By Topic

Image Processing, IET

Issue 6 • Date August 2012

Filter Results

Displaying Results 1 - 21 of 21
  • Accuracy and numerical stability of high-order polar harmonic transforms

    Publication Year: 2012 , Page(s): 617 - 626
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1032 KB)  

    Invariant moments and transforms are a special kind of statistical measure designed to remain constant after some transformations, such as object rotation, scaling, translation or image illumination changes, in order to improve the reliability of a pattern recognition system. Polar harmonic transforms (PHTs) are one of such image descriptors based on a set of orthogonal projection bases, which are better rated in comparison with other moments and transforms because they possess less redundant information. PHTs suffer from geometric error and numerical integration error. Geometric error is caused when a square image is mapped into a unit circular disk, which cannot exactly match the square domain. Numerical integration error is caused when the integration is approximated by zeroth-order summation. We propose a computational framework based on numerical integration approach that reduces the geometric error and the numerical integration error simultaneously. The effect of the improved accuracy in the PHTs is analysed for rotation invariance, scale invariance and image noise. Various numerical experiments demonstrate that the effect of these errors is much more prominent in small images and, therefore the proposed method is suitable for applications involving small images, such as in optical character recognition and template-matching applications. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Image zooming using directional cubic convolution interpolation

    Publication Year: 2012 , Page(s): 627 - 634
    Cited by:  Papers (7)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (862 KB)  

    Image-zooming is a technique of producing a high-resolution image from its low-resolution counterpart. It is also called image interpolation because it is usually implemented by interpolation. Keys' cubic convolution (CC) interpolation method has become a standard in the image interpolation field, but CC interpolates indiscriminately the missing pixels in the horizontal or vertical direction and typically incurs blurring, blocking, ringing or other artefacts. In this study, the authors propose a novel edge-directed CC interpolation scheme which can adapt to the varying edge structures of images. The authors also give an estimation method of the strong edge for a missing pixel location, which guides the interpolation for the missing pixel. The authors' method can preserve the sharp edges and details of images with notable suppression of the artefacts that usually occur with CC interpolation. The experiment results demonstrate that the authors'method outperforms significantly CC interpolation in terms of both subjective and objective measures. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An orthogonal polynomials transform-based variable block size adaptive vector quantisation for colour image coding

    Publication Year: 2012 , Page(s): 635 - 646
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (485 KB)  

    In this paper, a new adaptive vector quantisation with variable block size based on orthogonal polynomials is proposed for colour image coding. The input colour image is partitioned into fixed larger block sizes and the homogeneity/complexity of the local region is tested with the mean-normalised variance classifier. The homogeneous regions are encoded with fixed larger blocks and the complex regions are further divided into smaller block sizes for encoding as it gives the wealth of low-level information such as the high detail and high number of contrast edges. The integer-based orthogonal polynomials transform is then applied on the variable block sizes utilising the inter-correlation property of individual colour planes and interactions among the colour planes so as to reduce the vector dimension and hence the less computational cost. The proposed technique generates two dynamic codebooks and two backup codebooks adaptively and use rate-distortion cost criteria to encode both the homogeneous and complex regions separately. Since the proposed method encodes all the three colour components in single pass, the colour image encoding time is slightly higher than that of grey scale image coding but in contrast to three times in the existing colour image coding techniques. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Efficient inverse transform architectures for multi-standard video coding applications

    Publication Year: 2012 , Page(s): 647 - 660
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1042 KB)  

    Hardware designs that can support multiple standards are required for versatile media players. The study proposes a unified inverse transform architecture that can be efficiently used in Moving Picture Expert Group and ITU International Telecommunication Standardisation Sector (ITU-T) H.264/advanced video coding (AVC), Microsoft video codec 1 (VC-1) and Chinese Audio Video Coding Standard (AVS) decoders. For H.264/AVC 8-, 4- and 2-point inverse transforms, the computational complexity in the proposed architecture is similar to that defined in the H.264/AVC standard. By using the symmetry of the transform matrices, the matrix product operations of the inverse transforms in VC-1 and AVS are efficiently decomposed to use only shifters, adders and subtractors. All the computations are verified and designed using a hardware unit to achieve a low-cost hardware kernel. The proposed multiple-transform architecture contains fast 1-D transforms and rounding operations for the computation of H.264/AVC, VC-1 and AVS 8- and 4-point inverse transforms. Simulation results show that the total number of gates for the proposed architecture is 8983, which is much lower than that required for architectures without hardware sharing. Compared with individual designs, the proposed shared architecture reduces the number of logic gates by a factor of two with a penalty of 20% in data throughput. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Improved kernel-based IRIS recognition system in the framework of support vector machine and hidden markov model

    Publication Year: 2012 , Page(s): 661 - 667
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (567 KB)  

    IRIS biometric is one of the most efficient and trusted biometric methods for authenticating users owing to invariance with age or with physical activities. IRIS recognition techniques are broadly categorised in three groups: phase, texture and kernel-based methods, which out of kernel-based methods are proven to be the best suited for IRIS recognition problem. In this work a multiclass kernel Fisher analysis and its consequent feature set for IRIS recognition is proposed. The authors use support vector machine (SVM) classifier to group the large database into smaller groups where each group is linearly separable from the other. Once an image is grouped as one of the groups by SVM, it is classified to be recognised by hidden Markov model (HMM) classifier which compares the features of the given image only with the other images of the same group. Results show 93.2% overall accuracy for the system if we consider seven features and improved to 99.6% when 1200 features are used. In order to meet this efficiency an average convergence time needed by the algorithm is found to be lesser than existing SVM-based technique. Results also show fast convergence time for optimisation process in comparison to with other conventional kernel and SVM-based techniques. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Image quality augmented intramodal palmprint authentication

    Publication Year: 2012 , Page(s): 668 - 676
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (474 KB)  

    This study serves dual objectives. First, as an efficient, in terms of speed and accuracy, wavelet-based intramodal palmprint authentication approach. Second, quantification of illumination, quality of palmprints and its incorporation in the score fusion for the classification performance enhancement. The later objective is realised using localised contrast measurement and quality-augmented fusion based on illumination sensitiveness of the features. The former objective is realised through intramodal feature extraction and fusion exploiting the multi-scale analysis of palmprint using wavelet transform. The intramodal features (energy, principal lines, dominant wrinkles and high-scale spatial patterns) are extracted in the wavelet domain thus significantly minimising the computational disadvantage of intramodal approach. Significant reduction in the equal error rate (EER) is observed upon match score fusion. Experimental results on PolyU-Online-Palmprint-Database (PolyU) of 386 classes show a relative improvement index of 71.75% with an overall EER of 0.14%; better than the state-of-the-art intramodal and wavelet-based approaches. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • New adaptive steganographic method using least significant- bit substitution and pixel-value differencing

    Publication Year: 2012 , Page(s): 677 - 686
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (604 KB)  

    This study presents a new adaptive data-hiding method based on least-significant-bit (LSB) substitution and pixel-value differencing (PVD) for grey-scale images. The proposed method partition the cover image into some non-overlapping blocks having three consecutive pixels and select the second pixel of each block as the central pixel (called base-pixel). Then, k-bits of secret data are embedded in the base pixel by using LSB substitution and optimal pixel adjustment process (OPAP). The difference between the base-pixel value and other pixel values in the block are utilised to determine how many secret bits can be embedded in the two pixels. Also, the method divides all possible differences into lower level and higher level with a number of ranges. Then, it obtains the number of the secret bits that will be embedded into each block depending on the range which the difference values belong to. The experimental results show that the proposed method can embed a large amount of secret data while maintaining a high visual quality of the stego-images. The peak signal-to-noise ratio (PSNR) values and the embedding capacity of our method are higher than those of three other data-hiding methods which are investigated in this study. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Automated extraction of three-dimensional cereal plant structures from two-dimensional orthographic images

    Publication Year: 2012 , Page(s): 687 - 696
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (470 KB)  

    The authors present a novel approach to automatically extract three-dimensional (3D) structures of cereal plants from 2D orthographic images. The idea of this approach is to use 2D skeletons to represent elongated shapes of plants in 2D images and then to generate 3D plant structures from 2D skeletons. However, existing skeletonisation algorithms generate artefacts because of boundary extremities and noise. Moreover, cereal plant leaves appear as broken segments in images because of leaf twists. In this approach, oriented Gaussian filters are used to obtain ridges from 2D images via guided skeletonisation, so that one can obtain smooth skeleton representation of 2D elongated shapes without redundant branches. For broken leaf segments, a cost function is proposed to identify whether the two segments belong to the same leaf and thus should be connected. For a given leaf tip of a cereal plant, the z-coordinators of two 2D side-view images are same. Based on this, it is easy to identify the same leaf tip in two 2D side-views. Thus, the authors are able to track skeletons of leaves from their tips to the plant roots. Experimental results show that the proposed approach is able to handle various issues such as broken segments and overlapping among plant parts, and is also able to automatically extract 3D structures of cereal plants. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Computer-aided diagnosis of lung cancer based on analysis of the significant slice of chest computed tomography image

    Publication Year: 2012 , Page(s): 697 - 705
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (286 KB)  

    In this study, a computer-aided diagnosis system capable of selecting a significant slice for the analysis of each nodule from a set of slices of a computed tomography (CT) scan in digital imaging and communications in medicine (DICOM) format has been developed for the diagnosis of lung cancer. First, the CT image was preprocessed by segmenting the lung parenchyma from each slice using a greedy snake algorithm. The regions of interest (ROIs) were then extracted from the lung parenchyma using a region-growing algorithm. The extracted ROIs were labelled as cancerous or non-cancerous nodules with the aid of a human expert and then the shape and texture features were extracted from each ROI. The extracted features and the label of the corresponding ROI were used to train a radial basis function neural network (RBFNN). When a CT image is given to the system for diagnosis, it is first preprocessed to extract the ROIs from each slice. Only those ROIs that are greater than nine pixels and that exist in at least three slices are considered as nodules. For each nodule, the slice with the largest area is chosen as the significant slice and this slice is taken up by the feature extraction subsystem for further analysis of the nodule. The features are extracted and fed to the RBFNN, which classifies the nodule as cancerous or non-cancerous. From the experimental results, the system was found to achieve an accuracy of 94.44%. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Image restoration and decomposition using nonconvex non-smooth regularisation and negative Hilbert-Sobolev norm

    Publication Year: 2012 , Page(s): 706 - 716
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (771 KB)  

    A new model for image restoration and decomposition has been presented here. The proposed model applies the non-convex non-smooth regularisation and the Hilbert-Sobolev spaces of negative degree of differentiability to capture oscillatory patterns. The existence of a pseudosolution to the proposed model is proved. Moreover, two numerical algorithms for solving the minimisation problem are provided by applying the variable splitting and the penalty techniques. Finally, extended experiments on denoising, deblurring and decompositions of both real and synthetic images demonstrate the effectiveness and efficiency of the proposed numerical schemes. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Image thresholding framework based on two-dimensional digital fractional integration and Legendre moments'

    Publication Year: 2012 , Page(s): 717 - 727
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1000 KB)  

    In this study, the authors present a new image segmentation algorithm based on two-dimensional digital fractional integration (2D-DFI) that was inspired from the properties of the fractional integration function. Although obtaining a good segmentation result corresponds to finding the optimal 2D-DFI order, the authors propose a new alternative based on Legendre moments. This framework, called two dimensional digital fractional integration and Legendre moments' (2D-DFILM), allows one to include contextual information such as the global object shape and exploits the properties of the 2D fractional integration. The efficiency of 2D-DFILM is shown by the comparison to other six competing methods recently published and it was tested on real-world problem. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance of reversible digital image watermarking under error-prone data communication: a simulation-based study

    Publication Year: 2012 , Page(s): 728 - 737
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1574 KB)  

    Reversible digital watermarking techniques enable the recovery of the original `cover image` from a watermarked image in a distortion-free way. Reversible watermarking techniques find application in military and medical imagery, where integrity of the cover image is of utmost importance. However, in practice, many military data transmissions occur over communication channels whose noise levels are so high that the receiving system is unable to correct all errors in the received data. In such a case, the authors are bound to get non-zero distortion in the recovered cover image as well as the extracted watermark, in spite of using reversible watermarking techniques. In this study, they investigate the effect of high data error rates on different state-of-the-art reversible watermarking algorithms. Their simulation results help to choose the most suitable reversible watermarking scheme, depending on whether the distortion of the retrieved cover image or the distortion of the retrieved watermark is the primary concern. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Performance evaluation methodology for document image dewarping techniques

    Publication Year: 2012 , Page(s): 738 - 745
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (910 KB)  

    The performance evaluation of dewarping techniques is currently addressed by concentrating in visual pleasing impressions or by using optical character recognition (OCR) as a means for indirect evaluation. In this study, the authors present a performance evaluation methodology that calculates a comprehensive evaluation measure which reflects the entire performance of a dewarping technique in a concise quantitative manner. The proposed evaluation measure takes into account the deviation of the dewarped text lines from a horizontal straight reference which is considered to be the optimal result. This measure is expressed by the integral over the dewarped text line curves. To reduce the manual effort for identifying the text lines in the dewarped image, the authors propose a point-to-point matching procedure that finds the correspondence between the manually marked warped document image and the dewarping counterpart. This enables an evaluation for unlimited number of methodologies addressing a marking procedure which is applied only once. The validity of the proposed performance evaluation methodology is demonstrated by a concise experimental work that comprises four state-of-the-art dewarping techniques as well as the involvement of different users in the interactive part of the procedure. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Approximately even partition algorithm for coding the Hilbert curve of arbitrary-sized image

    Publication Year: 2012 , Page(s): 746 - 755
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1050 KB)  

    The Hilbert curve is one of space filling curves and it requires that the region is of size 2k × 2k, where kN. This study relaxes this constraint and generates a pseudo-Hilbert curve of arbitrary dimension. The intuitive method such as Chung et al.'s algorithm is to use Hilbert curves in the decomposed areas directly and then have them connected. However, they must generate a sequence of the scanned quadrants additionally before encoding and decoding the Hilbert order of one pixel. In this study, by using the approximately even partition approach, the authors propose a new Hilbert curve, the Hilbert* curve, which permits any square regions. Experimental results show that the clustering property of the Hilbert* curve is similar to that of the standard Hilbert curve. Next, the authors also propose encoding/decoding algorithms for the Hilbert* curves. Since the authors do not need to additionally generate and scan the sequence of quadrants, the proposed algorithm outperforms Chung et al.'s algorithms for the square region. Then, the authors apply the Hilbert* curves in Chung et al.'s algorithms for the Hilbert curve of arbitrary dimension and experimental results show that the proposed encoding/decoding algorithms out perform the Chung et al.'s approach. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Wavelet-based image denoising using three scales of dependency

    Publication Year: 2012 , Page(s): 756 - 760
    Cited by:  Papers (3)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (280 KB)  

    The denoising of a natural image corrupted by the Gaussian white noise is a classical problem in image processing. A new image denoising method is proposed by using three scales of dual-tree complex wavelet coefficients. The dual-tree complex wavelet transform is well known for its approximate shift invariance and better directional selectivity, which are very important in image denoising. Experiments show that the proposed method is very competitive when compared with other existing denosing methods in the literature. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Fast and robust skew estimation in document images through bilinear filtering model

    Publication Year: 2012 , Page(s): 761 - 769
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (749 KB)  

    Skew estimation of scanned document is important for document analysis and recognition. Owing to the complexity inherent in the document, some methods only process the documents with a small skew angle or specified content, layout and have high computational cost. A fast and robust skew estimation method is proposed based on a bilinear filtering model, which is used to detect edges existing in the document. Some foreground areas in the document have been extracted without considering document layouts or contents. The proposed approach enhances the structure of the document and reduces the effects of the noises in the document. It combines filtering operators into a single approach, so noise filtering which is an unavoidable pre-processing in the previous skew detection methods has been overcome. A dominant angle has been estimated based on the detected edges. According to the estimated dominant angle, a skew angle can be determined efficiently without confining the search space or making assumptions including skew angle range and layout of the document in advance. The proposal greatly reduces the computational time and works in an unsupervised style. Comparative tests with the state-of-the-art skew estimation methods indicate the superior performance of the developed approach. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Human body segmentation based on independent component analysis with reference at two-scale superpixel

    Publication Year: 2012 , Page(s): 770 - 777
    Cited by:  Papers (1)
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (676 KB)  

    In this study, a novel method to segment human body in static image is proposed based on independent component analysis with reference (ICA-R) at two-scale superpixel. In this work, the task is mainly decomposed into torso and lower body recovery. With the detected face, we obtain the reference signal of torso in the coarse torso region estimated by an augmented deformable torso model on the basis of the first-scale superpixel. The hip region is estimated based on the segmented torso for the lower body reference at the second-scale superpixel. Experiments on our dataset show that the proposed approach is robust and can accurately segment human body in images with a variety of poses, backgrounds and clothing. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Workspace for image clustering based on empirical mode decomposition

    Publication Year: 2012 , Page(s): 778 - 785
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (696 KB)  

    This study presents a new approach for image clustering, which is based on a novel workspace derived from the empirical mode decomposition (EMD). The proposed algorithm exploits the EMD, which can decompose any non-linear and non-stationary data into a number of intrinsic mode functions (IMFs). The intermediate IMFs of the image histogram have very good characteristics and provide a robust workspace that is utilised in order to detect the clusters of an image in a fast way. The proposed method was applied to several images and the obtained results show good image clustering robustness and low computational time, overcoming the disadvantages of the existing image clustering algorithms. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Blending-weight diffusion for image colourisation

    Publication Year: 2012 , Page(s): 786 - 794
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (829 KB)  

    An image colourisation method based on diffusion of colour-blending weights is presented. The authors propose to use reciprocal function for smoothness measurement to determine the priority order of colour propagation in image colourisation. The reciprocal measurement can improve the recent exponential-measurement-based propagation, which has a trend towards being greedy for smooth pixels, and is likely to cause wrong boundaries nearby the object juncture joining the grey levels without significant difference. In addition, the proposed method uses the blending-weight diffusion instead of the direct propagation of chrominance values. By doing so, it can respond quickly to the user demand for stroke editing. Experimental results show that compared with the exponential-measurement-based propagation, the proposed method filled colours more precisely in the positions representing real object surfaces, and saved 85.92% of computation time for adding a stroke and 98.60% for deleting a stroke, to an illustrative image. Compared with two conventional techniques which also implement-blending weights, this method can provide the blending weights for acceptable colourisation by using the economical strokes which positions are less constrained and which sizes are reduced. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Robust kernel-based learning for image-related problems

    Publication Year: 2012 , Page(s): 795 - 803
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (839 KB)  

    Robustness is one of the most critical issues in the appearance-based learning techniques. This study develops a novel robust kernel for kernel machines, and consequently improves their robustness in resisting noise for solving the image-related learning problems. By incorporating a robust ρ-function to reduce the influence of outlier components, this kernel gives more reasonable kernel values when images are seriously corrupted. The authors incorporate the proposed kernel into different kernel-based approaches, such as support vector machine (SVM) and kernel Fisher discriminant (KFD) analysis, to validate its performance on various visual learning problems of face recognition and data visualisation. Experimental results indicate that the proposed kernel can provide the superior robustness to the classical approaches. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Partitioning of feature space by iterative classification for degraded document image binarisation

    Publication Year: 2012 , Page(s): 804 - 812
    Save to Project icon | Click to expandQuick Abstract | PDF file iconPDF (1116 KB)  

    Proper partitioning of feature space into text and background regions is very important in document image binarisation. This study presents an iterative classification algorithm that efficiently partitions a two-dimensional feature space into text and background regions. It uses the result of Niblack's binarisation algorithm as training data and employs its characteristics to define classification rules. In each iteration, it labels only some points of the feature space, which can be classified reliably and leaves the classification of other points to the next iterations. The classification result of a point in current iteration affects the classification of its neighbours in the next iterations and makes them more probable to be classified correctly. After a few iterations, it partitions the feature space into two regions associated with the text and background pixels. After partitioning, two global thresholding methods were used as an extra text class refinement to make the proposed algorithm robust against bleeding-through and shadow-through degradations. Finally, each pixel is labelled as either text or background according to its corresponding region in the feature space. The authors' binarisation algorithm demonstrated superior performance against six well-known algorithms on three datasets. It is appropriate for various types of degraded images. View full abstract»

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

Aims & Scope

The range of topics covered by IET Image Processing includes areas related to the generation, processing and communication of visual information.

Full Aims & Scope

Meet Our Editors

Publisher
IET Research Journals
iet_ipr@theiet.org