A novel part-based approach to mean-shift algorithm for visual tracking

Visual tracking is one of the most important problems considered in computer vision. To improve the performance of the visual tracking, a part-based approach will be a good solution. In this paper, a novel method of visual tracking algorithm named part-based mean-shift (PBMS) algorithm is presented. In the proposed PBMS, unlike the standard mean-shift (MS), the target object is divided into multiple parts and the target is tracked by tracking each individual part and combining the results. For the part-based visual tracking, the objective function in the MS is modified such that the target object is represented as a combination of the parts and iterative optimization solution is presented. Further, the proposed PBMS provides a systematic and analytic way to determine the scale of the bounding box for the target from the perspective of the objective function optimization. Simulation is conducted with several benchmark problems and the result shows that the proposed PBMS outperforms the standard MS.

     

A high performance parallel DCT with OpenCL on heterogeneous computing environment

A noteworthy thing in desktop PCs is that they can provide a great opportunity to increase the performance of processing multimedia data by exploiting task- and data-parallelism with multi-core CPU and many-core GPU. This paper presents a high performance parallel implementation of 2D DCT on this heterogeneous computing environment. For this purpose, Intel TBB (threading building blocks) and OpenCL (Open Compute Language) are utilized for task- and data-parallelism, respectively. The simulation result shows that the parallel DCT implementations far the serial ones in processing speed. Especially, OpenCL implementation shows a linear speedup, a typical SIMD characteristic as the increase of 2D data sets.     

Polar-Cartesian Hybrid Transforms: A novel 2D range scan registration algorithm

Registration of range scans is commonly required in many localization and mapping algorithms. In this paper, we introduce a novel approach called Polar-Cartesian Hybrid Transforms for pair-wise registration of range scans. The proposed algorithm iteratively establishes correspondences by searching the points with closest polar angles in the polar coordinate frame. An angular look-up table is constructed based on the properties of the laser range finder to accelerate the searching procedure. In order to speed up the convergence, we compute the difference of polar range of every matched point pair to select the most contributing correspondences. After the correspondences are determined, the transformation is computed in Cartesian coordinate frame using a point-to-line metric. Combining the advantages of the polar and Cartesian coordinate frames, both robustness and efficiency are greatly improved compared with an up-to-date ICP algorithm.     

Neutron diffraction texture study of deformed uranium plates

The texture of two depleted uranium (DU) samples, labelled DUWR and DUWR2, were studied by neutron diffraction. DUWR was prepared by warm rolling of a cast ingot, and DUWR2 was prepared by adding 20% tensile strain to the warm-rolled DUWR. Complete three-dimensional orientation distribution functions were determined using four neutron pole figures for the DUWR, and using six neutron pole figures for the DUWR2 sample, by the WIMV method of the program popLA. The textures of the two samples were essentially identical to each other. They could be described by a twisted helical density tube spiralling continuously along the -axis of the Euler space. The projection of the backbone of the density tube along the -axis cast a linear shadow running parallel to the diagonal of the - plane, which could be defined by a =+90° (and =+270°) relation. The helical tube was confined within narrow -angle limits, from 14° to 30° with the peak orientation at (103) 0 10. The diffraction patterns of the DUWR2 sample were measured from the normal direction to the rolling surface of the sample, up to the scattering angle of 108° using a 0.15 nm neutron beam. The Rietveld profile refinement using the textured diffraction pattern was quite satisfactory when the texture effect to the entire diffraction profile was corrected for by the corresponding pole density from the inverse pole figure.     

Mobile robot localization with gyroscope and constrained Kalman filter

The odometry information used in mobile robot localization can contain a significant number of errors when robot experiences slippage. To offset the presence of these errors, the use of a low-cost gyroscope in conjunction with Kalman filtering methods has been considered by many researchers. However, results from conventional Kalman filtering methods that use a gyroscope with odometry can unfeasible because the parameters are estimated regardless of the physical constraints of the robot. In this paper, a novel constrained Kalman filtering method is proposed that estimates the parameters under the physical constraints using a general constrained optimization technique. The state observability is improved by additional state variables and the accuracy is also improved through the use of a nonapproximated Kalman filter design. Experimental results show that the proposed method effectively offsets the localization error while yielding feasible parameter estimation.     

Improving traversability of quadruped walking robots using body movement in 3D rough terrains

This paper presents a study on improving the traversability of a quadruped walking robot in 3D rough terrains. The key idea is to exploit body movement of the robot. The position and orientation of the robot are systematically adjusted and the possibility of finding a valid foothold for the next swing is maximized, which makes the robot have more chances to overcome the rough terrains. In addition, a foothold search algorithm that provides the valid foothold while maintaining a high traversability of the robot, is investigated and a gait selection algorithm is developed to help the robot avoid deadlock situations. To explain the algorithms, new concepts such as reachable area, stable area, potential search direction, and complementary kinematic margin are introduced, and the effectiveness of the algorithms is validated via simulations and experiments.     

Autonomous topological modeling of a home environment and topological localization using a sonar grid map

This paper presents a method of autonomous topological modeling and localization in a home environment using only low-cost sonar sensors. The topological model is extracted from a grid map using cell decomposition and normalized graph cut. The autonomous topological modeling involves the incremental extraction of a subregion without predefining the number of subregions. A method of topological localization based on this topological model is proposed wherein a current local grid map is compared with the original grid map. The localization is accomplished by obtaining a node probability from a relative motion model and rotational invariant grid-map matching. The proposed method extracts a well-structured topological model of the environment, and the localization provides reliable node probability even when presented with sparse and uncertain sonar data. Experimental results demonstrate the performance of the proposed topological modeling and localization in a real home environment.     

Hop chains: Secure routing and the establishment of distinct identities

We present a secure routing protocol that is immune to Sybil attacks and that can tolerate collusion of Byzantine routers. It can tolerate either initial collusion of Byzantine routers or runtime collusion of non-adjacent Byzantine routers, both in the absence of runtime collusion between adjacent routers. For these settings, the calculated distance from a destination to a node is not smaller than the actual shortest distance from the destination to the node. The protocol can also simultaneously tolerate initial collusion of Byzantine routers and runtime collusion of adjacent Byzantine routers but in the absence of runtime collusion between non-adjacent routers. For this setting, it guarantees a bound on the difference between the calculated distance and the actual shortest distance. The bound depends on the number of Byzantine routers on a path. The protocol makes very weak timing assumptions and requires synchronization only between neighbors or second neighbors. We propose to use this protocol for secure localization of routers using hop-count distances, which can be then used as a proof of identity of nodes.     

Biofouling control by quorum sensing inhibition and its dependence on membrane surface

Biofouling control by quorum sensing (QS) inhibition and the influence of membrane surface characteristics on biofilm formation and QS inhibition were investigated. Pseudomonas putida isolated from the bio-fouled reverse osmosis (RO) membranes in a real plant was used. Acylase was chosen as a model QS inhibitor. Bacteria on the membrane coupons were quantified with the heterotrophic plate count method. Cell distribution was imaged by a confocal laser scanning microscope. Results showed that biofilm formation on the membrane was reduced by acylase as it inhibits the activity of N-acylhomoserine lactone (AHL) which is a signal molecule of QS. It was also shown that membrane surface characteristics were influential factors affecting bacterial adhesion, biofilm formation, and QS inhibition.     

Proactive Human Search for the Designated Person with Prior Context Information in an Undiscovered Environments

This paper describes a scheme for proactive human search for a designated person in an undiscovered indoor environment without human operation or intervention. In designing and developing human identification with prior information a new approach that is robust to illumination and distance variations in the indoor environment is proposed. In addition, a substantial exploration method with an octree structure, suitable for path planning in an office configuration, is employed. All these functionalities are integrated in a message- and component-based architecture for the efficient integration and control of the system. This approach is demonstrated by succeeding human search in the challenging robot mission of the 2009 Robot Grand Challenge Contest.