Advanced Materials Research Vols. 532-533

Abstract: Removing noise from the original image plays an important role in many important applications involving image-based medical diagnosis and visual material examination for public security, and so on. Among them, there have been several published methods to solve the related problem, however, each approach has its advantages, and limitations. This paper examines a new measure of denosing in space domain based on 2-D kernel regression which overcomes the difficulties found in other measures. The idea of this method mainly let the values of a row or a column from an image are taken as the measured results of a fitting function. The following step is to estimate the weight coefficients using least square method. Finally, we obtain an denoised image by resampling the estimated function, and the variable x denotes the coordinate of an image. Results of an experimental applications of this method analysis procedure are given to illustrate the proposed technique, and compared with the basic wavelet-thresholding algorithm for image denoising.

Abstract: The independency between two attribute subsets can be verified based on Chi square statistic to reduce candidate sets. Based on this measure, heuristic algorithm employing information entropy for reduction of decision systems is presented by combining rough sets and statistics. And the validity of this algorithm is analyzed.

Abstract: Universal steganalysis include feature extraction and steganalyzer design. Most universal steganalysis use Support Vector Machine (SVM) as steganalyzer. However, most SVM-based universal steganalysis are not to be very much effective at lower embedding rates. The reason why selective SVMs ensemble improve the generalization ability was analyzed, and an algorithm to select a part of individual SVMs according to their difference to build the ensemble classifier was proposed, which based on the selected ensemble theory-Many could be better than all. In this paper, the selective SVMs ensemble algorithm was used to construct a strong steganalyzer to improve the performance of steganographic detection. The twenty five experiments on the benchmark with 2000 different types of images show that: for popular steganography methods, and under different conditions of embedding rate, the average detection rate of proposed steganalysis method outperforms the maximum average detection rate for the steganalysis method based on single SVM with improving by 3.05%-12.05%; and for the steganalysis method based on BaggingSVM with improving by 0.2%-1.3%.

Abstract: A novel image segmentation algorithm based on fuzzy C-means (FCM) clustering and improved particle swarm optimization (PSO) is proposed. The algorithm takes global search results of improved PSO as the initialized values of the FCM, effectively avoiding easily trapping into local optimum of the traditional FCM and the premature convergence of PSO. Meanwhile, the algorithm takes the clustering centers as the reference to search scope of improved PSO algorithm for global searching that are obtained through hard C-means (HCM) algorithm for improving the velocity of the algorithm. The experimental results show the proposed algorithm can converge more quickly and segment the image more effectively than the traditional FCM algorithm.

Abstract: The local dynamic threshold segmentation algorithm, Niblack algorithm, is more suitable for hand vein images segmentation by comparing the common- used algorithms. But the traditional Niblack algorithm has weakness, so this paper proposed an improved Niblack algorithm in which the coefficient is adaptive estimated. The results show that the improved algorithm can better segment the hand vein outlines, preserve the vein original characteristics, and meet the needs of the follow-up feature extraction and recognition.

Abstract: The basic principles for stabilized gyrocompass initial alignment are analyzed in platform inertial navigation system (PINS), then similar principles and initial alignment algorithms suitable for programming are proposed for strapdown inertial navigation system (SINS). The scheme of SINS gyrocompass initial alignment can be divided into four steps, including leveling alignment with header uncertainty, coarse header alignment, leveling realignment and gyrocompass alignment for header. By simplifying SINS nonlinear error model under header uncertainty, the formula of coarse header alignment is deduced. On the assumption of navigation computer having large memory and powerful computing ability, and basing on the ‘multiformity’ of SINS mathematical platform and the ability to attitude reverse control, a specific progress for SINS rapid gyrocompass alignment is introduced and designed in detail. Finally, some tests prove that the proposed alignment algorithm in this paper is effective.

Abstract: By improving on the Quad-tree LOD algorithm, this paper proposes a new algorithm based on the terrain partitioning and simplifying the terrain by using grid models of different levels of detail dynamically.This algorithm can be reduced the number of triangular facets in the terrain scene.The experimental results show that this algorithm can generate continuous multi-resolution terrain model and improve the efficiency of real-time rendering.

Abstract: To optimize scheduling pertaining to two production units and many tasks in flexible production, a dynamic scheduling model was proposed considering capacity balance. Each production unit was modeled as a kind of container whose capacity symbolized the container capacity. The capacity was divided as planned and spared parts. Joining spared capacities in two units, the state-to-art dynamic scheduling rules were created according to hydro-mechanical theory. A practical heuristic algorithm was developed for this model. The model was tested as feasible and effective in a case study conducted at an actual business.

Abstract: Granular Computing theory is a interesting research direction in artificial intelligence field. In this paper, granular computing theory is applied to medical image segmentation. Granularity thinking in image segmentation is expounded, and a novel medical image segmentation method is proposed. Firstly, we construct different granularities according to different features that the image contained, secondly, do the attributes combination to the obtained quotient spaces according to the quotient space granularity synthesis principle, and then complete the image segmentation. Compared with the methods adopting single image feature, this method may fully use the image information in a more effective way and may obtain better segmentation effects.

Abstract: Chan-Vese model often leads to poor segmentation results for images with intensity inhomogeneity. Aiming at the gray uneven distribution in the night vehicle images, a new local Chan–Vese (LCV) model is proposed for image segmentation. The energy functional for the proposed model consists of three terms, i.e., global term, local term and regularization term. By incorporating the local image information into the proposed model, the images with intensity inhomogeneity can be efficiently segmented. Finally, experiments on nighttime plate images have demonstrated that our model can segment the nighttime plate images efficently. Moreover, comparisons with recent popular local binary fitting (LBF) model also show that our LCV model can segment images with few iteration times.