Applied Mechanics and Materials Vols. 321-324

Abstract: The first step of track surface detection based on contactless optical detection is to obtain the surface image. This paper introduces a vehicle-mounted image collecting system base on area-array camera. This system avoids image morphing and edge deformation of system base on linear-array camera.Camera could works in both frequently flash and persistent bright without influence image collecting and will not miss information of the rail track surface when vehicle keeps the speed of 342Km/h. The exposure time should be chosen according to railcar speed and maximum speed.

Abstract: We propose a DCT domain watermarking algorithm, this algorithm transforms the watermark image before the watermark embedding and embeds watermark images by key vector, which improves the robustness and security greatly. The experiments indicate that the program has a better robustness on Gaussian filtering, white noise, image shearing, so digital watermark technology is useful.

Abstract: Compressed sensing (CS) is a new developed theoretical framework for information acquisition and processing that breaks through the conventional Nyquist sampling limit. This paper proposes a sparse representation schemes based on principal component analysis (PCA) for CS that will be used for hyperspectral images compressed sampling. This scheme employs the prediction transform matrix to remove the correlations among successive hyperspectral measurement vectors. Experiment processes using the hyperspectral image from Earth Observing One (EO-1), and it shows a desired result both at reconstruction and denoising.

Abstract: This paper introduces the theory of photovoltaic battery and gives circuit model of the battery. Its Simulink simulation model and the output characteristic curve are achieved by the derivation of the circuit model. It also introduces the working principle of MPPT and different kinds of MPPT algorithm. The paper focuses on the implementation of the Conductance Increment Method , and carries on the simulation results.

Abstract: Purpose of this paper, coupled with Kinect controller, utilizes handwritten recognition for digit home interactive TV channel controlling. The results of this paper are divided into three parts. The first part presents the development of handwritten digit rules and operations environment using Principal curves method. The second part shows the skeleton tracking system for features extraction. The third part developes an accomplishment of digit recognition function through neural network. In this system, it enables users to operate video and finishs interactive operation in a more intuitive and friendly way, instead of the traditional button control to achieve interactive effect of human-machine interface.

Abstract: Battle Damage Assessment (BDA) is to assess the damage degree of enemy’s target after being attacked. In modern war, combat commanders always make decision on the basis of BDA. In this paper, an automatic method for assessing the damage extent of an attacked airport based on the image taken before and after a strike is described. Firstly, the airport blockade condition is analyzed and damage assessment criteria of airport are proposed. Secondly, three steps of the image information pretreatment are carried out and a reliability analysis method of image information is proposed. Lastly, damage assessment result is calculated to verify the validity and availability of the proposed method.

Abstract: This paper explains and demonstrates how to apply the digital technique of sound wave contrast-into foreign oral teaching, using the software of media player and cool edit 2 to acquire the experimenters speech sound waves. The key is to offer the obvious sound wave contrast among the wave of the native speaker, the original and unmodified wave of the oral experimenter and the new wave of the oral experimenter after modifications and teachers instructions, which makes the oral autonomic learning more possible and more scientific. From long wave or short wave, wave trough or wave crest, smooth wave or sharp wave, it can be easily identified the deviations and the weaknesses of the experimenter from the standard during the daily practices so it is necessary to make corrections efficiently. Additionally, experimenter also can observe the improvements frequently, which means this experiment more instructive.

Abstract: The main approaches currently used for digital preservation of historical architecture heritage are 3D scanning. These approaches have successfully realized the acquirement of 3D shapes of architectural heritages. But due to some critical constrains, they still cannot provide enough support for recording high quality color and texture information so far. In the mean time, large amount of information of point cloud acquired by 3D scanning also create big challenges for data processing and digital content utilization. We adopted an entirely new ways and techniques in the process of digital preservation of ancient buildings in Foshan Ancestral Temple & Donghuali area, which employed High Dynamic Range Image (HDRI), panoramic imaging, photogrammetry measurement to record geometry information together with accurate color texture information of cultural heritage. In addition, some practical applications of digital content were also carried out, which shows a promising future of our methods.

Abstract: The way of kernel function has been widely applied in machine learning field, such as artificial neural network and support vector machine, for avoiding dimensional disaster of feature space and improving performance of learning machine effectively. The selection of kernel function and construction of new kernel are the core problems, which have a direct relation with the performance of classification, and the research of this field is not enough. In this paper support vector machine (SVM) was used as an example, and the performance of common kernel functions was evaluated through observing and computing the features of kernel matrix. Base on this, a new mixed kernel function was gotten by optimization of kernel functions, and the experimental data proved that the performance of SVM was improved by the mixed kernel function. If the weighting coefficient was selected properly, the correct rate could even reach to 100%. What’s more, not only a method to construct a new learning machine was given, but also a reference for selecting kernel function was given.

Abstract: This paper according to the video features using the method of evolution of video image classification processing, suitable for video performance and human visual characteristics of video watermarking, and put forward a kind of using artificial swarm evolution algorithm for video image classification processing of video watermarking algorithm, improve the watermark performance.