In this paper, we investigate a fundamental assignment problem for a type of constrained group service. A group is considered successfully formed only if the number of assigned patrons falls within the specified lower and upper-bound constraints. We introduce the concept of rank-consistent to characterize a useful subset of the assignment problem. Although the problem is still NP-hard, we introduce the technique of allocation vectors to handle the complexity. This technique not only allows us to exploit several characteristics for performance optimization but also enables us to design a generic procedure to obtain the optimal physical assignment under a given allocation vector. We lay the foundation of a 1/2 approximation algorithm and a branch and bound algorithm for seeking the optimal solution. The branch and bound algorithm employs a specific pattern of optimal allocation vector and several newly proposed pruning techniques, especially the one that utilizes the dominance relations between allocation vectors. Extensive experiments demonstrate that our algorithm is much more effective than a set of heuristic greedy algorithms.

Glaucoma is caused by damaged optic nerves, and can lead to permanent vision loss. The cup-to-disk ratio (CDR) is a key criterion for glaucoma diagnosis, therefore an accurate automatic segmentation of the optic disc (OD) and optic cup (OC) in retinal fundus images has become a major research topic. However, with deeper deep learning models being used to complete a fundus segmentation, the segmentation results remain acceptable only on certain datasets. In this research, a lightweight deep-learning encoder-decoder architecture, which adopts polar coordinate transformation and histogram equalization to simplify the learning complexity, is proposed for the simultaneous segmentation of OD and OC. The proposed model employs an encoder-decoder architecture consisting of a feature extraction module and a multi-output module to both simplify the model and to adjust the result through different outputs and multilabel loss functions. Experiment results demonstrate that the proposed approach outperforms other deep network models on OD and OC segmentation over the datasets REFUGE, MESSIDOR, and RIM-ONE, which contain images captured from cameras with various specifications. In addition, better results on glaucoma screening through CDR calculation were obtained on the REFUGE dataset with the proposed method.

Traditional Chinese is significantly different from other major languages, as compared to western countries use Latin alphabet letters system, Chinese contains over 13,000 common and sub-commonly used characters. Each of these Chinese characters is composed of multiple parts, which are arranged and combined according to various orthographies. Therefore, how to handle these enormous character counts and how to design of more efficient fonts are longstanding issues in the creation of Chinese fonts. In this study, we seek to address these issues using deep learning based on the concept of style transfer. Previous font generating methods generally require multiple pre-processing steps for feature extraction, and these cannot be simultaneously adapted to typographic and handwritten characters. In this study, we propose a neural network architecture for Chinese fonts conversion, which directly uses images as inputs. In this architecture, neural networks are trained using a small number of selected characters, which are then used to extract and convert the features of an input font. The outputs of the architecture are characters with identical content albeit different font styles. We apply our method to ten different font style including typographic and handwriting and demonstrate it capable of converting Chinese-character fonts into fonts that are similar to the original fonts.

Group services are highly important for a variety of computing application domains. In this paper, we study the fundamental problem of allocating a set of service patrons to a set of service groups in an attempt to maximize the total profit gained by the grouping platform. The problem under consideration is unique in that group service is not provided at all unless its lower bound requirement is satisfied. In addition, we allow each service patron to join multiple groups. In this paper, after proving the hardness property of the problem, we focus first on a special case of the problem. To this end, we propose two approaches. One aims at providing a suboptimal solution using a 1/2-approximation algorithm. The other approach turns to seeking an optimal solution using a branch and bound technique. For this purpose, we introduce a theorem that captures a useful property of an optimal allocation. Based on this theorem, we design an efficient branch and bound algorithm to find an optimal solution. We then extend these methods to solve the general problem. Extensive experiments show that our branch and bound algorithm is able to obtain an optimal solution with a small amount of computation time in many different settings.

Target detection is an important problem in wireless sensor networks where a number of sensors form a network to detect the presence or absence of a certain target or event. Data fusion is a potential method broadly used to improve detection performance when the sampling data are noisy. However, low detection probability cannot be avoided if detection decisions are made based on a collection of sampling data taken at just one particular moment. This paper adopts fusion-based sequential detection to guarantee the quality of detection results. A fusion center is used to collect local data from individual sensors periodically. A final detection decision is made only after the pre-defined constraints of false alarm and missing probability are satisfied. Rules for each sensor to make local decisions and for the fusion center to make global decisions are derived. Simulations are conducted to show the latency of making the final decisions based on the proposed fusion scheme.

Along with the growth of cloud computing and mobile devices, the importance of client device identity concern over cloud environment is emerging. To provide a lightweight yet reliable method for device identification, an application layer approach based on clock skew fingerprint is proposed. The developed experimental platform adapts AJAX technology to collect the timestamps of client devices in the cloud server during connection time, then calculate the clock skews of client devices. Few methods based on linear regression and piecewise minimum algorithm are developed to optimize the precision and shorten timestamp collection process. A jump point detection scheme is also proposed to resolve the offset drifting problem, which is usually caused by switching network or temporary disconnection. Finally, two experiments are conducted to study the effectiveness of clock skew fingerprint, and the results illustrate that the false positive rate and the false negative rate, in the worst case, are both no more than 8% when the tolerance threshold is set appropriately.

This paper develops a tensor-based 3D face reconstruction approach from a single color image. Instead of considering grayscale images, this paper also takes additional color factors into account while constructing the tensor model. Canonical correlation analysis then is applied to establish the relationship between the color image and the depth information in the face database. During the face reconstruction, given a single color face image, the depth estimation is recovered from the CCA-based mapping between the tensor models. Experiments show that the proposed approach improves the reconstructed results under different lighting conditions and poses.

A sequence-based visual navigation system for autonomous robots is proposed. A sequence of reference images representing a transportation route is taken as the navigation clues. When a robot navigates in a route, two navigation modes, with better reduction in total transportation time required, are switched according to the difference of scales between the robot's current view and the targeted reference image. When the scale is large, log-polar transform is applied to match regions in two images. A less accurate moving direction is inferred from the matching result. After the difference of scale is reduced when the robot approaches the targeted waypoint, SIFT features are matched between two images. An accurate moving direction is decided from the epipolar geometry which is computed from matched SIFT features. The robot's moving velocity is adjusted accordingly, thus reducing the overall transportation time. While other sequence-based navigation methods moves the robot at a maximal velocity of 10 cm/sec, the proposed method moves the robot in an average velocity of 26.5 cm/sec during a 117.3 m long route with 11 reference images. Comparing to a similar sequence-based navigation research, the proposed method does not only reduce transportation time by at least half, but also requires less number of reference images in route transportation applications.

Broadcast presents a special challenge for wireless sensor networks (WSNs). In some situation such as time synchronization or building routing path, broadcasting messages must be securely transmitted to all nodes, but this process is subject to attack by adversaries. For example, an adversary may try to waste the battery power of intermediate nodes by forcing a compromised node to repeatedly rebroadcast, thus causing a denial-of-service (DoS) attack. One way to solve this problem is use only part of the nodes in the network as the intermediate nodes, limiting the effects of the attack. In this paper, we propose a novel broadcast protocol: broadcast power preserving (BOPP). In BOPP, a packet reception reliability metric of each network component is discovered. This reliability score gives the packet reception rate of each communication edge in the network. With the scoring metric, BOPP can judge the network reliability from time to time and adapt the network to provide maximum reliability while minimizing the energy cost. This enables the network to resist DoS attacks.

No abstract available.