Advanced Materials Research Vols. 989-994

Abstract: The Lord of the formula of disjunctive normal form is an important content in propositional logic. When the propositional variable is large, the paper gives a formula for the main disjunctive normal form of simple method.

Abstract: Online long-term tracking is a challenging problem as data streams change over time. In this paper, sparse representation has been applied to visual tracking by finding the most correct sample with minimal reconstruction error using compressed Haar-like features. However, most sparse representation tracking algorithm introduce l₁ regularization into the PCA reconstruction using samples directly, which leads to complexity computation and can not adapt to occlusion, rotation and change in size. Our model updating not only uses the samples from the training set, but also generates the warped versions (include scale variation, rotation, occlusion and illumination changes) for the previous tracking result. Also, we do not use the samples in models for sparse representation directly, but the Haar-like features instead which are compressed in a very low-dimensional space. In addition, we use a robust and fast algorithm which exploits the spatio-temporal context for predicting the target location in the next frame. This step will lead to the reduction of the searching range by the detector. We demonstrate the proposed method is able to track objects well under pose and scale variation, rotation, occlusion and illumination with great real-time performance on challenging image sequences.

Abstract: In this paper, on the basis of the robot visual servo system, a virtual visual servo algorithm for pose estimation is put forward. By reducing the image error between the current image features of real object and the features of virtual models, the rotation and translation relation of the virtual model is changed by visual servo control law, until the pose of virtual model overlaps the one of real object, and then object pose estimation is obtained. In addition, RANSAC algorithm is also introduced to improve the robustness of the proposed algorithm. Finally the experiment about camera external parameters estimation shows that our algorithm gets better results than the available parameter estimation algorithms.

Abstract: This paper proposes a new path planning algorithm based on the CNN model. The path planning problem is completed with the dynamics of CNN by establishing a relationship between path control points and CNN cells. Based on the analysis of one dimensional space of CNN algorithm, a CNN equation is constructed and the path updating algorithm under the curvature constraint is obtained, then the stability of the algorithm is discussed. Path planning simulation based on two-dimensional space shows that this algorithm can avoid re-planning or falling into the local minimum, which means it can be successfully used in the path planning and maintenance of robots on the ground in dynamic environment.

Abstract: Brain Storm Optimization (BSO) is a novel proposed swarm intelligence optimization algorithm which has a fast convergent speed. However, it is easy to trap into local optimal. In this paper, a new model based on niche technology, which is named Niche Brain Storm Optimization (NBSO), is proposed to overcome the shortcoming of BSO. Niche technology effectively prevents premature and maintains population diversity during the evolution process. NBSO shows excellent performance in searching global value and finding multiple global and local optimal solutions for the multi-peak problems. Several benchmark functions are introduced to evaluate its performance. Experimental results show that NBSO performs better than BSO in global searching ability and faster than Niche Genetic Algorithm (NGA) in finding peaks for multi-peak function.

Abstract: Product data is the source data of product lifecycle in manufacturing enterprise. The quality of product data largely determines the effect of the application of Engineering analysis, simulation assembly and CNC programming work and so on. In order to solve the problems of the existing product data quality, such as validation custom trival, lack of high efficiency and flexibility, etc. The validation method of product data quality (PDQ) based on class was proposed in NX software environment, the representation of validation rules classes of product data quality, validation rules customization and implementation of validation process were introduced in detail in this study. Finally, an application case was employed to verify the practicability and effectiveness of the proposed method.

Abstract: In view of the deficiency of the standard back-propagation algorithm based on steepest descent method, a new kind of optimization strategy called invasive weed optimization (IWO) algorithm is introduced into the training process of feed-forward neural networks, and then a prediction model based on IWO feed-forward neural network (IWO-NN) is given. By the dynamic adjustment of standard deviation of the distribution of offspring individuals in IWO, the local convergence speed of networks is improved and the defect of trapping into a local optimum is reduced. By the empirical study of stock price prediction in Sany Heavy Industry, the results show that this method has better global astringency, robustness, and it is insensitive to initial values.

Abstract: Interval data is a range of continuous values which can describe the uncertainty. The traditional clustering methods ignore the structure information of intervals. And some modified ones have been developed. We have already used Taylor technique to perform well in the fuzzy c-means clustering algorithm. In this paper, we propose a new way based on the mixed interval slopes technique and interval computing. Experimental results also show the effectiveness of our approach.

Abstract: In view of the deficiency of the basic back-propagation (BP) algorithm based on steepest descent method. Bat algorithm (BA) in intelligent optimization is introduced into the training process of feed-forward neural networks, capturing the optimal solution of the objective function with a small population size and less number of iterations, and a prediction model based on BA feed-forward neural network (BA-NN) is given. By the empirical study of stock price prediction in Sany Heavy Industry, the results show that this method has advantages of frequency tuning and dynamic control of exploration and exploitation by automatic switching to intensive exploitation if necessary.

Abstract: A vehicle plate location algorithm under complex scenes is presented that based on interest region extraction and morphology. First of all, extract the edge of the lane, obtain the driveway area from the edge of the road, confirm the region of interest with the experiences in the driveway area, it can reduce the scope for searching. Then preprocessing is used in the vehicle image, including gray, edge detection and binary transformation is used. Then a series of morphological operations are used to look for candidate regions that probably contain the characters in range of sizes. Finally, the vehicle license plate can be found according to the median filter.