Fuzzy Connectedness Image Segmentation in Graph Cut Formulation: A Linear-Time Algorithm and a Comparative Analysis

A deep theoretical analysis of the graph cut image segmentation framework presented in this paper simultaneously translates into important contributions in several directions. The most important practical contribution of this work is a full theoretical description, and implementation, of a novel powerful segmentation algorithm, GC(max). The output of GC(max) coincides with a version of a segmentation algorithm known as Iterative Relative Fuzzy Connectedness, IRFC. However, GC(max) is considerably faster than the classic IRFC algorithm, which we prove theoretically and show experimentally. Specifically, we prove that, in the worst case scenario, the GC(max) algorithm runs in linear time with respect to the variable M=|C|+|Z|, where |C| is the image scene size and |Z| is the size of the allowable range, Z, of the associated weight/affinity function. For most implementations, Z is identical to the set of allowable image intensity values, and its size can be treated as small with respect to |C|, meaning that O(M)=O(|C|). In such a situation, GC(max) runs in linear time with respect to the image size |C|. We show that the output of GC(max) constitutes a solution of a graph cut energy minimization problem, in which the energy is defined as the a"" (a) norm ayenF (P) ayen(a) of the map F (P) that associates, with every element e from the boundary of an object P, its weight w(e). This formulation brings IRFC algorithms to the realm of the graph cut energy minimizers, with energy functions ayenF (P) ayen (q) for qa[1,a]. Of these, the best known minimization problem is for the energy ayenF (P) ayen(1), which is solved by the classic min-cut/max-flow algorithm, referred to often as the Graph Cut algorithm. We notice that a minimization problem for ayenF (P) ayen (q) , qa[1,a), is identical to that for ayenF (P) ayen(1), when the original weight function w is replaced by w (q) . Thus, any algorithm GC(sum) solving the ayenF (P) ayen(1) minimization problem, solves also one for ayenF (P) ayen (q) with qa[1,a), so just two algorithms, GC(sum) and GC(max), are enough to solve all ayenF (P) ayen (q) -minimization problems. We also show that, for any fixed weight assignment, the solutions of the ayenF (P) ayen (q) -minimization problems converge to a solution of the ayenF (P) ayen(a)-minimization problem (ayenF (P) ayen(a)=lim (q -> a)ayenF (P) ayen (q) is not enough to deduce that). An experimental comparison of the performance of GC(max) and GC(sum) algorithms is included. This concentrates on comparing the actual (as opposed to provable worst scenario) algorithms' running time, as well as the influence of the choice of the seeds on the output.

Towards a hybrid approach for adapting web graphical user interfaces to heterogeneous devices using context

Ubiquitous Computing promises seamless access to a wide range of applications and Internet based services from anywhere, at anytime, and using any device. In this scenario, new challenges for the practice of software development arise: Applications and services must keep a coherent behavior, a proper appearance, and must adapt to a plenty of contextual usage requirements and hardware aspects. Especially, due to its interactive nature, the interface content of Web applications must adapt to a large diversity of devices and contexts. In order to overcome such obstacles, this work introduces an innovative methodology for content adaptation of Web 2.0 interfaces. The basis of our work is to combine static adaption - the implementation of static Web interfaces; and dynamic adaptation - the alteration, during execution time, of static interfaces so as for adapting to different contexts of use. In hybrid fashion, our methodology benefits from the advantages of both adaptation strategies - static and dynamic. In this line, we designed and implemented UbiCon, a framework over which we tested our concepts through a case study and through a development experiment. Our results show that the hybrid methodology over UbiCon leads to broader and more accessible interfaces, and to faster and less costly software development. We believe that the UbiCon hybrid methodology can foster more efficient and accurate interface engineering in the industry and in the academy.

ANALYSIS OF THE NATURAL VIBRATION FREQUENCY TEST METHOD TO OBTAIN THE MODULUS OF ELASTICITY OF WOOD STRUCTURAL COMPONENTS

There are several techniques to characterize the elastic modulus of wood and those currently using the natural frequencies of vibration stand out as they are non-destructive techniques, producing results that can be repeated and compared over time. This study reports on the effectiveness of the testing methods based on the natural frequencies of vibration versus static bending to obtain the elastic properties of reforested structural wood components usually employed in civil construction. The following components were evaluated: 24 beams of Eucalyptus sp. with nominal dimensions (40 x 60 x 2.000 mm) and 14 beams of Pinus oocarpa with nominal dimensions (45 x 90 x 2.300 mm) both without treatment; 30 boards with nominal dimensions (40 x 240 x 2.010 mm) and 30 boards with nominal dimensions (40 x 240 x 3.050 mm), both of Pinus oocarpa and with chromate copper arsenate (CCA) preservative treatment. The results obtained in thiswork show good correlation when compared to the results obtained by the static bending mechanical method, especially when applying the natural frequency of longitudinal vibration. The use of longitudinal frequency was reliable and practical, therefore recommended for determining the modulus of elasticity of wood structural elements. It was also found that no specific support is needed for the specimens using the longitudinal frequency, as well as no previous calibrations, reducing the execution time and enabling to test many samples.

An online data access prediction and optimization approach for distributed systems

Current scientific applications have been producing large amounts of data. The processing, handling and analysis of such data require large-scale computing infrastructures such as clusters and grids. In this area, studies aim at improving the performance of data-intensive applications by optimizing data accesses. In order to achieve this goal, distributed storage systems have been considering techniques of data replication, migration, distribution, and access parallelism. However, the main drawback of those studies is that they do not take into account application behavior to perform data access optimization. This limitation motivated this paper which applies strategies to support the online prediction of application behavior in order to optimize data access operations on distributed systems, without requiring any information on past executions. In order to accomplish such a goal, this approach organizes application behaviors as time series and, then, analyzes and classifies those series according to their properties. By knowing properties, the approach selects modeling techniques to represent series and perform predictions, which are, later on, used to optimize data access operations. This new approach was implemented and evaluated using the OptorSim simulator, sponsored by the LHC-CERN project and widely employed by the scientific community. Experiments confirm this new approach reduces application execution time in about 50 percent, specially when handling large amounts of data.

MR-Radix: a multi-relational data mining algorithm

Abstract Background Once multi-relational approach has emerged as an alternative for analyzing structured data such as relational databases, since they allow applying data mining in multiple tables directly, thus avoiding expensive joining operations and semantic losses, this work proposes an algorithm with multi-relational approach. Methods Aiming to compare traditional approach performance and multi-relational for mining association rules, this paper discusses an empirical study between PatriciaMine - an traditional algorithm - and its corresponding multi-relational proposed, MR-Radix. Results This work showed advantages of the multi-relational approach in performance over several tables, which avoids the high cost for joining operations from multiple tables and semantic losses. The performance provided by the algorithm MR-Radix shows faster than PatriciaMine, despite handling complex multi-relational patterns. The utilized memory indicates a more conservative growth curve for MR-Radix than PatriciaMine, which shows the increase in demand of frequent items in MR-Radix does not result in a significant growth of utilized memory like in PatriciaMine. Conclusion The comparative study between PatriciaMine and MR-Radix confirmed efficacy of the multi-relational approach in data mining process both in terms of execution time and in relation to memory usage. Besides that, the multi-relational proposed algorithm, unlike other algorithms of this approach, is efficient for use in large relational databases.

Matrix-vector multiplication and triangular linear solver using GPGPU for symmetric positive definite matrices derived from elliptic equations

The modern GPUs are well suited for intensive computational tasks and massive parallel computation. Sparse matrix multiplication and linear triangular solver are the most important and heavily used kernels in scientific computation, and several challenges in developing a high performance kernel with the two modules is investigated. The main interest it to solve linear systems derived from the elliptic equations with triangular elements. The resulting linear system has a symmetric positive definite matrix. The sparse matrix is stored in the compressed sparse row (CSR) format. It is proposed a CUDA algorithm to execute the matrix vector multiplication using directly the CSR format. A dependence tree algorithm is used to determine which variables the linear triangular solver can determine in parallel. To increase the number of the parallel threads, a coloring graph algorithm is implemented to reorder the mesh numbering in a pre-processing phase. The proposed method is compared with parallel and serial available libraries. The results show that the proposed method improves the computation cost of the matrix vector multiplication. The pre-processing associated with the triangular solver needs to be executed just once in the proposed method. The conjugate gradient method was implemented and showed similar convergence rate for all the compared methods. The proposed method showed significant smaller execution time.

Influence of cueing on the preparation and execution of untrained and trained complex motor responses

This study investigated the influence of cueing on the performance of untrained and trained complex motor responses. Healthy adults responded to a visual target by performing four sequential movements (complex response) or a single movement (simple response) of their middle finger. A visual cue preceded the target by an interval of 300, 1000, or 2000 ms. In Experiment 1, the complex and simple responses were not previously trained. During the testing session, the complex response pattern varied on a trial-by-trial basis following the indication provided by the visual cue. In Experiment 2, the complex response and the simple response were extensively trained beforehand. During the testing session, the trained complex response pattern was performed in all trials. The latency of the untrained and trained complex responses decreased from the short to the medium and long cue-target intervals. The latency of the complex response was longer than that of the simple response, except in the case of the trained responses and the long cue-target interval. These results suggest that the preparation of untrained complex responses cannot be completed in advance, this being possible, however, for trained complex responses when enough time is available. The duration of the 1st submovement, 1st pause and 2nd submovement of the untrained and the trained complex responses increased from the short to the long cue-target interval, suggesting that there is an increase of online programming of the response possibly related to the degree of certainty about the moment of target appearance.

Nerve fiber layer in glaucomatous hemifield loss: a case-control study with time- and spectral-domain optical coherence tomography

Purpose: To evaluate the retinal nerve fiber layer measurements with time-domain (TD) and spectral-domain (SD) optical coherence tomography (OCT), and to test the diagnostic ability of both technologies in glaucomatous patients with asymmetric visual hemifield loss. Methods: 36 patients with primary open-angle glaucoma with visual field loss in one hemifield (affected) and absent loss in the other (non-affected), and 36 age-matched healthy controls had the study eye imaged with Stratus-OCT (Carl Zeiss Meditec Inc., Dublin, California, USA) and 3 D OCT-1000 (Topcon, Tokyo, Japan). Peripapillary retinal nerve fiber layer measurements and normative classification were recorded. Total deviation values were averaged in each hemifield (hemifield mean deviation) for each subject. Visual field and retinal nerve fiber layer "asymmetry indexes" were calculated as the ratio between affected versus non-affected hemifields and corresponding hemiretinas. Results: Retinal nerve fiber layer measurements in non-affected hemifields (mean [SD] 87.0 [17.1] mu m and 84.3 [20.2] mu m, for TD and SD-OCT, respectively) were thinner than in controls (119.0 [12.2] mu m and 117.0 [17.7] mu m, P<0.001). The optical coherence tomography normative database classified 42% and 67% of hemiretinas corresponding to non-affected hemifields as abnormal in TD and SD-OCT, respectively (P=0.01). Retinal nerve fiber layer measurements were consistently thicker with TD compared to SD-OCT. Retinal nerve fiber layer thickness asymmetry index was similar in TD (0.76 [0.17]) and SD-OCT (0.79 [0.12]) and significantly greater than the visual field asymmetry index (0.36 [0.20], P<0.001). Conclusions: Normal hemifields of glaucoma patients had thinner retinal nerve fiber layer than healthy eyes, as measured by TD and SD-OCT. Retinal nerve fiber layer measurements were thicker with TD than SD-OCT. SD-OCT detected abnormal retinal nerve fiber layer thickness more often than TD-OCT.