International Journal of Pattern Recognition and Artificial Intelligence

Since the concept of frequent pattern mining was proposed, there have been many efforts to obtain useful pattern information from large databases. As one of them, applying weight conditions allows us to mine weighted frequent patterns considering unique importance of each item composing databases, and the result of analysis for the patterns provides more useful information than that of considering only frequency or support information. However, although this approach gives us more meaningful pattern information, the number of patterns found from large databases is extremely large in general; therefore, analyzing all of them may become inefficient and hard work. Thus, it is essential to apply a method that can selectively extract representative patterns from the enormous ones. Moreover, in the real-world applications, unexpected errors such as noise may occur, which can have a negative effect on the values of databases. Although the changes by the error are quite small, the characteristics of generated patterns can be turned definitely. For this reason, we propose a novel algorithm that can solve the above problems, called AWMax (an algorithm for mining Approximate weighted maximal frequent patterns (AWMFPs) considering error tolerance). Through the algorithm, we can obtain useful AWMFPs regardless of noise because of the consideration of error tolerance. Comprehensive performance experiments present that the proposed algorithm has more outstanding performance than previous state-of-the-art ones.

Multi-label classification is to assign an instance to multiple classes. Naive Bayes (NB) is one of the most popular algorithms for pattern recognition and classification. It has a high performance in single label classification. It is naturally extended for multi-label classification under the assumption of label independence. As we know, NB is based on a simple but unrealistic assumption that attributes are conditionally independent given the class. Therefore, a double weighted NB (DWNB) is proposed to demonstrate the influences of predicting different labels based on different attributes. Our DWNB utilizes the niching cultural algorithm (NLA) to determine the weight configuration automatically. Our experimental results show that our proposed DWNB outperforms NB and its extensions significantly in multi-label classification.

This paper proposes a new approach of features extraction based on structural and statistical techniques for handwritten, printed and isolated numeral recognition. The structural technique is inspired from the Freeman code, it consists first of contour detection and closing it by morphological operators. After that, the Freeman code was applied by extending its directions to 24-connectivity instead of 8-connectivity. Then, this technique is combined with the statistical method profile projection to determine the attribute vector of the particular numeral. Numeral recognition is carried out in this work through k-nearest neighbors and fuzzy min-max classification. The recognition rate obtained by the proposed system is improved indicating that the numeral extracted features contain more details.

Finger vein pattern, as a promising hand-based biometric technology, has been well studied in recent years. In this paper, a new superpixel-based finger vein recognition method is presented. In the proposed method, we develop two types of effective superpixels, i.e. stable superpixel and discriminative superpixel to represent finger vein image and these superpixels are expected to play different roles in matching stage. In detail, the stable and discriminative superpixels are firstly learned from the training images for each enrolled class. When verifying a testing image, we just compare the superpixels at the same location as the two types of superpixels in template. Then, the two types of superpixels are combined utilizing a reversible weight-based fusion method in score level. Additionally, to further improve the recognition performance, we explore the superpixel context feature (SPCF). For each superpixel the SPCF is obtained by comparing the current superpixel with its surrounding neighbors. In the final matching stage, we integrate the matching score of two types of superpixels and it of the SPCF using the weighted SUM fusion method. The experimental results on two open finger vein databases, i.e. PolyU and SDUMLA-FV, show that our method not only performs better than the existing superpixel-based method, but also has advantages in comparison with some traditional ones.

In this paper, we propose an improved version of the neighbor embedding super-resolution (SR) algorithm proposed by Chang et al. [Super-resolution through neighbor embedding, in Proc. 2004 IEEE Computer Society Conf. Computer Vision and Pattern Recognition(CVPR), Vol. 1 (2004), pp. 275–282]. The neighbor embedding SR algorithm requires intensive computational time when finding the K nearest neighbors for the input patch in a huge set of training samples. We tackle this problem by clustering the training sample into a number of clusters, with which we first find for the input patch the nearest cluster center, and then find the K nearest neighbors in the corresponding cluster. In contrast to Chang’s method, which uses Euclidean distance to find the K nearest neighbors of a low-resolution patch, we define a similarity function and use that to find the K most similar neighbors of a low-resolution patch. We then use local linear embedding (LLE) [S. T. Roweis and L. K. Saul, Nonlinear dimensionality reduction by locally linear embedding, Science 290(5500) (2000) 2323–2326] to find optimal coefficients, with which the linear combination of the K most similar neighbors best approaches the input patch. These coefficients are then used to form a linear combination of the K high-frequency patches corresponding to the K respective low-resolution patches (or the K most similar neighbors). The resulting high-frequency patch is then added to the enlarged (or up-sampled) version of the input patch. Experimental results show that the proposed clustering scheme efficiently reduces computational time without significantly affecting the performance.

Accurate and fast target image recognition is an important function of applications such as remote sensing imaging and medical imaging. However, an operator such as speeded up robust feature (SURF) cannot be accurately matched in the recognition process of a target image. This led us to propose the use of a method capable of matching identification, i.e. binary robust invariant scalable keypoints (BRISK) operators, in combination with SURF operators. The proposed algorithm combines the accuracy of SURF operators and the rapidity of BRISK operators to obtain a quick and accurate way of matching. The initial matching of image feature extraction for targets is performed using the SURF-BRISK algorithm, and similarity measurements of feature matching are performed for the feature points of initial matching using the Hamming distance. Then, secondary fine matching is performed using the M-estimator Sample and Consensus (MSAC) algorithm to eliminate mismatched point pairs in order to achieve recognition of target images. Then, the three-dimensional coordinates of the work piece are obtained by using a binocular stereo vision system to provide location coordinates for the robots to grasp the work pieces accurately. In the experiment, stereo vision matching is conducted for targets obtained using the SURF-BRISK algorithm, and the location coordinates of targets are passed to the robot controller. The experimental results show that if the special geometric distortion is neglected, this method can be adapted for accurate positioning of the target; hence, it can identify the target in complex environments, access the location coordinates of the target, and achieve accurate robotic grasping of the work piece in real time.

A human action recognition system based on image depth is proposed in this paper. Depth information features are not easily disturbed by noise; and due to this characteristic, the system can quickly extract foreground targets. Moreover, the target data, namely, depth and two-dimensional (2D) data, are projected to three orthogonal planes. In this manner, the action featured in the depth motion along the optical axis can clearly describe the trajectory. Based on the change of motion energy and the angle variations of motion orientations, the temporal segmentation (TS) method automatically segments the complex action into several simple movements. Three-dimensional (3D) data is further applied to acquire the three-viewpoint (3V) motion history trajectory, whereby a target’s motion is described through the motion history images (MHIs) from the 3Vs. The weightings corresponding to the gradients of the MHIs are included for determining the viewpoint that bests describe the target’s motion. In terms of feature extraction, the application of multi-resolution motion history histograms can effectively reduce the computational load and achieve a high recognition rate. Experimental results demonstrate that the proposed method can effectively solve the self-occlusion problem.

The problem of discovering influential users is important to understand and analyze online social networks. The user profiles and interactions between users are significant features to evaluate the user influence. As these features are heterogeneous, it is challengeable to take all of them into a proper model for influence evaluation. In this paper, we propose a model based on personal user features and the adjacent factor to discover influential users in online social networks. Through taking the advantages of Bayesian network and chain principle of PageRank algorithm, the features of the user profiles and interactions are integratedly considered in our model. Based on real data from Sina Weibo data and multiple evaluation metrics of retweet count, tweet count, follower count, etc., the experimental results show that influential users identified by our model are more powerful than the ones identified by single indicator methods and PageRank-based methods.

In view of the exponential growth of information generated by social networks, social network analysis and recommendation have become important for many web applications. This paper examines the problem of social collaborative filtering to recommend items of interest to users in a social network setting. Many social networks capture the relationships among the nodes by using trust scores to label the edges. The bias of a node denotes its propensity to trust/mistrust its neighbors and is closely related to truthfulness. It is based on the idea that the recommendation of a highly biased node should be removed. In this paper, we propose a model-based approach for recommendation employing matrix factorization after removing the bias nodes from each link, which naturally fuses the users’ tastes and their trusted friends’ favors together. The empirical analysis on real large datasets demonstrate that our approaches outperform other state-of-the-art methods.

Non-Chinese speakers hold increasing opportunities and need to process Chinese information and communicate in Chinese. This paper, with the purpose of facilitating the handwriting input of Chinese characters for non-Chinese speakers, is directed towards the development of the handwriting rules and vocabulary for Latin-style anti-cursive characters and the ways of their selection and classification. This aims to build a practical platform by utilizing three characteristics of wavelet neural network — automatically ascertaining the number of hidden layer unit, converging rapidly and never running into the partial minimum of networks — for a simple Latin-style online handwriting input and processing, meanwhile, taking the customary handwriting habits of non-Chinese speakers. The paper, based on profound information of cursive characters, deciphered the genetic code of ancient cursive symbols and made clear the rules for characters changing into its cursive style. As a result, it breaks the bottleneck, which enables non-Chinese speakers to easily input information through handwriting Chinese characters.

The Suffix Array (SA) is a fundamental data structure which is widely used in the applications such as string matching, text index and computation biology, etc. How to sort the suffixes of a string in lexicographical order is a primary problem in constructing SAs, and one of the widely used suffix sorting algorithms is qsufsort. However, qsufsort suffers one critical limitation that the order of suffixes starting with the same $2^k$ characters cannot be determined in the kth round. To this point, in our paper, an efficient suffix sorting algorithm called dsufsort is proposed by overcoming the drawback of the qsufsort algorithm. In particular, our proposal maintains the depth of each unsorted portion of SA, and sorts the suffixes based on the depth in each round. By this means, some suffixes that cannot be sorted by qsufsort in each round can be sorted now, as a result, more sorting results in current round can be utilized by the latter rounds and the total number of sorting rounds will be reduced, which means dsufsort is more efficient than qsufsort. The experimental results show the effectiveness of the proposed algorithm, especially for the text with high repetitions.