Tracking fish in their natural environment is an important aspect of marine ecosystem research. However, real-world fish tracking is challenging due to unconstrained environments and complex scenarios. The purpose of this study is to develop a sparse sample collection and representation method (SSCR) based on the compressive sensing concept for fish tracking. The SSCR consists of sample collection and sparse sample representation procedures. The sample collection procedure obtains sets of positive, negative, and predictive samples by using the proposed speed-up background modeling method (SuBM). The SuBM adopts nonparametric histogram concept for each pixel to build a background model, and efficiently accelerates the tracking speed. In addition, the sparse sample representation procedure represents each predictive sample as a sparse linear combination of all positive and negative samples. The weights of the predictive samples are computed using our proposed re-weighting and dynamically updating orthogonal matching pursuit method (RwDuOMP). The RwDuOMP, which includes three concepts (picking extra samples, re-weighting the picked samples, and dynamically updating negative samples), efficiently improves the performance of sparse signal reconstruction. The predictive sample with the maximum weight is regarded as the target object tracking result. We evaluate the SSCR method using several complicated real-world underwater sequences. Furthermore, we compare the SuBM with the Gaussian Mixture Model, and also compare the RwDuOMP method with the orthogonal matching pursuit (OMP), regularized OMP, and compressive sampling matching pursuit methods. Experimental results indicate that our proposed method achieves efficiently higher tracking results than other methods, and accelerates fish tracking.

Fish recognition and identification in an underwater environment are important research topics. In this study, several real-world underwater videos were collected to construct a fish category database for further fish recognition and identification. Recently, compressive sensing, using reconstruction algorithms to reconstruct a sparse signal, has been successfully applied to face recognition. Reconstruction algorithms can be roughly categorized into two groups: basic pursuit (BP) and matching pursuit (MP). BP-related methods adopt a convex optimization technique, while MP-related methods utilize greedy search and vector projection ideas. This study reviews concepts for these reconstruction algorithms and analyzes their performance. Moreover, an over-atoms accumulation orthogonal matching pursuit (OAOMP) method based on OMP is proposed. OAOMP includes two procedures: picking over atoms, and accumulating weighting coefficients of each subject to assign as new weights. OAOMP was compared with existing reconstruction algorithms in terms of reconstruction performance and run time. Experiments were implemented in a fish category database by using eigenfaces and fisherfaces for feature extraction. The experimental results demonstrated that BP-related methods have better recognition rates, while MP-related methods have shorter run times. Moreover, OAOMP is able to achieve better accuracy than OMP and other MP-related methods.

MOTIVATION: High-accuracy de novo assembly of the short sequencing reads from RNA-Seq technology is very challenging. We introduce a de novo assembly algorithm, EBARDenovo, which stands for Extension, Bridging And Repeat-sensing Denovo. This algorithm uses an efficient chimera-detection function to abrogate the effect of aberrant chimeric reads in RNA-Seq data. RESULTS: EBARDenovo resolves the complications of RNA-Seq assembly arising from sequencing errors, repetitive sequences and aberrant chimeric amplicons. In a series of assembly experiments, our algorithm is the most accurate among the examined programs, including de Bruijn graph assemblers, Trinity and Oases. AVAILABILITY AND IMPLEMENTATION: EBARDenovo is available at http://ebardenovo.sourceforge.net/. This software package (with patent pending) is free of charge for academic use only. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Face recognition is a popular topic in computer vision applications. Compressive sensing is a novel sampling technique for finding sparse solutions to underdetermined linear systems. Recently, a sparse representation-based classification (SRC) method based on compressive sensing is presented. It has been shown to be robust for face recognition. In this paper, we proposed a maximum probability of partial ranking method based on the framework of SRC, called SRC-MP. It computes the maximum probability from the largest λ weighting coefficients for the individuals, respectively. Experiments are implemented on Extended Yale B and ORL face databases using eigenfaces, fisherfaces, 2DPCA and 2DLDA for feature extraction. Furthermore, we compare our proposed method with classical projection-based methods such as principal component analysis (PCA), linear discriminant analysis (LDA), 2DPCA and 2DLDA. The experimental results demonstrate our proposed method is able to achieve higher recognition rate than other methods.

Experimental results of applying two similarity measurements, Euclidean distance and chord distance, to test a set of six Brodatz's textures are reported. Experiments show that in addition to feature extraction, a similarity measurement between images should be simultaneously considered. We also review some other similarity measurements.

MOTIVATION: Differential detection on symptom-related pathogens (SRP) is critical for fast identification and accurate control against epidemic diseases. Conventional polymerase chain reaction (PCR) requires a large number of unique primers to amplify selected SRP target sequences. With multiple-use primers (mu-primers), multiple targets can be amplified and detected in one PCR experiment under standard reaction condition and reduced detection complexity. However, the time complexity of designing mu-primers with the best heuristic method available is too vast. We have formulated minimum-set mu-primer design problem as a set covering problem (SCP), and used modified compact genetic algorithm (MCGA) to solve this problem optimally and efficiently. We have also proposed new strategies of primer/probe design algorithm (PDA) on combining both minimum-set (MS) mu-primers and unique (UniQ) probes. Designed primer/probe set by PDA-MS/UniQ can amplify multiple genes simultaneously upon physical presence with minimum-set mu-primer amplification (MMA) before intended differential detection with probes-array hybridization (PAH) on the selected target set of SRP. RESULTS: The proposed PDA-MS/UniQ method pursues a much smaller number of primers set compared with conventional PCR. In the simulation experiment for amplifying 12 669 target sequences, the performance of our method with 68% reduction on required mu-primers number seems to be superior to the compared heuristic approaches in both computation efficiency and reduction percentage. Our integrated PDA-MS/UniQ method is applied to the differential detection on 9 plant viruses from 4 genera with MMA and PAH of 11 mu-primers instead of 18 unique ones in conventional PCR while amplifying overall 9 target sequences. The results of wet lab experiments with integrated MMA-PAH system have successfully validated the specificity and sensitivity of the primers/probes designed with our integrated PDA-MS/UniQ method.

The security of multimedia over network transmission has recently attracted a lot of research. This paper combines schemes of cryptography with steganography for hiding secret messages. Given secret messages, for example, an English sentence, our scheme first converts the messages to an M/spl times/N binary image which is then covered by a binary random texture synthesized from a 2D Ising Markov random field using the seed, a shared secret key, between the sender and the receiver, generated by the strategy of elliptic curve cryptography (ECC). The concealed messages are then encrypted, based on the RSA scheme for transmission. An experiment shows that using an unauthorized key gets messages totally different from the original ones even when the error key is very close to the authorized one.

Fractal image compression is famous for its particular iterated decoder and the magic Collage theorem. This paper proposes an efficient decoding scheme. In fractal image compression, an image is partitioned into nonoverlapped range blocks and overlapped domain blocks. That is, there are more domains than ranges. Hence many pixels {p/sub i/} in the image do not belong to any domain blocks and these pixels need not be computed iteratively. We can compute them only once in the last iteration. Moreover, some other pixels {p/sub i/} can also be computed noniteratively if they only map to {p/sub i/}. Therefore iterative computations on {p/sub i/} and {p/sub i/} are redundant. We can eliminate the redundancy to accelerate the decoder without any loss on fidelity. In our experiment, the polished procedure can speed up on a large scale. It takes only 0.2-0.3 seconds to decode a 512 by 512 image on a Pentium II 450 PC running Windows 98.

With iterated function systems (IFS), there are two classical algorithms of generating artificial fractal pictures: the deterministic algorithm and the random iteration algorithm. The deterministic algorithm spends huge computations on checking convergence iteratively. On the other hand, the random iteration algorithm is nondeterministic and time consuming. People seldom check convergence by the algorithm. Moreover, many pixels in the picture are computed again and again by both of the algorithms. Thus a new algorithm is given. The algorithm promises convergence without iterative checks. Also, our algorithm runs faster than the existing algorithms do on a large scale.

Summary form only given. Image data compression by fractal techniques has been widely investigated. Although its high compression ratio and resolution-free decoding properties are attractive, the encoding process is computationally demanding in order to achieve an optimal compression. This article proposes a fast fractal-based encoding algorithm (ACC) by using the intensity changes of neighboring pixels to search for a suboptimal domain block for a given range block. Experimental results show that our algorithm achieves close to the optimal algorithm (OPT) for 256/spl times/256 images Jet, Lenna, Mandrill, and Peppers, with a compression ratio of 16. A comparison of the performance of algorithms OPT and ACC on a Sun Ultra 1 Sparc workstation is given.

This paper attempts to give a recipe for selecting one of the popular image compression algorithms based on: 1) wavelet, 2) JPEG/DCT, 3) vector quantisation, and 4) fractal approaches. We review and discuss the advantages and disadvantages of these algorithms for compressing gray-scale images, give an experimental comparison on four 256/spl times/256 commonly used images (Jet, Lenna, Mandrill, Peppers, and one 400/spl times/400 fingerprint image). Our experiments show that all of the four approaches perform satisfactorily when the 0.5 bits per pixel (bpp) is desired. However, for a low bit rate compression like 0.25 bpp or lower, the embedded zerotree wavelet approach and DCT-based JPEG approach are more practical.

The authors empirically compare three algorithms for segmenting simple, noisy images: simulated annealing (SA), iterated conditional modes (ICM), and maximizer of the posterior marginals (MPM). All use Markov random field (MRF) models to include prior contextual information. The comparison is based on artificial binary images which are degraded by Gaussian noise. Robustness is tested with correlated noise and with object and background textured. The ICM algorithm is evaluated when the degradation and model parameters must be estimated, in both supervised and unsupervised modes and on two real images. The results are assessed by visual inspection and through a numerical criterion. It is concluded that contextual information from MRF models improves segmentation when the number of categories and the degradation model are known and that parameters can be effectively estimated. None of the three algorithms is consistently best, but the ICM algorithm is the most robust. The energy of the a posteriori distribution is not always minimized at the best segmentation.>

The authors examine two important problems, estimation and goodness of fit, in modeling binary single-texture images by discrete Markov random fields. A methodology for comparing parameter estimators is proposed and applied to evaluate four estimation procedures. The classes of models considered are four-parameter Derin-Elliot models and four-parameter autobinomial models with second-order neighborhoods. A Min- chi /sup 2/ estimator is proposed and shown to outperform estimators described in the literature. The methodology is based on a hardcore sampling process over the parameter space and a Wilcoxon rank-sum statistic. A static for assessing the goodness of fit between a specific model and an arbitrary texture image is also proposed and used in a Monte Carlo ranking test. The statistic is experimentally validated on synthetic textures. Experiments on natural textures suggest that second-order binary models do not fit natural textures well.>

The design of a pattern recognition system requires careful attention to error estimation. The error rate is the most important descriptor of a classifier's performance. The commonly used estimates of error rate are based on the holdout method, the resubstitution method, and the leave-one-out method. All suffer either from large bias or large variance and their sample distributions are not known. Bootstrapping refers to a class of procedures that resample given data by computer. It permits determining the statistical properties of an estimator when very little is known about the underlying distribution and no additional samples are available. Since its publication in the last decade, the bootstrap technique has been successfully applied to many statistical estimations and inference problems. However, it has not been exploited in the design of pattern recognition systems. We report results on the application of several bootstrap techniques in estimating the error rate of 1-NN and quadratic classifiers. Our experiments show that, in most cases, the confidence interval of a bootstrap estimator of classification error is smaller than that of the leave-one-out estimator. The error of 1-NN, quadratic, and Fisher classifiers are estimated for several real data sets.