Search

In this paper, we propose a partially-blurred-image classification and analysis framework for automatically detecting images containing blurred regions and recognizing the blur types for those regions without needing to perform blur kernel estimation and image deblurring. We develop several blur features modeled by image color, gradient, and spectrum information, and use feature parameter training to robustly classify blurred images. Our blur detection is based on image patches, making region-wise training and classification in one image efficient. Extensive experiments show that our method works satisfactorily on challenging image data, which establishes a technical foundation for solving several computer vision problems, such as motion analysis and image restoration, using the blur information.

In image deconvolution, the boundary value problem, if not appropriately handled, often causes serious ringing artifacts in the restored results. This paper proposes a simple method to tackle this problem without any assumption on the noise level and the symmetry of the Point Spread Function (PSF). We establish new boundary conditions by smoothly expanding the input image to a large tile. It helps reducing the boundary discontinuities and accordingly makes all restoration filters based on Fast Fourier Transform (FFT) not produce obvious image border artifacts.

Inspired by the foraging behaviors of real ants, an ant-based routing algorithm for mobile ad hoc networks (MANET) should adapt the special characteristics of the networks. It should be self-configurable, self-built and distributed. In this paper, we present a bi-directional and hop-by-hop pheromone-updated ant-based routing (BHPAR) algorithm for MANET. The algorithm can accelerate the convergence rate of the existing ant-based routing algorithms and improve the response speed to abnormal cases in MANET. Besides, BHPAR provides redundant paths for routing and thus improves the reliability and survivability of the network. Simulation results show that BHPAR performs well in average end-to-end delay and routing overhead as well as packet delivery rate.

Mobile ad hoc networks (MANET) are characterized by the unpredictable change in the quality of wireless communication due to node movement and failure. To cope with such a special characteristic, various ant-based routing algorithms have been proposed for MANET. However, these algorithms still have several problems unsolved, including the congestion problem and the shortcut problem. In this paper, we present a new ant-based routing algorithm, called ant-based routing algorithm with admission control and noise encouraged route-selection (ARACNE). The algorithm adopts two mechanisms, called admission control (AC) and noise encouraged route-selection (NE), to solve the congestion problem and the shortcut problem, respectively. Simulation results show that both mechanisms can improve the performance of ant-based routing algorithm for MANET by reducing the routing overhead and increasing the packet delivery rate.

The paper describes a novel ant-based distributed route algorithm (ADRA) for ad-hoc networks. The ants move across the network between randomly chosen pairs of nodes; as they move they deposit simulated pheromones as a function of their distance from their source node, the quality of the link, the congestion encountered on their journey, the current pheromones the nodes possess and the velocity at which the nodes move. Of course, the node changes the pheromones by itself according to the quality of the link, and ages the link by pheromones evaporating. An ant selects its path at each intermediate node according to the distribution of simulated pheromones at each node. In order to accelerate the convergence rate of the congestion problem and the shortcut problem, we give the parameters different weight values to update the probability routing table. The performance of the algorithm is measured by the packet loss ratio, control overhead as well as end-to-end packet delay. The results of using the improved ant-based control protocols are compared with those achieved by using fixed shortest-path routes previously proposed for use in ad-hoc networks. The ADRA system is shown to result in fewer call failures than the other methods, while exhibiting many attractive features of distributed control.

The paper presents a novel dynamic load-aware based load-balanced routing (DLBL) algorithm for ad hoc networks. DLBL considers an intermediate node routing load as the primary route selection metric while discovering a route; this can lead to less network congestion and fewer bottlenecks. During route maintaining periods, DLBL deals, by a distributed mechanism, with the congestion of active routes when nodes or the routes have their queue overload, at the same time with little overhead increment. When a link breaks because of node mobility or power off, etc., DLBL provides efficient path maintenance to patch up broken links to help get a robust route from the source to the destination. By presenting and analyzing simulation results, DLBL is shown to result in good performance of packet delivery ratio and average end-to-end delay, while exhibiting many attractive features of distributed control to adapt to dynamic ad hoc networks.