EMPIRE10 (Evaluation of Methods for Pulmonary Image REgistration 2010) is a public platform for fair and meaningful comparison of registration algorithms which are applied to a database of intrapatient thoracic CT image pairs. Evaluation of nonrigid registration techniques is a nontrivial task. This is compounded by the fact that researchers typically test only on their own data, which varies widely. For this reason, reliable assessment and comparison of different registration algorithms has been virtually impossible in the past. In this work we present the results of the launch phase of EMPIRE10, which comprised the comprehensive evaluation and comparison of 20 individual algorithms from leading academic and industrial research groups. All algorithms are applied to the same set of 30 thoracic CT pairs. Algorithm settings and parameters are chosen by researchers expert in the configuration of their own method and the evaluation is independent, using the same criteria for all participants. All results are published on the EMPIRE10 website (http://empire10.isi.uu.nl). The challenge remains ongoing and open to new participants. Full results from 24 algorithms have been published at the time of writing. This paper details the organization of the challenge, the data and evaluation methods and the outcome of the initial launch with 20 algorithms. The gain in knowledge and future work are discussed.

We introduce a labeled point set registration algorithm based on a family of novel information-theoretic measures derived as a generalization of the well-known Shannon entropy. This generalization, known as the Havrda-Charvat-Tsallis entropy, permits a fine-tuning between solution types of varying degrees of robustness of the divergence measure between multiple point sets. A variant of the traditional free-form deformation approach, known as directly manipulated free-form deformation, is used to model the transformation of the registration solution. We provide an overview of its open source implementation based on the Insight Toolkit of the National Institutes of Health. Characterization of the proposed framework includes comparison with other state of the art kernel-based methods and demonstration of its utility for lung registration via labeled point set representation of lung anatomy.

We propose an efficient image registration strategy that is based on learned prior distributions of transformation parameters. These priors are used to constrain a finite- time multi-start optimization method. Motivation for this approach comes from the fact that standard affine brain image registration methods, especially those based on gradient descent optimization alone, are affected by the initial search position. While global optimization methods can resolve this problem, they are are often very time consuming. Our goal is to build an explicit prior model of the gap between a typical registration solution and the solution gained by a global optimization method. We use this learned prior model to restrict randomized search in the relevant parameter space surrounding the initial solution. Global optimization in this restricted parameter space provides, in finite time, results that are superior to both gradient descent and the general multi-start strategy. The performance of our method is illustrated on a data set of 67 elderly and neurodegenerative brains. Our novel learning strategy and the associated registration method are shown to outperform other approaches. Theoretical, synthetic and real-world examples illustrate this improvement.

We introduce a novel topological formulation for contour grouping. Our grouping criterion, called untangling cycles, exploits the inherent topological 1D structure of salient contours to extract them from the otherwise 2D image clutter. To define a measure for topological classification robust to clutter and broken edges, we use a graph formulation instead of the standard computational topology. The key insight is that a pronounced ID contour should have a clear ordering of edges, to which all graph edges adhere, and no long range entanglements persist. Finding the contour grouping by optimizing these topological criteria is challenging. We introduce a novel concept of circular embedding to encode this combinatorial task. Our solution leads to computing the dominant complex eigenvectors/eigenvalues of the random walk matrix of the contour grouping graph. We demonstrate major improvements over state-of-the-art approaches on challenging real images.

This paper presents autonomous control for micro-flying robot (muFR) and small helicopter X.R.B. In case of natural disaster like earthquake, a MAV is very effective for surveying the site and environment in dangerous area or narrow space, where human cannot access safely. In addition, it is a help to prevent secondary disaster. This paper is concerned with autonomous hovering control, guidance control of muFR, and automatic takeoff and landing control of X.R.B

One important aspect of constructing an overlay network is how to exploit network locality in the underlying network. In this paper, we propose a scalable protocol for constructing an overlay network that takes account of locality of network hosts. The constructed overlay network can significantly decrease the communication cost between end-hosts. Our simulation results show that the average distance between a pair of hosts in the constructed overlay network is only about 11% of the one in a traditional, randomly connected overlay network. Furthermore, our proposed overlay considered to be more scalable than tree-based or mesh-based overlays.

There are many research interests in peer-to-peer (P2P) overlay architectures. Most widely used unstructured P2P networks rely on central directory servers or massive message flooding, clearly not scalable. Structured overlay networks based on distributed hash tables (DHT) are expected to eliminate flooding and central servers, but can require many long-haul message deliveries. An important aspect of constructing an efficient overlay network is how to exploit network locality in the underlying network. We propose a novel mechanism, mOverlay, for constructing an overlay network that takes account of the locality of network hosts. The constructed overlay network can significantly decrease the communication cost between end hosts by ensuring that a message reaches its destination with small overhead and very efficient forwarding. To construct the locality-aware overlay network, dynamic landmark technology is introduced. We present an effective locating algorithm for a new host joining the overlay network. We then present a theoretical analysis and simulation results to evaluate the network performance. Our analysis shows that the overhead of our locating algorithm is O(logN), where N is the number of overlay network hosts. Our simulation results show that the average distance between a pair of hosts in the constructed overlay network is only about 11% of the one in a traditional, randomly connected overlay network. Network design guidelines are also provided. Many large-scale network applications, such as media streaming, application-level multicasting, and media distribution, can leverage mOverlay to enhance their performance.

In this paper, a novel model named MultiServ is proposed to alleviate the congestion and to provide better quality of service for end-host using overlay network. In MultiServ, a special overlay is built so that end-host and its neighbors can cooperatively transmit data efficiently. Meanwhile, a joint congest control scheme is proposed for multiple path data transmission. As a result, the traffic in the underlying network can be balanced and smoothed and the congestion can be alleviated or avoided. This provides a promising solution for application with demand of good quality of service for throughput sensitive transmissions.

In this paper, we propose a performance analysis model to study the reach-ability in unstructured overlay networks. Given a node's degree distribution and a network size, n, the model describes the flooding query pattern in a P2P network accurately. Also, we prove that in such an overlay network, the average distance between any two hosts is limited by O(log N). This model is simple and accurate, and therefore is a very useful tool in exploring many properties of massive overlay networks. It can be applied to P2P based content distribution networks and ad hoc wireless networks, for example.