针对脆性物体破碎模拟时存在的表面破碎边界难确定和计算速度慢的问题,提出一种新的基于三维Voronoi图和布尔运算的混合破碎模拟方法.首先为脆性物体构建AABB包围盒,并采用三维Voronoi图方法对包围盒进行剖分以生成碎块,提出用BSP树将物体划分成多个凸集的树状集合,并求出碎块与这些凸集的交集从而快速且准确地获取表面破碎边界.最后根据能量守恒和动量守恒定理对脆性物体的破碎进行仿真计算.实验结果表明,算法在保证实时性的同时,具有较强的画面真实感,尤其适用于具有复杂表面的脆性物体破碎模拟.

目的 探讨基于改进移动立方体算法的腹部器官CT图像三维重建效果.方法 提出一种基于区域增长法的通用树结构和移动等值点法的自适应改进移动立方体算法,先进行医学图像分割,选取种子点后标记出与阈值相交的所有体元;创建通用树结构,将相交体元插入子节点中,确定基于通用树的顶点索引方式;通过移动等值点合并共面三角形,简化等值点信息的获取.基于1名志愿者的腹部CT图像,采用传统移动立方体算法和改进移动立方体算法构建肾脏三维模型,并比较其效果.结果 与传统算法比较,改进的移动立方体算法生成的三角面片个数减少39.20％,算法执行效率提高37.59％,三维模型表面平滑逼真,局部细节真实性较好.结论 基于改进移动立方体的算法可更快速精确地实现CT图像腹部器官三维重建.

To solve the problem that the selection of the noise parameter relies heavily on experience value,an improved WNNM algorithm for adaptive parameter selection was proposed.The advantage of the algorithm was to build the noise evaluation model based on the traditional WNNM algorithm.The image features were constructed by combining the statistical features of the mean subtracted contrast normalized (MSCN) coefficients and its four direction neighboring MSCN coefficients,and the sample set was constituted by the image feature and the noise concentration corresponding to it.The sample set was trained using the support vector regression method to obtain the noise evaluation model,which provided the optimal parameter for WNNM algorithm.Experimental results show that the proposed algorithm is more efficient and has better robustness and generalization performances than traditional method.

Modeling and collision detection algorithm is the premise of the real-time in virtual surgery. Here,we extract the CT point cloud data from patients.Then,we model for soft tissues and organs based on the algorithm of octree subdivision and hierarchical structure of bounding sphere.In order to im-prove the real-time of the collision detection,the physical models of the surgical instruments are simpli-fied to balls or lines.The geometry remains unchanged,and it thus does not affect the operation of the virtual visual while the speed of the collision detection is improved greatly.The experimental results show that the algorithm can accurately detect the point of virtual contact with the virtual model of surgi-cal instruments,and collision detection in real-time have increased considerably.Simplified average colli-sion detection time is 10% of the original.

In order to control the dip angle and speed of the robot simultaneously on robot modeling,based on the mathematical models of two-wheeled patrol robot,Linear Quadratic Regulator(LQR) controller is designed for the system.The condition for system stability is developed so that the proper control parameters can be chosen to guarantee the whole system stability.In actual application,Kalman filtering algorithm is used to acquire dip angle by fusing data from the accelerometer and the gyroscope.The simulation in Matlab-Simulink platform and real-time control in real robot platform experiments result show that based on Kalman filtering and LQR algorithm,the two-wheeled robot can run erectly at the specified speed and has strong anti-interference performance.

In view that the real-time attitude calculation of caterpillar robot solved by using the gyroscope or accelerometer and magnetometer separately has the problem of low accuracy and is susceptible to interference,the multi-sensor data fusion algo-rithm was researched.The attitude of the robot was represented by quaternion.The sensor information was fused using Kalman filtering algorithm.The state equation was built through gyroscope data combining the fourth order Runge-Kutta method for im-proving calculation accuracy,and the observation equation was built by accelerometer and magnetometer data using factored qua-ternion algorithm (FQA)for saving hardware calculation time.The attitude measurement system was established using STM32 controller.Experimental results show that the system can effectively finish the fusion of multi-sensor information.The errors of pitch and roll are within ±1°and the error of yaw is within ±2°,which met the accuracy requirement of the robot’s attitude.

What operation instruments interact with are human tissues and organs in virtual surgery. Because its impedance is nonlinear and unpredictable, the haptic interaction is difficult to be stable, especially when interacting with rigid tissues such as bone. To solve this difficult problem, this paper presents a sliding mode control algorithm based on Lyapunov theory to realize stable operation for virtual surgery which is never seen in previous research. The simulation results show that 1) the haptic interaction can remain stable when interacting with both soft tissues and large impedance tissues such as bone; 2) the haptic interaction can remain stable even when operation instruments interact with nonlinear impedance tissues.

针对点云数据的Delaunay三角网格纹理映射速度慢、映射效果不够细腻及不适合大规模点云数据纹理映射等问题,提出一种基于球面纹理映射的点云数据重建改进方法,并在Qsplat算法的基础上进行实现。采用Qsplat算法对大规模点云数据进行模型重建,利用球面等比约束纹理映射算法建立纹理坐标、球面、点云重建模型三者之间的数学关系,实现大规模点云数据的球面纹理映射。实验结果表明,与传统的三角网格纹理映射相比,该方法可明显提高纹理映射的速度和质量,拓宽球面等比约束纹理映射方法的应用范围,适用于大规模点云数据的纹理映射。

In order to obtain high quality simulation effect with low time and space complexity, and the stability of system,a novel improved algorithm for shape mathcing based on Splat graphic element is presented. In the algorithm,a new Splat graphic element is adopted instead of the classical point graphic element,the surface of object is seamlessly covered with the least number of Splats to ensure rendering quality,which can be achieved by controlling the sampling density and automatically adjusting the radius of circular Splats. The deformation of Splats is calculated with shape matching algorithm. Experimental results show that for the same geometric model,the new algorithm can reduce about 50% storage space and improves the computational efficiency by about two times compared with the classical algorithm. The algorithm has stability in dynamic simulation.

Two wheeled robot is a multi variable,higher order non linear,strong coupling system. A novel double loop PID control method was presented,and the control system for the two wheeled robot was designed. The real time inclination of robot was measured by fusing the gyro signal with accelerometer signal with a Kalman filtering method. The speed of the robot was measured with encoders. The double loop PID controller combined positive feedback with negative feedback controller. The sum of inclination negative feedback and speed positive feedback were output to control the robot balance and walking with a certain expected speed. Actual experiments show that the double loop PID control method can make two wheeled robot walking smooth in accordance with the expected speed and the system has a good anti interference performance. The control method also has the advantages of traditional PID controller,which does not rely on an accurate model,easy to implement and parameter setting,and has strong robustness.

By the Hermite polynomicals method,an approach to approximate Conic sections in the form of a rational Bezier curve with Hermite polynomial curves is studied.The property condition of constructed Hermite polynomial curve such as G2-continuity with the Conic section at the end points and G1-continuity at the parametric mid-point and shape-preserving has been proposed.Explicit error bound is also derived and discussed.The validity of the proposed method for approximating Conic sections with Hermite polynomial curves is further proved through multiples sets of different types of comparative tests.

The local binary pattern(LBP) descriptors employed in the content-based image retrieval system lack the abilities to describe the spatial relationships and have longer dimension of feature vector.In this paper,an improved LBP(ILBP) texture descriptor based on spatial statistical feature of LBP code pair is proposed.The original image is converted to the LBP pseudo image using LBP coding method for micro-pattern,and then several statistics of LBP code pair are extracted to form the feature vector for describing the texture attributes of images.Experiments are preformed on the content-based image retrieval prototype platform.Experimental results show that compared with other LBP descriptors the ILBP descriptor further enhances the description ability of LBP descriptor and substantially reduces the feature vector dimension with better query accuracy and query efficiency.

Effective image segmentation is an important task in computer vision.In view of computational complexity and poor description of the image segmentation by maximal similarity based region merging(MSRM),a novel fast image segmentation algorithm,i.e.,improved MSRM(IMSRM),using local binary pattern(LBP) to calculate the similarity between the adjacent regions is proposed.LBP texture descriptor,which encodes the local micro-structure between the image pixels to achieve a description of their spatial relationship,effectively improves the description capability of the region feature,the obtained feature vector dimension is much smaller than the color histogram,and greatly improves calculation efficiency of the adjacent area similarity.The proposed algorithm automatically merges the regions which are over segmented by mean shift algorithm,with the marker indicating the region of the object and background.The region merging process is adaptive to the image content and it does not need to set the similarity threshold in advance.A large number of experiments compared with MSRM algorithm show that the IMSRM algorithm can effectively extract outline of the object from a variety of complex backgrounds with better edge details,and the efficiency of the algorithm can be improved by about 50%.

Aiming at promoting execution efficiency and registration accuracy of dense scale invariant feature transform(SIFT)flow based image registration algorithm,in this paper,a new improved dense local self-similarity(ILSS)flow based registration algorithm is proposed.By separating color and luminance information of color images through the color space conversion,the improved image registration algorithm only extracts dense self-similar feature on the channel of luminance to improve efficiency of image feature extraction stage substantially.With constraints of maintaining the characteristics of the flow field smoothness,we then employ pyramid multi-resolution iterative method to improve the computational efficiency of the estimation of the self-similar flow field stage.The basic idea of the multi-resolution iterative method is to roughly estimate the dense ILSS flow at a coarse level of image grid,then gradually propagate and refine the dense flow from coarse to fine.A large number of experimental results show that compared with dense SIFT flow based image registration algorithm,the dense ILSS flow based one allows robust matching across different scene appearances and has higher computational efficiency and registration accuracy.The success on these experiments suggests that the image registration using dense ILSS flow can be a useful tool for various applications in computer vision and computer graphics.

A fast image segmentation algorithm based on region feature is proposed to estimate centroid number.In the preprocessing analysis stage,the feature vector based on the cooccurrence matrix statistics is used to describe the regional characteristics of sub-image,and the proposed algorithm combines with cluster validity function to estimate accurate centroid number and initialization of membership matrix.In the main clustering stage,the implicit feature of color and texture extracted by Gabor filter is used to accomplish clustering,which not only produces a more reasonable quality of region segmentation,but also has fine noise immunity.The experimental results show that the proposed algorithm effectively overcomes the deficiencies of pixel-level estimations,greatly accelerates the iterative speed of the FCM main clustering stage and achieves higher efficiency in the implementation.

We introduce principle of underground pipelines detection,and also simulate electromagnetic induction curve of single pipe,two-crossed pipes,two parallel-contour pipes.From those curves,we know that this method is quite appropriate to pipe detection.We simulate electromagnetic induction curve of two parallel and non-contour pipes.So we can know exact position of pipes.In the last,we point out that shortcoming and insufficiency of this method in this case,and specify future direction of this field.

For permanent magnet synchronous motor(PMSM) existing the chaotic movement,The paper puts forw ard a kind of existence of the disturbance of permanent magnet synchronous motor(PMSM) of the sliding mode variable structure control of chaos synchronization method.Based on sliding mode variable structure control theory,it designs a high robust adaptive controller,even outside interference and parameter conditions of uncertainty,also can realize the sliding mode.The results of numerical simulation show that the method can be further realize the PMSM chaotic motions of the synchronization control,and has good control effect.

Multi-degree elevation and degree reduction of B-Spline curves in a matrix representation was proposed.The control points of the degree elevated or reduced B-Spline curve can be obtained as a product of the transfor-mation matrix and the vector of original control points.The method does not need knots inserting or knots refin-ment.In this way,the process of degree changing of B-spline simply express.

In order to describe the complex nonlinear relationship between yield and the six factors,including soil nitrogen(N),phosphorus(P),potassium(K) concentration and N,P,K fertilizer input,we have improved the original BP neural network.Firstly,by revising the parameters,namely weight value and threshold value of ANN repeatedly with Simulated Annealing(SA),the performance of ANN was improved tremendously.Secondly,we adopted Genetic Algorithm(GA) to improve the same parameters of the neural network.At last,we compared the performances between the two methods and made the conclusion that the BP neural network which is based on Simulated Annealing and Genetic Algorithm has a better performance.

In the panoramic camera image there is serious noise interference, and it is not easy to extract and other issues. It is imminent to do the image restoration, image recognition to these image. This paper focuses on the equidistant discretization algorithm, in image pretreatment, and in image feature extraction study, focuses on image feature extraction algorithms based on attribute importance,converts the abstract features into a meaning explicit knowledge representation system, and then design effective simulation experiment to verify the correctness of the algorithm.