A simple and effective method is presented for discrete-cosine-transform (DCT)–domain downsizing. Various methods employed in DCT-domain downsizing simply reuse the frequency component of DCT, which shows a severe aliasing effect. The proposed approach extends the downsizing method for alleviating or reducing the aliasing effect with a windowing operation, which adjusts the magnitude of the DCT coefficient. Visual inspection showed satisfactory results, with no complexity overhead and performance degradation regarding the peak-signal-to noise ratio (PSNR) after upsampling.

When estimating partial volume effects in the presence of noise, using neighboring information improves the estimation. The optimal linear transformation (OLT) is an unbiased minimum variance estimator. However, it does not use neighboring information and thus is sensitive to noise. We employ polynomial and B-spline continuous representations of the data to mathematically incorporate the neighboring information into the OLT. To evaluate the method, we use synthetic and actual images generated by simulation and acquired from phantoms and the human brain. Standard deviations of new estimators are up to 60% less than that of the OLT when the signal-to-noise ratio (SNR) is 25. As the SNR decreases, the proposed method demonstrates more improvements. Overall, B-spline estimators provide larger estimations of the standard deviation compared to polynomials. However, B-spline estimators outperform polynomials, providing an arbitrary degree of continuity. B-spline estimators are up to 10 times faster than polynomials and about 10 times slower than the OLT.

The secure verification is important for watermarking protocols. A malicious arbitrator is able to remove an original watermark from an unauthorized copy of the digital content as a result of a security breach in the phase of arbitration and resell multiple copies of it with impunity. We propose a novel buyer-seller watermarking protocol of secure verification. In this scheme, a seller permutes an original watermark provided by a trusted Watermarking Certification Authority (WCA) and embeds it into digital content in an encrypted domain. In case an unauthorized copy is found, the seller can recover the original watermark from the watermark extracted from the copy and sends it to an arbitrator. Without the knowledge of permutations applied by the seller, the arbitrator is unable to remove the permuted watermark from the digital content. Hence, verification is secured. As an additional advantage of the proposed protocol, arbitration can be conducted without the need for the cooperation of the WCA or the buyer.

Confocal microscopy provides a powerful tool for biologists to investigate gene expression in a 3D manner. However, due to the inherent properties of confocal images, it is difficult to accurately segregate foreground signals from the background using direct thresholding. Therefore, there is a need for a segmentation algorithm that can be used with fluorescent confocal images of gene expression. We present an automatic segmentation algorithm for thresholding confocal images of gene expression in biological samples. The algorithm, called density-based segmentation (DBS), is modified from a noise-tolerant data clustering algorithm (DENCLUE). We demonstrate the utility of this algorithm in different synthetic images as well as in confocal images of zebrafish embryos, with comparison to Otsu's algorithm, which employs direct thresholding. The results of segmentation in synthetic images show that the DBS algorithm is noise-tolerant and is able to distinguish two objects located close to each other. In addition, the results of segmentation in confocal images show that the DBS algorithm can threshold objects while preserving morphological details of internal structures. Therefore, the proposed DBS algorithm is a better segmentation technique than direct thresholding in the segmentation of fluorescent confocal images.

We propose a pairwise Rayleigh quotient (PRQ) classifier and apply it to discriminate between malignant tumors and benign masses in mammograms. The PRQ classifier employs a Rayleigh quotient based on a set of pairwise constraints, which leads to a generalized eigenvalue problem with low complexity of implementation. Kernel functions are used to incorporate nonlinearity. Studies were conducted with features of 57 breast masses, of which 20 are related to malignant tumors and 37 to benign masses. The linear PRQ classifier provided results comparable to those obtained with Fisher's linear discriminant analysis (FLDA), support vector machines (SVMs), and convex pairwise SVMs (CPSVMs). The linear PRQ classification performance of the comparatively weak feature sets with edge sharpness and texture features was significantly improved by about 5%, as compared to those obtained by FLDA, SVM, and CPSVM. The nonlinear PRQ classifier with the triangle kernel provided the perfect performance of 1.0 in terms of the area under the receiver operating characteristics curve, for nearly all feature combinations, but with good robustness limited to the kernel parameter in a certain range. We propose a measure of robustness to evaluate the PRQ classifier.

We describe a method for automatically detecting streaks in printed images using adaptive window-based image projections and mutual information. The proposed approach accepts a scanned image enclosing the defect and computes the projections across the entire image at different window sizes. The resulting traces collected from the projections are analyzed with a peak detection algorithm and subsequently correlated using normalized mutual information to pinpoint the location and width of streak(s). Finally, for a given peak, the window size is changed adaptively to identify and locate the intensity and length of the corresponding streak(s) while maximizing the signal-to-noise ratio. Results on synthetic and real-life images are provided to demonstrate the effectiveness of our proposed technique.

We propose novel matrix-form representations of the conventional forward three-dimensional (3-D) discrete cosine transform (DCT) and the inverse 3-D DCT. The pair of the computationally efficient 3-D DCTs causes the formulation to be more concise with a systematic structure, which has a high potential for the development of other applications via the 3-D DCT pair. Applications for enlargement of image/video implemented by the DCT pair are also presented. These new algorithms effectively eliminate ripple and blocky effects and maintain the original characteristic. Moreover, due to the essentially even-function property of the DCT, the proposed enlargement of an image demonstrates the desired symmetric property. As a result, they can also find numerous applications in image processing and video coding. Simulation results show that proposed mechanisms enable good performance for scaling algorithms of sequences, images, and videos, as expected.

Based on the novel matrix representations of the forward three-dimensional (3-D) discrete cosine transform (DCT) and the inverse 3-D DCT, applications for reduction of sequence/image/video are proposed. The original sequence/image/video is divided into subunits, and then the reduction is implemented in the frequency-domain by DCT. Two efficient algorithms are presented in accordance with two kinds of reducing multiples, 1/s and o₁/o₂, respectively. The proposed algorithms significantly eliminate ripple and blocky effects. As a result, they can provide numerous applications in data compression as well. Simulation results show that proposed mechanisms enable good performances for reductions of sequences, images, and videos as expected.

Digital video watermarking has attracted a great deal of research interest in the past few years in applications such as digital fingerprinting and owner identification. The H.264/AVC (advanced video coding) standard is the latest and most advanced video coding standard, but to this date, very few watermarking schemes have been designed for it. This is mainly due to its complexity and compression efficiency, which present a major challenge for any video watermarking approach. We developed a new, quantization-based video watermarking scheme that is designed to work with H.264/AVC. We propose a locally adaptive, rate-distortion optimized watermark that is inserted in the transform coefficients of the macro blocks. We use a unique perceptual mask in order to control the levels of spatial and temporal distortion. Our scheme is designed with a built-in bit allocation mechanism that ensures optimum distribution of watermark bits between different macro blocks. Our watermark offers constant robustness to H.264 video compression at different bit rates, without significantly affecting the overall bit rate and quality of the video stream. Experimental results show that our scheme outperforms existing watermarking methods under compression, transcoding, filtering, scaling, rotation, and collusion attacks.

We propose a novel scanner characterization approach for applications requiring color measurement of hardcopy output in calibration, characterization, and diagnostics applications. The method is advantageous for common practical color printing systems that use more than the minimum of three colorants necessary for subtractive color reproduction; printing with cyan (C), magenta (M), yellow (Y), and black (K) is the most prevalent example we use in our description. The proposed method exploits the fact that for the scenarios in consideration, in addition to the scanner RGB values for a scanned patch, the CMYK control values used to print the patch are also available and can be exploited in characterization. An indexed family of 3D scanner characterizations is created, each characterization providing a mapping from scanner RGB to CIELAB for a fixed value of K, the latter constituting the index for the characterization. Combined together, the family of 3D characterizations provides a single 4D characterization that maps scanner RGB obtained from scanning a patch and the K control value used for printing the patch to a colorimetric CIELAB measurement for the patch. A significant improvement in the robustness of the method to variations in printing is obtained by modifying the K index to utilize the scanned output for a black-only patch printed with the corresponding K value instead of directly utilizing the control K value used at the printer. Results show that the proposed 4D scanner characterization technique can significantly outperform standard 3D approaches in the target applications.

The reconstruction of an image from its projections through filtered backprojection is not new and has gained popularity since its application on computerized tomography (CT) for medical imaging. However, the number of projections to reconstruct a single image can be in the thousands. Thus, significant computational power is required. Today's digital signal processors (DSPs) offer an alternative for implementations of a backprojector to process large amounts of data. We demonstrate an implementation in a fixed-point DSP. This implementation is optimized for cache memory to reduce CPU stalls, and we achieved up to 99.6% of effective CPU cycles. With a DSP running at 1 GHz, an image of 1200×1200 pixels can be reconstructed from 180 projections in 1 s.

We propose a fast and simple shot-cut detection algorithm, which directly operates in a compressed domain and is suitable for real-time implementation. The proposed algorithm exploits the existing MPEG techniques by examining the prediction status for each macro-block inside the B and P frames. As a result, locating both abrupt and dissolved shot cuts is operated by a sequence of comparison tests; thus, no feature extraction or histogram differentiation is needed. Although the description of the algorithm is primarily based on MPEG-1 and MPEG-2 streams, the scheme can be readily extended to other video compression standards such as MPEG-4 and H.264 by following the principle on monitoring: (1) balance between forward prediction and backward prediction; and (2) boundaries among the P, B, and I frames. Extensive experiments illustrate that the proposed algorithm outperforms similar existing algorithms, providing a useful technique for fast and online video content processing.

We deal with video shot-cut detection in digital videos using the singular-value decomposition (SVD). SVD is performed on a matrix whose columns are the 3D frame color histograms. We have used SVD for its capabilities to derive a refined low-dimensional feature space from the high-dimensional raw feature space, where similar video patterns are placed together and can be easily clustered. After SVD is performed, a two-phase process is employed to detect the shots. In the first phase, a dynamic clustering method is used to create the frame clusters. In the second phase, every two consecutive clusters, obtained by the clustering procedure, are tested for a possible merging in order to reduce false shot-cut detections. In the merging phase, statistical hypothesis testing is used. The detection technique was applied to several TRECVID video test sets that exhibit different types of shots and contain significant object and camera motion inside the shots. We demonstrate that the method detects cuts and gradual transitions, such as dissolves and fades, with high accuracy.

We propose novel techniques for microarray image analysis. In particular, we describe an overall pipeline able to solve the most common problems of microarray image analysis. We propose the microarray image rotation algorithm (MIRA) and the statistical gridding pipeline (SGRIP) as two advanced modules devoted to restoring the original microarray grid orientation and to detecting, the correct geometrical information about each spot of input microarray, respectively. Both solutions work by making use of statistical observations, obtaining adaptive and reliable information about each spot property. They improve the performance of the microarray image segmentation pipeline (MISP) we recently developed. MIRA, MISP, and SGRIP modules have been developed as plug-ins for an advanced framework for microarray image analysis. A new quality measure able to effectively evaluate the adaptive segmentation with respect to the fixed (e.g., nonadaptive) circle segmentation of each spot is proposed. Experiments confirm the effectiveness of the proposed techniques in terms of visual and numerical data.

Spectral separation is the process of obtaining printer control values to reproduce a given spectral reflectance. Given a multispectral image where each pixel represents a spectral reflectance, separation could be implemented by inverting a physical printer model on a pixel-by-pixel basis. Such a process would obviously need to be very fast to handle high-resolution images in a reasonable time. For a printer whose spectral response is characterized by the Yule–Nielsen spectral Neugebauer model, the linear regression iteration (LRI) method can be used to invert the model. We introduce the subspace linear regression iteration (SLRI) method, a modification of LRI shown to be significantly accelerated due to performing its calculations within the subspace determined by the Neugebauer primaries. Using this subspace approach, the number of multiplications becomes independent of the spectral sampling rate. Using a standard six color printer and a common spectral sampling rate, the number of multiplications can be decreased by about two-thirds without changing the convergence behavior.

We develop a spatially adaptive Bayesian image denoising method using a mixture of a Gaussian distribution and a point mass function at zero. In estimating hyperparameters, we present a simple and noniterative method. We use a hypothesis-testing technique in order to estimate the mixing parameter, the Bernoulli random variable. Based on the estimated mixing parameter, the variance for a clean signal is obtained by using the maximum generalized marginal likelihood (MGML) estimator. We simulate our denoising method using both orthogonal wavelet and dual-tree complex wavelet transforms and compare our algorithm to well-known denoising schemes. Experimental results show that the proposed method can generate good denoising results.

Three-dimensional surface reconstruction of the coronary arterial from intravascular ultrasound (IVUS) images plays an important role in diagnosis and treatment of the atherosclerosis. A simplified 3-D topologically adaptable snake model (3-D T-snake) is proposed; nevertheless, it can successfully address the significant problem of conventional T-snakes when dealing with the model self-intersections. A method of surface reconstruction of IVUS coronary arterial walls based on the simplified 3-D T-snake is presented. The experimental results indicate that the method is accurate and robust, and the simplified 3-D T-Snake is feasible and valid.

This PDF file contains the errata for “JEI Vol. 16 Issue 04 Paper 2803892” for JEI Vol. 16 Issue 04

This PDF file contains the editorial “Pattern Recognition and Machine Learning” for JEI Vol. 16 Issue 04