Graph-based representation of images is becoming a popular tool since it represents in a compact way the structure of a scene to be analyzed and allows for an easy manipulation of sub-parts or of relationships between...
详细信息
ISBN:
(数字)9783709164877
ISBN:
(纸本)9783211831212
Graph-based representation of images is becoming a popular tool since it represents in a compact way the structure of a scene to be analyzed and allows for an easy manipulation of sub-parts or of relationships between parts. Therefore, it is widely used to control the different levels from segmentation to interpretation.
The 14 papers in this volume are grouped in the following subject areas: hypergraphs, recognition and detection, matching, segmentation, implementation problems, representation.
Existing approaches for automatic image annotation usually suffer from two issues: (1) lacking a good quality distance metric for image semantic similarity measure; (2) rarely considering the correlation between label...
详细信息
Existing approaches for automatic image annotation usually suffer from two issues: (1) lacking a good quality distance metric for image semantic similarity measure; (2) rarely considering the correlation between labels assigned to each image. In this paper, we aim to resolve both of the problems simultaneously in a novel unified framework. Specifically, a proper distance metric is learned based on the structural SVM in a discriminative manner, which can optimize the ranking of the images induced by distances from a test image. Subsequently, a collaborative label propagation algorithm is leveraged to model the correlation between class labels in an explicit manner. Also, the learned metric is embedded in the propagation model. The integration of the two components leads to more accurate annotation results. The experiments conducted on the Corel dataset demonstrate the effectiveness of the proposed unified framework.
Deblurring camera-based document image is an important task in digital document processing, since it can improve both the accuracy of optical character recognition systems and the visual quality of document images. Tr...
详细信息
With the unprecedented growing of city economy and the crying need for public transportation,the ease of traffic related problems such as traffic jams has been a hard nut to crack everywhere. In this paper,a new metho...
详细信息
ISBN:
(纸本)9781424463473
With the unprecedented growing of city economy and the crying need for public transportation,the ease of traffic related problems such as traffic jams has been a hard nut to crack everywhere. In this paper,a new method based on convexity and concavity to better position the roads and streets of urban areas in order for further GPS/GIS integrated system,which is used for traffic flow analysis,traffic forecast,traffic induction and car positioning and tracking etc,is proposed. Compared with our previous work,this method can yield better results,and the accuracy that we get can be comparable to Google map. Our experiments are based on the data of the city of Shanghai and the results are very satisfying.
In recent years, iris recognition is becoming a very active topic in both research and practical applications. However, fake iris is a potential threat there are potential threats for iris-based systems. This paper pr...
详细信息
A new algorithm for linear instantaneous independent component analysis is proposed based on maximizing the log-likelihood contrast function which can be changed into a gradient *** iterative method is introduced to s...
详细信息
A new algorithm for linear instantaneous independent component analysis is proposed based on maximizing the log-likelihood contrast function which can be changed into a gradient *** iterative method is introduced to solve this equation *** unknown probability density functions as well as their first and second derivatives in the gradient equation are estimated by kernel density *** simulations on artificially generated signals and gray scale natural scene images confirm the efficiency and accuracy of the proposed algorithm.
Diabetic retinopathy (DR) has already been one of the leading causes of vision loss. A large number of researches about deep learning-based DR screening using color retinal photography images have been proposed in rec...
详细信息
In this paper, we present a novel method to upsample the depth map obtained by the Time-of-Flight (ToF) camera with the guidance of the companion high resolution color image. The problem is modeled with an optimizatio...
详细信息
ISBN:
(纸本)9781479983407
In this paper, we present a novel method to upsample the depth map obtained by the Time-of-Flight (ToF) camera with the guidance of the companion high resolution color image. The problem is modeled with an optimization framework where we use a novel exponential function as the error norm. By using this novel error norm, our model could take the properties of the depth map itself into account. Depth discontinuity cues are obtained not only from the color image but also the depth map itself. To further enhance the performance, we perform a data driven selection of the parameter in the model to better fit the property of the depth map. Experimental results show that our method has excellent performance in smoothing the noise, preserving sharp depth discontinuities and suppressing the texture copy effect.
Scale is a major concept in many sciences concerned with human activities and physical processes occurring in the world, and directly related to many investigations of spatial objects, including the procedure of spati...
详细信息
Scale is a major concept in many sciences concerned with human activities and physical processes occurring in the world, and directly related to many investigations of spatial objects, including the procedure of spatial data mining. In this paper, we attempt to apply the spatial data mining to the field of coal mining, and the technical notion is to generate patterns or rules by means of different scale databases that depict the same subject. The whole research procedure gives readers an understanding of how processes operate at different scales and how they can be linked across scales. At the same time, our study actually presents a new method of image mining as well.
Concept generalization under incomplete domain theory is a very important research aspect in artificial intelligence. Current multilayer perceptron and EBL (explanation based learning) approaches cannot deal with it e...
详细信息
ISBN:
(纸本)0780342534
Concept generalization under incomplete domain theory is a very important research aspect in artificial intelligence. Current multilayer perceptron and EBL (explanation based learning) approaches cannot deal with it effectively. We present a new method, hybrid multilayer perceptron/EBL approach for concept generalization, which can deal with concept generalization more effectively.
暂无评论