Advanced Materials Research Vols. 179-180

Abstract: Association rule mining is one of the hottest research areas that investigate the automatic extraction of previously unknown patterns or rules from large amounts of data. Finding association rules can be derived based on mining large frequent candidate sets. Aiming at the poor efficiency of the classical Apriori algorithm which frequently scans the business database, studying the existing association rules mining algorithms, we proposed a new algorithm of association rules mining based on relation matrix. Theoretical analysis and experimental results show that the proposed algorithm is efficient and practical.

Abstract: We diagnose the Electronic Gasoline failure according to the improved CPN neural network. In the normal CPN neural network, it allows one neuron to win. We improved the CPN neural network based on the developed CPN and allow two neurons to win in the competition layer. The two winning neurons affect the Weights simultaneously. This improved method can adjust the weight values in the CPN model more accurately and give more accurate output for testing data.

Abstract: Ordinary differential equation with two small parameters was considered. Since the presence of two small parameters, the solution of the problem will change rapidly near both sides of the boundary layer. Firstly, the equation was decomposed into several equations in order to have fourth order asymptotic decomposition. The asymptotic properties of all these equations were discussed. Secondly, high order numerical methods were constructed for left side and right side singular component. Thirdly, a class of high numerical methods were presented when the special case. Finally, the error estimations for all these numerical methods were given.

Abstract: A simple process to fabricate porous polyimide membranes (PPMs) and effect of warm-up temperature rates on morphologies of the PPMs were reported in this paper. The polymer solution layer consisting of the corresponding polyamic acid (PAA), solvent and the pore forming agents with high boiling temperature or high decomposition temperature was first treated under a lower temperature (about 150°C). The received solid membrane was further imidized by a higher temperature (about 270°C) and the pore forming agents were removed from the membrane at a temperature above their boiling temperature or decomposition temperature at last. Then a PPM was obtained. As the temperature was elevated gradually, the porosity of the received PAA membrane was lower than that of the PPM treated by a faster warm-up temperature rate.

Abstract: Microscopic and macroscopic deformation behavior of ultra-high molecular weight polyethylene (UHMWPE) under the drawing force was studied in this paper. The scanning electron microscopy (SEM) experimental results indicated that: under the force, non-crystalline materials distorted, crystal faces were torn along grain boundaries and crystal grains were disrupted. So crystal size became smaller, more grain boundaries and more voids were formed, which supplied more free space for the movement of crystal grains and chain segments. The torsion test results showed that the UHMWPE exhibited excellent flexibility and the specimen had more excellent deformation ability at a slower loading rate.

Abstract: Based on the fundamental test data and the mesoscopic approach, a rigid body spring model is developed for simulation of the behavior of large aggregate concrete subjected to uniaxial or multiaxial load. Firstly, the concrete is treated as a three-phase inhomogeneous system; random aggregate model for fully-graded concrete is used to form the aggregate distribution. Then, based on the rigid body spring discrete element model, a procedure for mesoscopic study behaviour of large aggregate concrete under the two-dimensional stress state is given. At last, the comparison of numerical and experimental results shows that this method could effectively describe the failure behavior of large aggregate concrete under various plane stress state.

Abstract: In this paper, Short-term bias stability of solid state vibration beam gyroscope is studied. Based on the dynamical equation of solid vibration beam gyroscope, the elements effecting the zero stability which effects the short term bias stability was discussed. The computed result shows that the main factors for short-term bias stability is transient solution error. The output signal has been filtered by the low pass filter, according to the characteristic of transient solution error. Simulation results demonstrate that low-pass filter can effectively suppress the error of output signal. In addition, the signal of solid state vibrating gyroscope filtered by low-pass filter agrees well with the simulation result. From the simulation and experiment data, the validity of the model and the feasibility of the filtering programs has been proved.

Abstract: The existence of holes in vehicle object makes it difficult to extract the feature. existing hole-filling method are capable of filling holes on small and smooth regions of a object, for large holes whit complex boundaries or in curved region in vehicle object, they may not obtain the satisfactory results. For resolving this problem, a novel hole-filling algorithm based on horizontal scan line was proposed to fill arbitrary holes in vehicle object obtained from vehicle image. Experiments show that the algorithm has good performance and efficiency, and in the filling processing, the feature of the area and perimeter of an object can be obtained at the same time.

Abstract: The traditional Canny algorithm can only seek the finite difference from horizontal and vertical directions. For the problem which exists in it, we improved the algorithm to seek the finite difference from horizontal direction, vertical direction, 45 degree direction and 135 degree direction. So the edge detection has got improvement in the effect that detail keeps, edge articulation and edge continuity, and it wiped off some false edge. The experiment has verified the validity and correctness of the improved algorithm.

Abstract: A quasi-blind adaptive video watermarking algorithm in uncompressed video wavelet domain based on human visual system is proposed in this paper. This algorithm established a visual model according to the human visual masking and the characteristics of wavelet coefficients. It did not require the scene segmentation of the video, choosing the video frames which used for embedding watermarking by the key, and embedding different watermarking which processed by Arnold scrambling into different frame, so it is robust to statistical analysis, frame crop and so on. To ensure the spatial synchrony when extracting the watermark information, zero-watermarking method and chaotic system are used to generate the synchronization information. The quantized central limit theorem is applied to adjust the low frequency coefficients, it makes the extracted watermark information keeps invariant when its element value changed in robust region. A new correlation detection method of watermarking information was put forward. Watermark detection does not require original video. It realized the quasi-blind watermark detection.