Abstract:The simulation of paper marbling through computers can help designers create unique, novel, and artistic paper marbling image patterns, which is important for the protection, record, and inheritance of this kind of ancient art. However, related techniques have not been researched intensively. In this paper, a new paper marbling method based on the techniques of fluid dynamics vector graphics is proposed. The method is fast and stable, and achieves clear silhouettes of the pattern boundaries. First, a semi-Lagrange method to solve the fluid equation and obtain the velocities of two-dimensional incompressible fluid field is used. Considering vector graphics has the advantage that can not be affected by the resolution, the study tracks the movement of the fluid boundary to generate the outline of the marbling image. Then, the paper renders the outline by the stencil method and exports the shape as the Scalable Vector Graphics (SVG) image. Moreover, the above method is extended for simulation of paper marbling generated with multiple dyes, which can generate more colorful marbling patterns. In addition, the study combines the method with the mathematical transform method to simulate marbling shapes with special patterns.

Abstract:The system design has been become an imperative part of the mechatronic system design with the increasing complexity. However, the current design platforms are all extended from the software modeling tools, which are essentially software-oriented and lack the support for the characteristics of complex mechatronics. Based on the analysis of key problems for the design of multi-domain complex mechatronics, a design-oriented modeling platform, M-Design, is developed for its design. First, the domain-independent fundamental semantics of SysML is represented in M-Design. Next, the domain-dependent design semantics is also modeled for various related domains of complex mechatronics. Finally, the transformation is designed to be conducted automatically from system design model to system simulation model to support the early simulation and find the irrational design schemes as early as possible. A case study is given to illustrate the proposed method.

Abstract:A complex sphere radial tree visualization technique is proposed for visualizing large scale unstructured images based on stereographic mapping. The requirement model of visual organization and interaction for hierarchical images is first studied. Next, a hierarchical organization pattern is proposed, and layout algorithm is introduced for the visualization technique based on modified radial tree method and stereographic mapping. It gives the application instance of this technique on large-scale medical images. Results show that this visualization method can illustrate image contents while preserving global context in a certain space and provides powerful and effective support for users' analysis.

Abstract:This paper presents an automatic line-drawing method designed for building images. The fast line segment detector with a false detection control and detects the outline and the door, or window edge features detected, which are used in the construction of edge tangent flow. Based on the modified ETF, the Difference-of-Gaussians is used to render the final line drawing styles. This scheme gives clear line features for building images without much noise, which is also simple and easy to implement. The test results verify the effectiveness of the presented algorithm.

Abstract:The Streamline based method is one of the most important vector field visualization methods. The physical-related feature was seldom considered in the past streamline placements algorithms. A streamline placement algorithm for 2D and 3D vector field is introduced in this paper. The study measures the difference between the inflow and the outflow to evaluate the local spatial-varying feature at a specified field point. A Difference of Inflow and Outflow Matrix (DIOM) is then calculated to describe the global appearance of the field. The study draws streamlines by choosing the local extreme points in DIOM as seeds. DIOM is somewhat like flow divergence and is physics-related and thus reflects the intrinsic characteristics of the vector field. The strategy performs well in revealing features of the vector field, even with relatively few streamlines in both 2D and 3D vector field.

Abstract:The Mumford-Shah model for planar image segmentation chas been extended to variational models for image segmentation on implicit surfaces in this paper. The closed surfaces are denoted as zero-level set of signed distance functions, and the open surfaces are expressed as interaction sets of zero-level set and binary functions. With the use of intrinsic gradients and intrinsic divergences, the Mumford-Shah model for image segmentation on implicit surfaces is formulated. In order to improve their generality, the paper uses a generalized smoothness term which can be exemplified for different types. In order to improve computation efficiency, the study designs the Split Bregman algorithms via introducing auxiliary variables and Bregman iterative parameters for the proposed models. Numerical examples validate the models and algorithms finally.

Abstract:First, through the wavelet transform of the thermopaint color image, the wavelet moduli are acquired and are used as eigenvalues together with color characteristics. Next, the image is segmented through a fuzzy C-means clustering of these eigenvalues. Experimental results show that this algorithm has a good effect on thermopaint color image segmentation compared to the conventional algorithms that uses sole wavelet transform and fuzzy clustering.

Abstract:The Projected Tetrahedra is a popular method in the field of tetrahedra database visualization. Tretrahedra must be sorted according to obstruction between them to achieve an accurate rendering image, but strong dependency among tetrahedra results in not only inefficient sorting, but also poor parallel execution. This paper proposes a tetrahedra sorting algorithm which is based on K-D tree spatial partitioning. The database in one leaf node are peeled into layers in natural order, and the tetrahedra in the same layer are unobstructed . The peeling of different leaf node is independent, and their sorted tetrahedra are organized together according to the obstruction between leaf nodes. The sorting efficiency has improved greatly through two-level parallelism and guarantees accurate sorting. The data structure can be implemented easily in a graphics processing unit (GPU). The experimental results show that the quick and accurate sorting based on K-D tree processed in GPU shortens the sorting time greatly.

Abstract:As an important part of book covers, characters contain rich semantic information. By extracting accurate information from complex color images, and combining it with content-based image retrieval technology, it is possible to further improve the accuracy of book retrieval. According to the characteristics of text information in Chinese book covers, this paper proposes connected components methods to locate the text regions. At first, the grayscale image is decomposed to a series of binary images and merged to connect components in each image, according to the structures of Chinese characters, generating candidate text regions. Additionally, text verification is used to rule out non-text regions. The result regions are regarded as the prominent regions of book cover, further this paper use Hu moment invariant to extract features for image matching. Experiments show the results of this method are fairly good, proving the importance of text information to book retrieval.

Abstract:The purpose of trademark retrieval is to ensure that new trademarks do not repeat any images of the vast number stored in the trademark registration system. This paper regards the contour of a trademark as a geometric object and represents it precisely in terms of mathematical expressions by employing an orthogonal complete function system, V-system, as a mathematical tool. The boundary feature vector, which captures the overall features of the trademark, is calculated in the frequency domain first. Next, two region feature vectors describing the local characteristics are created by dividing the trademark image into small blocks and considering the distributions of its pixels. Finally, a new trademark retrieval algorithm is achieved by utilizing the weighted Euclidean distance between boundary and region feature vectors. The study conducts different kinds of experiments and adopts several of the evaluation criterions to evaluate the performance of the proposed algorithm. Compared with the methods of classical Fourier descriptor, Zernike moments, Hu invariant moment and combination of Fourier descriptor and Zernike moments, the proposed algorithm has obvious advantages.

Abstract:The Gradient-Domain fusion is an effective method for image compositing, but suffers from color distortion when the hue of target image is quietly different from that of source image. The distortion can hardly be eliminated by boundary optimization. This paper proposes a method to correct the artifact of color distortion induced by gradient-domain fusion while keeping the seamless boundary. First, Poisson cloning is applied to luminance component for keeping the local contrast of intensity. Then, for each pixel of source image, a color belief is used to guide color correction and is estimated based on general geodesic distance transform for source image. Next, the color components of each pixel are corrected by minimizing an object function with corresponding color belief, while the object function takes neighborhood colors propagation, the original colors obtained by Poisson cloning are also taken into consideration. The final composites are obtained simply by combining the corrected color and the cloned luminance. Experimental results demonstrate that the proposed method reduces the color distortion of composition effectively while keeping the seamless boundary and requires only few interactions from user. Compared to other gradient-domain based image compositing methods that only optimize boundary conditions, the proposed method keeps both the original color and the relative changes of colors. The use of a soft color belief to compositing provides a smooth transition from foreground to background, which can not be done by hard constraints.

Abstract:As a key technology in content-based video indexing and management, the video shot boundary detection has attracted considerable research attention. However, the traditional detection algorithm can only handle a hard cut boundary. No satisfactory results have been achieved on gradual transition until now. A method based on the combination of adaptive threshold and Fourier fitting is presented in this paper to detect shot transition. A non-uniform histogram in HSV color space on each frame is first accumulated to calculate a similarity sequence of videos, depending on which thresholds are generated by the adaptive threshold to detect hard cut transition. Gradual transition has a much more complex transform in its span. After extensive experiments and observation, the study has found that fixed patterns exist in every gradual transition. Candidate transitions are detected by finding segments which show traits of a gradual transition pattern. Gradual transition boundaries of different types are collected to train a set of standard templates, which can be used to judge whether a candidate is a real gradual transition boundary and also, to further determine its transition type. The algorithm is accelerated by exploiting the parallel computation power of GPUs using CUDA. The effectiveness is verified through extensive experiments and is compared with other methods.

Abstract:Interactive tabletops and surfaces (ITS) are important components in computer supported co-located cooperation environments. ITS is different from the traditional WIMP interfaces and has the features of horizontal display, multiple input devices, and single display groupware. There is a lack of guidance in the development and design of ITS. A user interface model is proposed to solve the problem. First, a description method is proposed to describe multiple devices and the collaboration interaction, which adopts part of the methodology Penichet chosess to analyze person-computer-person interaction and also proposes an additional task-devices table (TDT) to ease the description of the interaction of different devices in ITS. Next, the ITS user interface metaphor OCAL (Objects Containers Adorners Languages) is described. Based on the analysis, Puerta's general computational framework is specific into ITSUIM (Interactive Tabletops and Surfaces User Interface Model) under the guide of OCAL. Finally, an application example is given. Experimental results show that the method proposed here is capable of being efficient enough to support the design and development of the user interfaces of ITS.

Abstract:The camera extrinsic parameters interpolation is one of the key technologies in the application of 3D scene roaming etc. Since the space constructed by camera extrinsic parameters is not linear, the traditional methods usually decompose the parameters into translation and rotation transformation, resulting in unnaturalness for interpolation. This paper presents a camera extrinsic parameters interpolation algorithm based on Lie Algebra. Firstl, the study transforms the camera extrinsic parameters matrix into a Lie Algebra space; then a linear interpolation in that space is finished; finally, the inverse transformation of the interpolation matrix into the space of camera extrinsic parameters obtains a series of sequence of interpolation result. Experiments show that, compared with the traditional interpolation directly in the space of camera extrinsic parameters and the interpolation based on quaternion, the interpolation sequences obtained by this algorithm is smoother and more in line with the people's observation habits.

Abstract:The study presents a new realistic rendering algorithm for a fluid scene with SPH (Smoothed Particle Hydrodynamics) in a screen space. First, the paper proposes an anisotropic based surface smoothing model: The study adopts the Laplacian Smoothing to acquire relative smooth particle position distribution in 3D space. Next an anisotropic modal is used to produce initial fluid surface. Last, a bilateral filtering depth map, we got the final desired smoothing fluid surface. Next, the study present a finer fluid surface illumination model to improve fluid scenes rendering, also using curvature based model to add realistic details of fluid. The study realizes that all these are in GPU. Compared with other methods, the approach can greatly improve the realism of fluid scenes rendering in screen space without losing the real-time performance, and could be found great potential applications in the domains such as real-time simulation and game, etc.

Abstract:Aliasing is a long-standing problem in computer graphics. Many techniques have been developed to anti-alias fast and efficiently. Based on the most advanced antialiasing algorithm, this article presents a new approach to deal with the drawback in edge detection and reconstructs the original morphological antialiasing technique. The new approach uses local edge information to calculate the range of the line slope, and then predicts more probable end-points for straight lines. While the reconstructing of straight lines is much better, this technique use minor additive cost. Compared with previous morphological antialiasing techniques, this work not only reconstructs the local edge shapes, but also captures the global edge shapes, which in return helps enormously in antialiasing lines. The use of such global slope-based shape information improves the gradients of straight lines, makes inclined lines more continuous, and therefore, obtains better rendering result.

Abstract:This paper proposes a high quality k-D tree constructing algorithm for interactive dynamic scene ray tracing. Combined with the fact that scene primitives were usually well distributed, a reasonable formula is derived from the basic k-D tree traversal cost function to represent the traversal cost for sequential positions in node. After dividing the bounding box of splitting node to uniform bin series, the best splitting position is directly computed by resolving the analytic solution of the cost formula. To guarantee superior rendering performance for different ray tracing applications, the study raised several sensible functions for sub-space number computing in different situations. Experimental results proved that this algorithm was suitable for scenes with various kinds of primitive distribution, and it can be used for full k-D tree construction. Great construction efficiency was obtained with the best splitting quality maintained.

Abstract:This paper proposes an algorithm for shadow-removal based on a single outdoor image. The algorithm consists of three steps: first, the study the characteristics of outdoor illumination and an color space to detect the shadow areas in the outdoor scene. Second, by taking detected shadows as background, the paper incororates the matting method to calculate the matting scale factor, with which the penumbra can receive consistent lighting. Finally, shadow pixels are relit by multiplying the corresponding scale factors derived from the matting scale factor. The experimental results demonstrate that the proposed approach can not only detect the self-shadow and cast-shadow, but also produce satisfactory shadow removal results in various outdoor scenes.

Abstract:Recently, many researchers have been attracted in link prediction, which is an effective technique \ used in graph based models analysis. By using the link prediction method the study understands associations between nodes. Most of previous works in this area have not explored the prediction of links in dynamic multi-dimension networks and have not explored the prediction of links which could disappear in the future. This paper argues that these kinds of links are important. At least they can serve as a complement for current link prediction processes in order to plan better for the future. This paper proposes a link prediction model, which is capable of predicting bi-direction links that might exist and may disappear in the future in dynamic multi-dimension networks. Firstly, the study presents the definition of multi-dimensional networks, reduction dimension networks, and dynamic networks. Then paper proposes a forward some algorithms which build multi-dimension networks, reduction dimension networks, and dynamic networks. Next, a give bi-direction link prediction algorithms in dynamic multi-dimension weighted networks. At the end, algorithms above are applied in recommendation networks. Experimental results show that the algorithm can improve the link prediction performance in dynamic multi-dimensional weighted networks.

Abstract:The visualization process usually only focuses on the results of the current visualization, losing the review and analysis of historical information. It implies that some important intermediate results are not tracked and detected timely, and often makes the information and the potential law invisible. The university graduate student information data is used as an example. When faculty in university visualize the students' personal information, they usually focus on the current view, ignoring the trace of the historical views, which contains important information or patterns. To address this problem and figure out the unknowns hidden in educational datasets, this paper proposes the historical views navigation through a similarity and closeness centrality based recommendation. In this approach, the useful intermediate views, or the interested views of the users are saved as history and compared with the current view. By analyzing the similarities between the closeness centrality and the path centrality, the most relative historical views are recommended to the user. Finally, the user study shows that most of the participants believe that this approach is a necessary alternative to properly integrate the current view and the historical views, which enhances user experience significantly.

Abstract:Obstructive Sleep Apnea Syndrome (OSAS) is the most common sleep-disordered breathing disease. Considering its severe effect on human health, long-term monitoring is necessary. However, the current OSAS monitoring systems require patients to stay in the hospital overnight and need considerable amount of wiring around the human body. This paper proposes a smartphone-based, wireless e-health system designed for OSAS, which offers anytime and anywhere monitoring. The study focuses on the energy issues of this battery-driven system, and makes full use of the low-power mode of Bluetooth, which has been proven to be the energy bottleneck, resulting in 50% increase of system lifetime. Compared with current methods for Bluetooth mode switch, this proposal, implemented on smartphone, improves system versatility and expandability, and provides technical support for holistic energy-saving approach based on load balancing among data storage, transmission and computation.

Abstract:The effect of carpooling is very significant in many aspects such as reducing the cost of logistics and traffic congestion. In this paper, a clustering heuristic strategy based on matching degree is introduced to assign the demand of services to one specific vehicle. In a single vehicle problem, a prior clustering idea to reduce the number of insertion operation is induced which may improve the efficiency of the algorithm. Additionally, migration operator is proposed to improve the success rate of matching and to reduce total costs. Real examples show that the algorithm not only cuts down a vehicle's idling rate, but also reduces the vehicle operation cost to some degree.