Sparse representation technique has been widely used in various areas of computer vision over the last decades. Unfortunately, in the current formulations, there are no explicit relationship between the learned dictio...
详细信息
ISBN:
(纸本)9781479906505
Sparse representation technique has been widely used in various areas of computer vision over the last decades. Unfortunately, in the current formulations, there are no explicit relationship between the learned dictionary and the original data. By tracing back and connecting sparse representation with the K-means algorithm, a novel variation scheme termed as self-explanatory convex sparse representation (SCSR) has been proposed in this paper. To be specific, the basis vectors of the dictionary are refined as convex combination of the data points. The atoms now would capture a notion of centroids similar to Kmeans, leading to enhanced interpretability. Sparse representation and K-means are thus unified under the same framework in this sense. Besides, an appealing property also emerges that the weight and code matrices both tend to be naturally sparse without additional constraints. Compared with the standard formulations, SCSR is easier to be extended into the kernel space. To solve the corresponding sparse coding subproblem and dictionary learning subproblem, block-wise coordinate descent and Lagrange multipliers are proposed accordingly. To validate the proposed algorithm, it is implemented in image classification, a successful applications of sparse representation. Experimental results on several benchmark data sets, such as UIUC-Sports, Scene 15, and Caltech-256 demonstrate the effectiveness of our proposed algorithm.
This paper presents an overview of our published work on physical principles, applications, and advances in integral imaging and digital holography. Various approaches for image capture, image reconstruction, and 3D d...
详细信息
Several recent proposals, namely Software Defined Networks (SDN), Network Functions Virtualization (NFV) and Network Service Chaining (NSC), aim to transform the network into a programmable platform, focusing respecti...
详细信息
Several recent proposals, namely Software Defined Networks (SDN), Network Functions Virtualization (NFV) and Network Service Chaining (NSC), aim to transform the network into a programmable platform, focusing respectively on the control plane (SDN) and on the data plane (NFV/NSC). This paper sits on the same line of the NFV/NSC proposals but with a more long-term horizon, and it presents its considerations on some controversial aspects that arise when considering the programmability of the data plane. Particularly, this paper discusses the relevance of data plane vs control plane services, the importance of the hardware platform, and the necessity to standardize northbound and southbound interfaces in future software-defined data plane services.
A novel sparse representation based multi-source localization method is presented in this work. We envision a wireless network infrastructure containing multiple phase arrays of acoustic sensors. With multiple arrays,...
详细信息
A novel sparse representation based multi-source localization method is presented in this work. We envision a wireless network infrastructure containing multiple phase arrays of acoustic sensors. With multiple arrays, direct estimation of a set of source locations is achieved using a new joint sparse representation of array covariance matrices (JSRACM). This representation transforms the source location estimation problem into a spatial sparse signal representation (SSSR) optimization problem. To mitigate the high computation complexity of JSRACM, a novel binary sparse indicative vector (SIV) is introduced to represent the support of joint SSSR of array covariance matrices. As such, the multiple source locations may be estimated by solving an unconstrained optimization problem of the SIV vector using existing FOCUSS-like algorithms. The resulting SIVR-JSRACM algorithm does not require prior information of the number of sources nor initial source location estimates. It promises super-resolution, robustness to noise, and low computing complexity which is independent of the number of sensor phase arrays. Simulation results demonstrate superior performance of the proposed algorithm.
As an new optical instrument, Fiber Optical Gyroscope (FOG) is used in guidance and navigation system more and more. But the problem is that it can be disturbed by electromagnetic environment easily. According to appl...
详细信息
Memory processes are based on large cortical networks characterized by non-stationary properties and time scales which represent a limitation to the traditional connectivity estimation methods. The recent development ...
详细信息
ISBN:
(纸本)9781457702150
Memory processes are based on large cortical networks characterized by non-stationary properties and time scales which represent a limitation to the traditional connectivity estimation methods. The recent development of connectivity approaches able to consistently describe the temporal evolution of large dimension connectivity networks, in a fully multivariate way, represents a tool that can be used to extract novel information about the processes at the basis of memory functions. In this paper, we applied such advanced approach in combination with the use of state-of-the-art graph theory indexes, computed on the connectivity networks estimated from high density electroencephalographic (EEG) data recorded in a group of healthy adults during the Sternberg Task. The results show how this approach is able to return a characterization of the main phases of the investigated memory task which is also sensitive to the increased length of the numerical string to be memorized.
This paper evaluates the possibility of applying a geno-fuzzy control strategy to a magnetorheological semi-active damper for seismic vibration control. The proposed control starategy is designed and then tested and v...
详细信息
This paper presents the dynamic modeling and control simulation of a novel robot that combines flying motion and on ground motion into an integrated single robot. The ground motion is based on four wheels configuratio...
详细信息
ISBN:
(纸本)9781467355599
This paper presents the dynamic modeling and control simulation of a novel robot that combines flying motion and on ground motion into an integrated single robot. The ground motion is based on four wheels configuration that provides more stability. The flying motion is depending on the flying mechanism of quadrotor system. Smart transformation mechanism is developed to switch the robot from the ground motion configuration to the flying motion configuration and vice versa without adding any additional actuators. A manipulator with 3 DOF is added to handle an object during the ground motion and it is useful to hold this object during the flying motion. A CAD model is developed using SOLIDWORKS. The dynamic model of this robot is derived to achieve the eccentricity of the payload, the weight of the eccentric manipulator and managing the variation of the payload in the dynamic model. The derived robot dynamics are highly nonlinear. A controller is designed based on feedback linearization technique to stabilize the robot attitude and altitude. controlling the horizontal movements' nonholonomic constraints is used to generate the desired trajectories of robot attitudes. Another dynamic model and controller have been established for the transformation mechanism. Finally, the simulation results using MATLAB/SIMULINK show that the controller successfully vanish the eccentric effect and stabilize the robot attitude.
Forecasting the future time series values from the past values plays important role in daily life. There are some techniques support for time series analysis such as statistics, Neural Network (NN), Fuzzy Logic (FL) a...
详细信息
Hilbert-Huang Transform (HHT), proposed by N. E. Huang in 1998, is a novel algorithm for nonlinear and non-stationary signal processing. The key part of this method is decomposition the signal into finite number of In...
详细信息
暂无评论