Earticle

Computation of extent of image visual excellence is of essential importance for many image and video processing appliances, where the objective of quality evaluation algorithms is to automatically evaluate the excellence of images. This paper is the detailed experimental study, classification, analysis and comparison of the subjective non-blind image quality measures. After analysis, evaluation and comparison, these schemes are classified into two groups on the basis of similarity and dissimilarity check. It also scrutinizes the statistical recital of all the quality measures.

Batik has important aesthetic and cultural values, which records and expresses national culture. Their basic graphical pattern comes from imitating object in natural environment that lots of patterns have self-similar. In this paper, we first analyze self-similar characteristic of butterfly pattern Batik and propose an auto-generation algorithm of butterfly pattern based on fractal geometry. The auto-generation algorithm adopts two-dimensional iterated function system (IFS) to construct fractal. First, it generates basis butterfly shape element by functions and it then repeatable iterates the functions until forming a butterfly pattern. Second, it generates butterfly beard to obtain an entire butterfly pattern by functions. Finally, it uses function to iterate the entire butterfly pattern to generate various pattern layouts. The experimental result shows that the algorithm can automatically generate various and beautiful butterfly fractal pattern. Based on the experimental result, we analyze change regularity between generation patterns and function parameters to obtain the most effective range of various function parameter values, which can perfect the proposed algorithm to quickly generate effective patterns. It realizes digital design of butterfly pattern and enriches pattern of Batik.

Speech is the most natural communication mode for human beings. The task of speech recognition is to convert speech into a sequence of words by a computer program. Speech recognition applications enable people to use speech as another input mode to interact with applications with ease and effectively. Speech recognition interfaces in native language will enable the illiterate/semi-literate people to use the technology to greater extent without the knowledge of operating with computer keyboard or stylus. For more than three decades, a great amount of research was carried out on various aspects of speech recognition and its applications. Today many products have been developed that successfully utilize automatic speech recognition for communication between human and machines. Performance of speech recognition applications deteriorates in the presence of reverberation and even low levels of ambient noise. Robustness to noise, reverberation and characteristics of the transducer is still an unsolved problem that makes the research in the area of speech recognition still very active. A detailed study on automatic speech recognition is carried out and presented in this paper that covers the architecture, speech parameterization, methodologies, characteristics, issues, databases, tools and applications.

In view of the problem that the joint type multiple DOF Manipulator is easy to collide with the surrounding equipment and jitter caused by the joint movement, this paper based on the stamping manipulator of five DOF cylindrical coordinate type, planned a straight path through actual stamping process in Cartesian space and quadratic programmed by using a bounded deviation theory. Then, this paper respectively used cubic polynomial, five polynomial and B-spline interpolation on interpolation points in joint space, and analyzed the displacement, velocity and acceleration of the end reference point. Finally, model in virtual prototype and real experiments were verified to ensure the safety and precision of the path. The results meet the requirements of the stamping process, succeeded in raising the stamping accuracy, and ensured the quality of stamping, and can arrange the entire auto production line equipment. So, it has important theoretical and practical significance to improve the stamping process.

Mutual information based medical image registration is one of the most common registration method in registration domain currently. This paper studies the mutual information of the calculation process for mutual information calculation speed is slow, low efficiency of registration, registration can not meet real-time requirements, an improved gray compression algorithm, by reducing the gray-scale of images to improve image the registration efficiency and gradient similarity combined with mutual information as a new registration measure, so that the registration in improving efficiency while meeting the requirement of registration accuracy. The comparative experiments this article made show that, under the premise of registration accuracy to ensure that the method can significantly improve the speed of image registration.

With the explosive growth of internet and the fast communication techniques available in recent years, the security and confidentiality of the sensitive data has become of supreme importance and concern. In order to protect this data from tampering and unauthorized access various methods for data hiding in cryptography and steganography have been developed and are in practice today. Steganography hides messages inside a carrier media. Cryptography, on the other hand obscures the content of the message. Visual cryptography, a Secret sharing Scheme, is a new technique which provides information security on digital information by distributing the secret information into many shares. The approach is simple unlike the complex, computationally intensive techniques used in traditional cryptography. Decryption can be performed by simply stacking the shares together and by interpreting it using the Human Visual System (HVS) without the involvement of any computational machines. An attempt has been made to extend the simple (2-2) secret sharing scheme based on Random Grids Visual Secret Sharing Scheme for authentication into a cuboid representation where each share represents a face of the cube. The representation also increases the carrying capacity of a system. Each face (share) of a cube is fragmented into a collection of randomly scrambled sub-blocks. The respective blocks are embedded into a carrier image to hide its existence by using a stenographic technique. The final information is revealed by overlapping the two shadows/shares containing the sub-blocks with the correct index.

In this paper, for wavelet modulus maxima edge detection affected by noise largely, we propose an improved image fusion algorithm. Firstly, wavelet denoising and smoothing filter are conducted on the original image. Secondly, making edge detection with wavelet transform modulus maxima edge detection and improved canny edge detection get each of edge detection images. Thirdly, let the wavelet fusion follow some certain fusion rules. Finally, making the inverse wavelet transform reconstruct the fused image. The experimental results show that the fused image combines the advantages of edge detection methods, which is an effective image edge detection method.

Cloud services is a typical trait of cloud computing. It allows tenants rent customizable services transparently, and guarantees substantial cost savings for the providers. With the significant number of cloud services, a key challenge when providing customizable cloud services is how to manage those services and find an optimized configuration service rapidly that maximizes tenants’ preferences, subject to resource constraints. To help address this challenge, this paper introduces a novel approach to support the modeling and configuration of customizable cloud services. In particular, we propose a customizable service selecting and configuring optimization approach based on a heuristic approach, which can quickly derive an optimized valid service selection by evaluation different configurations that optimize tenants’ preferences under resource limitations. The scalability and performance of the algorithms is investigated in detail. And the obtained experimental results demonstrate the feasibility of the proposed method.