The development of convolutional neural network has brought great achievements to image classification in recent ***,the classification performance is good only for natural images rather than medical *** important rea...
详细信息
The development of convolutional neural network has brought great achievements to image classification in recent ***,the classification performance is good only for natural images rather than medical *** important reason is that the medical image database used for training is always *** how to use these limited data to acquire more extensive features has become a hot research *** this paper,we first update the order and number of the whole training data every time in active and incremental *** we set different contribution rate for the data selected in our model,which based on the information quantity of the data in training stage and make our model converge *** that,a pre-trained model and our preprocessed datasets are employed,which allows us to further fine-tune our *** experiments evaluated on two different biomedical datasets shows that our model can achieve promising results.
The atomic norm minimization (ANM) has been successfully incorporated into the two-dimensional (2-D) direction-of-arrival (DOA) estimation problem for super-resolution. However, its computational workload might be una...
详细信息
ISBN:
(纸本)9781538646595
The atomic norm minimization (ANM) has been successfully incorporated into the two-dimensional (2-D) direction-of-arrival (DOA) estimation problem for super-resolution. However, its computational workload might be unaffordable when the number of snapshots is large. In this paper, we propose two gridless methods for 2-D DOA estimation with L-shaped array based on the atomic norm to improve the computational efficiency. Firstly, by exploiting the cross-covariance matrix an ANM-based model has been proposed. We then prove that this model can be efficiently solved as a semi-definite programming (SDP). Secondly, a modified model has been presented to improve the estimation accuracy. It is shown that our proposed methods can be applied to both uniform and sparse L-shaped arrays and do not require any knowledge of the number of sources. Furthermore, since our methods greatly reduce the model size as compared to the conventional ANM method, and thus are much more efficient. Simulations results are provided to demonstrate the advantage of our methods.
In order to reduce network energy consumption and prolong the lifetime of wireless sensornetworks, this paper improves the LEACH to an energy efficient multi-hop routing algorithm. LEACH in the cluster creation, data...
详细信息
In view of the existing web site credibility evaluation models failing to comprehensively consider the multiple attributes of trust,after learning the D- S evidence theory and fuzzy theory,a new web site credibility e...
详细信息
ISBN:
(纸本)9781510805750
In view of the existing web site credibility evaluation models failing to comprehensively consider the multiple attributes of trust,after learning the D- S evidence theory and fuzzy theory,a new web site credibility evaluation model is put forward in which the trust value is divided into historical trust value and the current trust *** study shows that the model can track the change of user preferences in real-time as well as can meet the dynamic nature of universal environment,so the trust assessment is more reasonable and accurate.
Graphical models have been widely applied in solving distributed inference problems in wireless sensornetworks (WSNs). In this paper, we formulate the sensor self-localization problem in a WSN as an inference problem...
详细信息
Graphical models have been widely applied in solving distributed inference problems in wireless sensornetworks (WSNs). In this paper, we formulate the sensor self-localization problem in a WSN as an inference problem on a factor graph. Using a sequential schedule of message updates, a sequential uniformly reweighted sum-product algorithm (SURW-SPA) is developed for self-localization problems. The proposed algorithm combines the distributed nature of belief propagation (BP) with the improved performance of sequential tree-reweighted message passing (TRW-S) algorithm. We apply the SURW-SPA to sensor self-localization in large-scale static networks, and evaluate its performance in terms of localization accuracy and convergence speed.
Real-time hand tracking is fundamental to human gesture recognition. However, due to the huge computation, previous studies are either off-line or limited to given poses. In order to satisfy the requirement of real-ti...
详细信息
Real-time hand tracking is fundamental to human gesture recognition. However, due to the huge computation, previous studies are either off-line or limited to given poses. In order to satisfy the requirement of real-time hand tracking, in this paper we propose a real time hand tracking method using Kinect. Firstly we extract the hand region from the depth image output from Kinect. Then we achieve the hand parameters. During the procedure of hand region extraction, we propose a cascade structure with recursive connected component algorithm to improve the efficiency and reserve connection relationships in 3D space. To determine the fingertips, we use the former 3D connections and geodesic distance over hand skeleton pixels to guarantee the accuracy and robustness, with acceptable loss. Experimental results show that our proposed solution can significantly improve the quality of real-time hand tracking.
In this paper, we explore network architecture anal key technologies for content-centric networking (CCN), an emerging networking technology in the big-data era. We descrihe the structure anti operation mechanism of...
详细信息
In this paper, we explore network architecture anal key technologies for content-centric networking (CCN), an emerging networking technology in the big-data era. We descrihe the structure anti operation mechanism of tl CCN node. Then we discuss mobility management, routing strategy, and caching policy in CCN. For better network performance, we propose a probability cache replacement policy that is based on cotent popularity. We also propose and evaluate a probability cache with evicted copy-up decision policy.
The Delay/Disruption Tolerant network (DTN) is characterized with intermittently connectivity and frequent partitions. Due to this, the DTN routing algorithm is very vulnerable to various kinds of malicious attacks. I...
详细信息
ISBN:
(纸本)9781479970926
The Delay/Disruption Tolerant network (DTN) is characterized with intermittently connectivity and frequent partitions. Due to this, the DTN routing algorithm is very vulnerable to various kinds of malicious attacks. In this paper, combined with many other DTN forwarding approaches, we propose a multi-stage routing forwarding algorithm so as to further improve network throughput. To ensure the security advantages of multi-period routing, we also need to take the efficiency of the routing algorithm into account. The concept of trade-off coefficient is introduced in this paper in order to balance the security and efficiency. Through the security analysis and simulation result, we can reach the conclusion that by setting a reasonable trade-off factor, the mechanism can effectively resist the joint attack and also optimize the network performance.
Face recognition research mainly focuses on traditional 2D color images, which is extremely susceptible to be affected by external factors such as various viewpoints and has limited recognition accuracy. In order to a...
详细信息
Face recognition research mainly focuses on traditional 2D color images, which is extremely susceptible to be affected by external factors such as various viewpoints and has limited recognition accuracy. In order to achieve improved recognition performance, as well as the 3D face holds more abundant information than 2D, we present a 3D human face recognition algorithm using the Microsoft's Kinect. The proposed approach integrates the depth data with the RGB data to generate 3D face raw data and then extracts feature points, identifies the target via a two-level cascade classifier. Also, we build a 3D-face database including 16 individuals captured exclusively using Kinect. The experimental results indicate that the introduced algorithm can not only achieve better recognition accuracy in comparison to existing 2D and 3D face recognition algorithms when the probe face is exactly in front of Kinect sensor, but also can increase 9.3% of recognition accuracy compared to the PCA-3D algorithm when it is not confronting the camera.
This paper addresses the issue of direction-of-arrival (DOA) estimation under the coexistence of mutual coupling and nonuniform noise. We first construct a selection matrix based on the structural characteristics of t...
详细信息
ISBN:
(数字)9781728119465
ISBN:
(纸本)9781728119472
This paper addresses the issue of direction-of-arrival (DOA) estimation under the coexistence of mutual coupling and nonuniform noise. We first construct a selection matrix based on the structural characteristics of the mutual coupling matrix (MCM) to form a new manifold matrix which is not affected by the mutual coupling effect. Then, we employ the covariance fitting criteria to form a semidefinite programming for DOA estimation. Compared to most of existing sparse methods, our method does not require discretization of the angle space, hence is able to achieve super-resolution. Simulations are carried out to verify the effectiveness of the propose method.
暂无评论