In this paper, we propose a novel method to implement fast detection of Common Visual Pattern (CVP). The purpose of CVP detection is to find the correspondences between the common visual regions of two given partial d...
详细信息
In this paper, we propose a novel method to implement fast detection of Common Visual Pattern (CVP). The purpose of CVP detection is to find the correspondences between the common visual regions of two given partial duplicate images. There are two major components of the proposed method which guarantee the good performance. First, we establish the Radiate-Geometric-Model (RGM). The RGM is represented by a set of radiate structures, and each structure is geometrically made up of a group of matched feature pairs. By utilizing the statistical information gained from the radiate structures, the RGM can not only quickly estimate the potential pairs of common regions but also organize the scale relationship between matched pairs into a compact form, hence increase the detection speed substantially. Second, we formulize the Radiate-Geometric-Model (RGM) into a graph optimization problem which could be solved by the method of graph-shift, thus make our algorithm capable of detecting the CVPs of all kinds of correspondences. Experimental results prove that the speed of our algorithm is at least 40 times faster than the state-of-the-art, while achieving a better detection performance at the same time.
Based on analysis of basic cubic spline interpolation, the clamped cubic spline interpolation is generalized in this paper. The methods are presented on the condition that the first derivative and second derivative of...
详细信息
As an important component of data mining, Cluster Analysis (CA) has being attached importance to artificial intelligence, machine learning and other fields. Traditional clustering methods have been studied for a relat...
详细信息
As an important component of data mining, Cluster Analysis (CA) has being attached importance to artificial intelligence, machine learning and other fields. Traditional clustering methods have been studied for a relatively long time;their technologies are mature and consequently they are well-applied. However, they are insufficient in clustering accuracy, noise sensitivity, along with effect on mass of data and non-convex clustering. Granular computing, which is regarded as a label of theories, methodologies, techniques, and tools, is an emerging conceptual and computing par information processing. It plays an important role information processing for fuzzy, uncertainty, partial truth and soft computing and is one of the main study stream in A.I. This paper introduces some new clustering methods, such as Fuzzy clustering, Clustering Algorithm Based on Rough Set, and clustering algorithm based on quotient space theory, emphatically expounds the basic thought and typical algorithms of these methods, and comparative analysis is carried out among these methods. Finally, new clustering algorithms are prospected and we put forward the value of research direction.
This paper systematically studies the problem of decision rule acquisition in inconsistent incomplete decision systems (IIDSs). First, a tolerance granular framework model based on tolerance granular computing is pres...
详细信息
This paper systematically studies the problem of decision rule acquisition in inconsistent incomplete decision systems (IIDSs). First, a tolerance granular framework model based on tolerance granular computing is presented, which is suitable for variety types of decision rules in IIDSs; secondly, with the proposed model, a framework for acquiring all minimum decision rule sets for each type is given, which solves the problem of decision rule acquisition in IIDSs to a certain degree; finally, an example is given to show the efficiency of our framework.
Based on analysis of cubic spline interpolation, the differentiation formulas of the cubic spline interpolation on the three boundary conditions are put up forward in this paper. At last, this calculation method is il...
详细信息
The rough neural networks (RNNs) are the neural networks based on rough set and one kind of hot research in the artificial intelligence in recent years, which synthesize the advantage of rough set to process uncertain...
详细信息
In image/video processing software and hardware products, low complexity interpolation algorithms, such as cubic and splines methods, are commonly used. However, these methods tend to blur textures and produce jaggy e...
详细信息
In image/video processing software and hardware products, low complexity interpolation algorithms, such as cubic and splines methods, are commonly used. However, these methods tend to blur textures and produce jaggy effect compared with other adaptive methods such as NEDI, SAI. Tanner graph based image interpolation algorithm has better effect in dealing with edge and texture, but with high computation complexity. Thanks to the high performance parallel processing capability of today's GPU, use of complex algorithms for real time application is becoming possible. In this paper, we present a fast algorithm for tanner graph based image interpolation and it's implementation on GPU. In our algorithm, the image model training process of tanner graph based image interpolation is greatly simplified. Experimental results show that the GPU implementation can be more than 47 times as fast as the CPU implementation.
The grey system forecasting model, neural network forecasting model and support vector machine forecasting model are proposed in this paper. Taking the road goods traffic volume from year of 1996 to 2003 in the whole ...
详细信息
In this paper, a new visual saliency detection method is proposed based on the spatially weighted dissimilarity. We measured the saliency by integrating three elements as follows: the dissimilarities between image pat...
详细信息
The bottleneck problem has emerged in feature selection when processing high-dimension and large-scale data, so in the past decade, the researches on feature selection have not adhere to the traditional algorithms and...
详细信息
The bottleneck problem has emerged in feature selection when processing high-dimension and large-scale data, so in the past decade, the researches on feature selection have not adhere to the traditional algorithms and ideas, showing a new trend of combining many new mathematical tools, which opens new space for feature selection applied in pattern recognition and makes further development in knowledge discovery and data mining. Granular computing has begun to take shape and show effect as a new idea of intelligent information processing, which creates the conditions for feature selection applied in data. The paper describes a new feature selection algorithm, basing on granular computing and making rough set approximation as background, the algorithm generates the granules, using a tolerance function, distinguishes noise data and inconsistent data, to achieve feature selection in the information table, and be effective for large-scale data sets.
暂无评论