Advanced Materials Research Vols. 846-847

Abstract: Camera self-calibration is a key step in photo based reconstruction works. Vanishing point detection is a very important method in single photo based camera self-calibration. We research current vanishing point detection method and put forward a detection algorithm based on a new clustering idea: J-Linkage. First, we construct a similar concept space from straight lines from images by edge detection and segmentation method. Then, we cluster the similar concept space to decide several main directions. At last, we can easily estimate vanishing points from the clustered categories. Experiment prove that, the method has a high performance efficiency and with good accuracy.

Abstract: Pose estimation of 3d object is a hot research in the field of computer vision. This paper presents a novel pose estimation method based on colored markers. To overcome the effect of the luminance and other colors, this method uses the HSV color space and isolates the colors operating only on chromaticity plane where value (V) has no actual effect for identifying the colored regions of interest. The template is then applied on the remaining colors in order to find the center of the region. The pixels which have the same color but are not in the marker area are excluded, since they are considered noisy. The template guarantees the stability and efficiency of the extraction of the feature points. Compared with the CDT algorithm, the proposed method can extract reliable center points, and has higher accuracy in pose estimation for planar rigid objects. At last, experimental results demonstrate the efficiency of the method.

Abstract: Taking use of VirtuoZos digital photogrametric workstations of Supresoft Inc., employing full digital aerophotogrammetric mapping method, according to existing aerial photographic image database, at the foundation of basic control survey, we carry out aerial photographic control survey, surveying and mapping with pile point method, digital aerotriangulation and full digital stereocompilation. When taking 1:1000 topographic map obtained by indoor edition to field work for compration, supplement and regulation, we mainly aim at the following purposes: determining the nature and quantity as well as filling vacancy for ground and land features obtained by indoor stereocompilation; positioning for habitation, wall and railing, incompletely surveyed, drop shadow-covered and trees-obstructed ground features, as well as newly-added planimetric features after aerial photogrammetry; especially measuring the solid width of a house to centimeter and making eaves correction after measuring the length of a house when needed; correcting the geomatric positioning relationand of ground features according to their solid shape and fairly drawing them in the map; afterwards, we carry out graphic edition again during indoor work. Only in this way can we guarantee the horizontal and vertical accuracy of large scale photogrammetric mapping.

Abstract: A modified propagator method based on L-shaped array for 2-Dimensional (2-D) direction of arrival (DOA) estimation in monostatic MIMO radar is proposed. A cross-correlation matrix, which can eliminate the influence of noise, is constructed by the received data from the two orthogonal uniform linear arrays (ULAs) at x-axis and z-axis. Then the matrix can be utilized to estimate signal subspace of 2-D DOA through propagator method. At last, the elevation and azimuth angles of the 2-D DOA is automatically paired by the complex eigenvalues of a low-order complex matrix. The 2-D DOA estimation performance of the proposed method is better than conventional propagator method and ESPRIT algorithm. Simulation results verify the effectiveness of the proposed method.

Abstract: Telemetry data frame structure is complicated and changeable, so telemetry pre-processing software cannot be universal. To solve this problem, a component method was proposed in this paper, which can effectively compensate for the deficiencies of the traditional method. XML files were employed to configure telemetry parameters, including the information of appropriate processing method for data processing. Based-on memory-mapped telemetry source data extraction can greatly improve source extraction speed, and data integrity is guaranteed by sub-frame data fusion. Subsequent telemetry software developing shows that the method can improve the reusability of pre-processing module and shorten the system development time.

Abstract: IETM has been an important role in equipment integrated support. IETM works mainly through the guidance of the equipment maintenance support information retrieval. So, IETM information retrieval technology is one of the key technologies in the application of integrated equipment support. This paper proposes a new method of IETM information query based on Bayesian network model. The model supports for the IETM information of structured query. IETM information query is based features, the use the IETM dataset words, the elements and structure of unit stats information on the topology of the model structure and the conditional probability learning. Combination of the probability functions of the method and using the model of the probabilistic inference process, the relevance structured IETM information query conditions were estimated. Finally, the information retrieval process is analyzed.

Abstract: This paper proposes an improved acquisition method for high-order binary-offset-carrier (BOC) modulated signals based on fractal geometry. We introduced the principle of our acquisition method, and outlined its framework. We increase the main peak to side peaks ratio in the BOC autocorrelation function (ACF), with a simple fractal geometry transform. The proposed scheme is applicable to both generic high-order sine-and cosine-phased BOC-modulated signals. Simulation results show that the proposed method increases output signal to noise ratio (SNR).

Abstract: This paper presents a parallel improved niche genetic algorithm (PINGA) for 3D stealth coverage corridors real-time planning of unmanned aerial vehicles (UAVs) operating in a threat rich environment. 3D corridor was suggested to meet the diversity kinematics constraints of UAVs. Niche genetic algorithm (NGA) was improved by merging neighborhood mutation operator and hill climbing algorithm, and performed in parallel. Additionally, the crowding strategy based on high value targets was used to generate coverage trajectories in the area of interest (AOI). Preliminary results in virtual environments show that the approach for UAVs high quality flight corridors planning is real-time and effective.

Abstract: Based on the theory that the steering vectors of coherently distributed targets for bistatic MIMO radar possess a property called Hadamard product rotational invariance (HPRI), a novel central angle estimation algorithm is derived. The proposed algorithm has the advantage of no searching and easily pair the estimated angles. Furthermore, it can deal with different coherently distributed targets with different angular distribution functions or unknown angular distribution functions. Simulation results show the proposed method is effective.

Abstract: As for JMVC (Joint Multi-view Video Coding), when executing motion and disparity estimation simultaneously, the implicit correlation of reference directions between current and neighboring macroblocks is neglected, and this will lead to useless searching. To reduce this redundancy, we propose an estimation algorithm which sets search ranges properly by exploiting the mentioned correlation. The proposed algorithm first gets the reference directions of macroblocks in the left, top and top right direction of current macroblock, and compares them with current reference direction, then sets search ranges by the comparison results, executes motion and disparity estimation in the new ranges at last. Experimental results show that the proposed algorithm can save 28.98%- 46.30% of coding time without degradation of coding quality comparing to JMVC.