Advanced Materials Research Vols. 756-759

Abstract: This paper converts a NURBS curve to piecewise rational Bézier curves by knot insertion algorithm, and then discusses the algorithm of continuous connection of NURBS curves. Meanwhile, explores the method to keep the same shape of the NURBS curves after connecting through the point translation and vector rotation theory. Finally, gives an instance to verify the validity of the algorithm.

Abstract: This paper investigates the problem of automated defect detection for textile fabrics and proposes a new optimal filter design method for solving this problem. Wavelet is used to extract the texture features of textile fabrics. In view of this optimal filter, a news semi-supervised segmentation algorithm is then proposed. The performance of the scheme is evaluated by using a variety of homogeneous textile images with different types of common defects. The test results exhibit accurate defect detection with low false alarm, thus confirming the robustness and effectiveness of the proposed scheme.

Abstract: The traditional LBP only considers the difference between the center pixel and the neighbor pixels. In order to use the relationship of the center pixel and neighbor pixels, a manipulative texture method B_LBP is proposed in this paper, which integrates the grayscale intensity of neighbor pixels into traditional LBP. The method keeps the local structure information of original images and enhances the identification ability. LDA (Linear discriminate analysis) is used to reduce the dimensionality of the original data. The experiments are conducted on Yale B and CMU PIE face databases with B_LBP, LEP and LBP. The results show that the B_LBP is superior to the traditional LBP and LEP.

Abstract: This paper proposes a blind source extraction (BSE) method on the basis of several time-delay autocorrelations from primary sources. This method is a second-order statistic learning algorithm, which can extract a desired biomedical signal while it shows a specific temporal structure. In contrast to conventional BSE methods, it is simple and do not need to choose any learning step size. Computer simulations on biomedical measurements demonstrate its validity and high performance in the process of revealing the underlying desired signal.

Abstract: The famous FastICA algorithm has been widely used for blind signal separation. For every process, it only converges to an original source which has the maximum negentropy of the underlying signals. To ensure the first output is the desired signal, we incorporate a priori knowledge as a constraint into the FastICA algorithm to construct a robust blind source extraction algorithm. One can extract the desired signal if its normalized kurtosis is known to lie in a specific range, whereas other unwanted signals do not belong to this range. Experimental results on biomedical signals illustrate the validity and reliability of the proposed method.

Abstract: A novel kernel based semi-supervised fuzzy clustering algorithm is proposed, and its iterative formula is given. This new algorithm can effectively improve the efficiency of the clustering algorithm. Combined with Fisher projection algorithm, two principal components are extracted from 7 hue statistics and 11 green value statistics, this new semi-supervised clustering method is applied to recognize the angular leaf spot disease of Bauhinia blakeana. The results showed that the consistent rate is 100% for the labeled leaves, and above 95% for other unlabeled leaves.

Abstract: Based on the research of the four kinds of algorithms of digital image segmentation, based on edge detection methods, based on region growing method, threshold segmentation method and digital image threshold segmentation method based on wavelet transform, using MATLAB simulation of all digital image enhancement and segmentation process, the obtained results are analyzed, proving the threshold segmentation wavelet transform method has unparalleled advantages in information extraction in medical image. Wavelet transform is a mathematical tool widely used in recent years, compared with the Fu Liye transform, the window of Fu Liye transform, wavelet transform is the local transform of space and frequency, it can be very effective in extracting information from the signal [[1.

Abstract: The influence on the high range resolution profile (HRRP) of the ballistic target caused by high speed motion is analyzed based on the wideband radar de-chirping echo. According to the characters of ballistic target echo ambiguity diagram (AD), the auto-terms line of echo ambiguity function is extracted from the corresponding binary image by skeleton extraction technology. Therefore, the problem of chirp rate estimation is transformed into the least square (LS) estimation problem. Furthermore, a new motion compensation method of HRRP for ballistic targets is proposed based on the ambiguity diagram and least square estimation theory. Simulation results show that the method can eliminate the influence of the high speed motion effectively, and it has a high estimate accuracy and low computational cost.

Abstract: With the rapid growth and wide application of Internet, everyday there are many of information generated and the existence of a large amount of information makes it hardly to mining the wanted information. The recommendation algorithm is the process to alleviative the problem. Collaborative filtering algorithm is one successful personalized recommendation technology, and is widely used in many fields. But traditional collaborative filtering algorithm has the problem of sparsity, which will influence the efficiency of prediction. In this paper, a collaborative filtering recommendation algorithm based on bipartite graph is proposed. The algorithm takes users, items and tags into account, and also studies the degree of tags which may affect the similarity of users. The collaborative filtering recommendation algorithm based on bipartite graph can alleviate the sparsity problem in the electronic commerce recommender systems.

Abstract: We present conditional random fields (CRFs), a framework for building probabilistic models to segment and label sequence data, and use CRFs to label pixels in an image. CRFs provide a discriminative framework to incorporate spatial dependencies in an image, which is more appropriate for classification tasks as opposed to a generative framework. In this paper we apply CRF to an image classification tasks: an image labeling problem (manmade vs. natural regions in the MSRC 21-object class datasets). Parameter learning is performed using contrastive divergence (CD) algorithm to maximize an approximation to the conditional likelihood. We focus on two aspects of the classification task: feature extraction and classifiers design. We present classification results on sample images from MSRC 21-object class datasets.