【摘 要】
：
This paper proposes a general framework for the problem of structured sparsity via half-quadratic(HQ)minimization.Based on the theory of convex conjugacy,we firstly define an \(l_{2,1}^\varepsilon \)-
【出 处】
：
第十一届图像图形技术与应用学术会议(IGTA2016)
论文部分内容阅读
This paper proposes a general framework for the problem of structured sparsity via half-quadratic(HQ)minimization.Based on the theory of convex conjugacy,we firstly define an \(l_{2,1}^\varepsilon \)-norm and induce a family of penalty functions for structured sparsity.Then we build and discuss some important properties of these functions.By introducing the multiplicative auxiliary variable in HQ,we further reformulate the structured sparsity problem as an augmented half-quadratic optimization problem,and propose a general iteratively reweighted framework to alternately minimize the cost function.The proposed framework can be used in sparse representation,group sparse representation and multi-task joint sparse representation.Finally,in terms of the task of multi-biometric information fusion,we apply our proposed methods to obtain a novel fusion strategy,named structured fusion.Experimental results on the multi-biometric problems corroborate our claims and validate the proposed methods.
其他文献
In this paper,we put forward a kind of visual landing algorithm which is straightforward and undemanding to realize for the special environment of deck.Firstly,the algorithm calculates roll angle acco
Temperature field measurement technology based on color charge-coupled device(CCD)image sensor is a new temperature measuring method with advantages of all the non-contact temperature measuring method
Hand detection is a challenging task in hand gesture recognition system and the detection results can be easily affected by changes in hand shapes,viewpoints,lightings or complex backgrounds.In order
Numerous image content authentication schemes have been proposed to solve potential security problems in transmitting digital images via the Internet.Aiming at achieving high detection successful rate
This paper presents a novel point set registration algorithm based on iterative linear optimization,which can be used to register both rigid and non-rigid point set.Firstly,a new cost function was con
This paper presents an improved pose estimation algorithm for vision-based space objects.The major weakness of most existing methods is limited convergence radius.In most cases they ignore the influen
This paper presents a novel random selective image encryption scheme.In the scheme,image is firstly permutated by position and high four-bits of pixel; then the permutated image is diffused by adaptiv
Irregular low-density parity-check(LDPC)codes can provide an unequal error protection(UEP)capability naturally by unequal degree distribution.In this paper we propose a new architecture of UEP-LDPC en
A proposed method was introduced to study the meso-pore structures of eco-material in this paper.Firstly,the samples were solidified by using the mixtures of epoxy resin,triethanolamine and blue organ
This paper presents an automatic grayscale image colorization method using convolutional neural network.Besides the gray target image,the user doesnt need to provide a reference color image nor manual