe distortion caused by watermarking audio in an image by developing a scaling function that achieves maximum robustness and transparency prior to its embedding. The property of texture is used as a criterion to identify the target area for embedding the watermark. The security of the algorithm is enhanced by performing a random permutation of the watermark. The random arrangement of the indices serves as a secret key. This technique is a blind watermarking scheme and the extraction procedure can be performed without the original host image.The experimental results demonstrate that watermark is imperceptible and prevents audio file from audible distortion. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Polar Run-Length Features in Segmentation of Retinal Blood Vessels

    Page(s): 72 - 75
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (264 KB) |  | HTML iconHTML  

    Manual segmentation of retinal blood vessels in optic fundus images is a tiresome task. Several methods have previously been proposed for the automatic segmentation of retinal blood vessels. In this paper we propose a classifier-based method. First the images are preprocessed so that the within class variability of the vessel and background classes are minimized. Next, the image is scanned with a window of a certain size. Polar run-length matrices are simply created by transforming the windows into polar coordinates and then constructing conventional run length matrices. Two features are then extracted for each gray level value in the polar run length matrix. The feature vectors are then classified using a multilayer perceptron artificial neural network. The performance of the proposed method is compared with that of the human observers and with those methods previously reported in literature. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Feature-Based Steganalysis for JPEG Images

    Page(s): 76 - 80
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (295 KB) |  | HTML iconHTML  

    The goal of blind steganalysis is to detect the presence of hidden data and to eventually extract them from the stego images generated by various data hiding schemes. In this paper, we construct a new blind classifier capable of detecting several steganographies for JPEG images. Thirteen statistics are collected in the DCT domain and spatial domain. By using the characteristic function and the center of mass (COM) for each statistic, we calculate an 82-dimensional feature vector for each image. Support vector function (SVM) is utilized to construct the blind classifier. Experimental results show that the proposed steganalytic method provides significant performance on various types of steganographies, such as Model-based steganography MB1[17]&MB2[18], non-blind spread spectrum data hiding method Cox[16], and five popular data hiding schemes-F5[11], JSteg[12], Jphide&Seek[13], Outguess[14] and Steghide[15]. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Connected Numeral Strings Segmentation Based on the Combination of Characteristic Position and Contour Detecting

    Page(s): 81 - 84
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (271 KB) |  | HTML iconHTML  

    Segmentation of Connected numeral strings plays an important role in number recognition systems. A new method to segment two connected numeral strings is proposed in the article, we define characteristic position first, using it the graph-representation of the image is derived, based on this and using contour detecting we attain the candidate segmentation position pairs used for determining the segmentation path, then a criterion is applied to choose the best segmentation from the candidates. Finally, an experiment is conducted. The results reveal that the method proposed has high speed and good result. View full abstract»

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