对边缘不突出的显微镜影像进行边缘特征提取与Hough直线检测

文章主要介绍如何对边缘不突出的显微镜进行边缘特征提取与Hough直线检测,其中涉及到数字图像处理中的边缘检测、图像二值化及Hough直线检测算法。     

精确实验数据曲线方程的求解

针对精确实验数据的曲线拟合方程化处理方法已不再适应高精度测试技术的问题,提出了用三次样条插值函数进行曲线方程化处理的新方法,并创建了用MATLAB语言编写遗传算法程序进行曲线斜率求解的方法,通过实例表明,采用遗传算法求解出的三次样条插值函数非常适合精确实验数据的曲线方程化处理。     

进口刀盘圆环失效原因及改进

首先从材料力学角度分析了抚顺石化公司干法腈纶装置原进口切断刀盘圆环失效的原因 ,认为此种切断刀盘圆环系二维增强结构层压板 ,层间没有纤维增强 ,需要解决由二维增强向三维增强的过渡。给出了改进办法 ,寻求到三维增强的替代材料。同时介绍了机械加工测试及在线实验     

Reliability analysis of a repairable duplex system

We analyse the survival time of a repairable duplex system characterised by cold standby and by a pre-emptive priority rule. We allow general probability distributions for failure and repair. Moreover, an important realistic feature of the system is the general assumption that the non-priority unit has a memory. This combination of features has not been analysed in the previous literature. Our (new) methodology is based on a concatenation of a Cauchy-type integral representation of the modified Heaviside unit-step function and a two-sided stochastic inequality. Finally, we introduce a security interval related to a security level and a suitable risk-criterion based on the survival function of the system. As a practical application, we analyse some particular cases of the survival function jointly with the security interval corresponding to a security level of 90.     

A parametric frequency response method for non-linear time-varying systems

A new parametric frequency response algorithm is introduced to investigate linear and non-linear dynamic systems with time-varying parameters. In the new algorithm the time-varying parameters are regarded as additional inputs of the systems and the non-linear generalised frequency response functions for multi-input-single-output systems are then employed to obtain Zadeh's system functions from a differential equation representation. The parametric frequency response method reveals how the time-varying parameters affect the behaviour of the systems through a time-varying term. The new method can be applied to both linear and non-linear time-varying systems.     

3D soft-tissue tracking using spatial-color joint probability distribution and thin-plate spline model

Visual tracking techniques based on stereo endoscope are developed to measure tissue motion in robot-assisted minimally invasive surgery. However, accurate 3D tracking of tissue surfaces remains challenging due to complicated deformation, poor imaging conditions, specular reflections and other dynamic effects during surgery. This study employs a robust and efficient 3D tracking scheme with two independent recursive processes, namely kernel-based inter-frame motion estimation and model-based intra-frame 3D matching. In the first process, target region is represented in joint spatial-color space for robust estimation. By defining a probabilistic similarity measure, a mean-shift-based iterative algorithm is derived for location of the target region in a new image. In the second process, the thin-plate spline model is used to fit the 3D shape of tissue surfaces around the target region. An iterative algorithm based on an efficient second-order minimization technique is derived to compute optimal model parameters. The two processes can be computed in parallel. Their outputs are combined to recover 3D information about the target region. The performance of the proposed method is validated using phantom heart videos and in vivo videos acquired by the daVinci^®${\mathrm{daVinci}}^{\circledR }$ surgical robotic platform and a synthesized data set with known ground truth.     

Sliding-mode and proportional-derivative-type motion control with radial basis function neural network based estimators for wheeled vehicles

An obstacle avoidance problem of rear-steered wheeled vehicles in consideration of the presence of uncertainties is addressed. Modelling errors and additional uncertainties are taken into consideration. Controller designs for driving and steering motors are designed. A proportional-derivative-type driving motor controller and a sliding-mode steering controller combined with radial basis function neural network (RBFNN) based estimators are proposed. The convergence properties of the RBFNN-based estimators are proven by the Stone–Weierstrass theorem. The stability of the proposed control law is proven using Lyapunov stability analysis. The obstacle avoidance strategy utilising the sliding surface adjustment to an existing navigation method is presented. It is concluded that the driving velocity and steering-angle performances of the proposed control system are satisfactory.     

Mixed H2/H∞ robust model predictive control with saturated inputs

In this paper, we investigate the mixed H₂/H_∞ robust model predictive control (RMPC) for polytopic uncertain systems, which refers to the infinite horizon optimal guaranteed cost control (OGCC). To fully use the capability of actuators, we adopt a saturating feedback control law as the control strategy of RMPC. As the saturating feedback control law can be effectively represented by the convex hull of a group of auxiliary linear feedback laws, the auxiliary feedback laws allow us to design the actual feedback control law without consideration of the input constraints directly to achieve the improved performance. Moreover, we suggest the relative weights on the actual and auxiliary feedback laws to the RMPC, which in turn improves the closed-loop system performance. Furthermore, an off-line design of the proposed RMPC is also developed to make it more practical. Numerical studies demonstrate the effectiveness of the proposed algorithm.     

The C-loss function for pattern classification

This paper presents a new loss function for neural network classification, inspired by the recently proposed similarity measure called Correntropy. We show that this function essentially behaves like the conventional square loss for samples that are well within the decision boundary and have small errors, and L₀ or counting norm for samples that are outliers or are difficult to classify. Depending on the value of the kernel size parameter, the proposed loss function moves smoothly from convex to non-convex and becomes a close approximation to the misclassification loss (ideal 0–1 loss). We show that the discriminant function obtained by optimizing the proposed loss function in the neighborhood of the ideal 0–1 loss function to train a neural network is immune to overfitting, more robust to outliers, and has consistent and better generalization performance as compared to other commonly used loss functions, even after prolonged training. The results also show that it is a close competitor to the SVM. Since the proposed method is compatible with simple gradient based online learning, it is a practical way of improving the performance of neural network classifiers.     

Data-driven asymptotic stabilization for discrete-time nonlinear systems

In this paper, we propose a data-driven feedback controller design method based on Lyapunov approach, which can guarantee the asymptotic stability of the closed-loop and enlarge the estimate of domain of attraction (DOA) for the closed-loop. First, sufficient conditions for a feedback controller asymptotically stabilizing the discrete-time nonlinear plant are proposed. That is, if a feedback controller belongs to an open set consisting of pairs of control input and state, whose elements can make the difference of a control Lyapunov function (CLF) to be negative-definite, then the controller asymptotically stabilizes the plant. Then, for a given CLF candidate, an algorithm, to estimate the open set only using data, is proposed. With the estimate, it is checked whether the candidate is or is not a CLF. If it is, a feedback controller is designed just using data, which satisfies sufficient conditions mentioned above. Finally, the estimate of DOA for closed-loop is enlarged by finding an appropriate CLF from a CLF candidate set based on data. Because the controller is designed directly from data, complexity in building the model and modeling error are avoided.