Real-world Coordinates Research Articles

Pattern recall skills of talented soccer players: Two new methods applied

In this study we analyzed the pattern recall skills of talented soccer players by means of two innovative methods of analysis and gaze behavior data. Twenty-two young female soccer players watched video clips of 3 vs. 3 small-sided games and, after occlusion, had to reproduce the positions of the players. Recall performance was measured by calculating the spatial error of the recalled player positions at the moment of occlusion and at consecutive 33ms increments. We analyzed player positions relative to each other, by assessing geometric pattern features in terms of angles between players, and we transformed the data into real-world coordinates to exclude the effects of the 2D perspective in the video clips. The results showed that the participants anticipated the movements of the patterns. In real-world coordinates, the more experienced players anticipated the pattern further in advance than the less experienced players and demonstrated a higher search rate, a shorter fixation duration and a higher fixation order. The differences in recall accuracy between the defensive and offensive elements were not consistent across the methods of analysis and, therefore, we propose that perspective effects of the video clip should be taken into account in further research.

Quantifying Tidal Mud Flat Elevations From Fixed-Platform Long-Wave Infrared Imagery

A procedure for estimating tidal mud flat topography from fixed-platform, long-wave infrared imagery is presented. Shallow water and low bearing capacity on many mud flats hinder traditional surveying methods by water craft, walking, or land vehicle. The approach utilizes identification of the intersection of water with the surface of the mud flat through a rising tide. Waterlines on mud flats are often indistinguishable to the naked eye and in visible-band imagery. Long-wave infrared imagery, relying on emitted radiance, provides a more distinct delineation between an exposed mud flat and water. The waterline is identified via an image intensity threshold and transferred to real-world coordinates using a geometrical transformation model that accounts for potential imager sway. Elevation estimates, interpolated to a uniform grid, show excellent agreement (absolute error generally less than 0.02 m) with ground truth elevations obtained using a sled-mounted global positioning survey system.

타원체 모형을 통한 어안 영상 보정

어안렌즈의 특성으로 인해 발생하는 왜곡을 보정하는 일반적인 방법은 크게 두 가지 방법으로 분류할 수 있다. 첫 번째 방법은 렌즈의 특성을 고려한 수학적 모델링을 통한 보정 방법이고, 두 번째 방법은 렌즈의 종류와 상관없이 획득되는 영상만을 통한 보정 방법이다. 렌즈의 특성을 고려하는 경우는 렌즈의 파라미터와 실세계에 존재하는 3차원 실세계좌표와 2차원 영상좌표의 관계를 통해 기하학적으로 보정 식을 구할 수 있다. 그러나 기존 어안렌즈 보정에 관한 논문들은 구형태의 어안렌즈를 기준으로 유도되었기 때문에 타원체 형태의 어안렌즈에 대해서는 맞지 않는다. 본 논문에서는 어안렌즈를 타원체로 모델링하여 기하학적으로 보정하는 방법을 제안한다. 보정한 결과를 통해 제안한 방법이 타당한 것을 확인할 수 있다. General method for correcting the distortion caused by the characteristic of the fish-eye lens may be classified in two ways. The first method is a calibration method using a mathematical model taking into account the characteristics of the lens, the second method is a method using only the distortion correction image, regardless of the lens. When considering the characteristics of the lens, calibration equation can be calculated geometrically from the relationship between the three-dimensional real-world coordinates and two-dimensional image coordinates and the parameters of lens. However, it is not suitable for ellipsoid type lens, because of existing research papers have been corrected on the spherical-type fisheye lens. In this paper, we propose a method for correcting geometrically using fish-eye lens as an ellipsoid model. Through a calibration picture, it can be seen that the proposed method is valid.

A boundary condition based algorithm for locating construction site objects using RFID and GPS

Construction sites are rugged, dynamic, and complex, involving a large number of objects that continuously change their locations and occupied spaces. Being able to accurately locate and track site objects site is critical to project management and control to meet the increasing demands for efficiency and safety in modern construction projects. Automated locating technologies such as Radio Frequency Identification (RFID) have proven to be beneficial in tracking some construction site objects. The biggest challenge in applying the RFID technology is that the received signal strength (RSS) varies over time and location, and there is no direct relationship between signal strength and detection range, leading to low positional accuracies in the estimated tag locations. This paper presents an algorithm termed BConTri that combines a “boundary condition method” and the trilateration concept to estimate tag location in three-dimensional (3D) real world coordinates from four or more RFID readers equipped with GPS. This study also developed a prototype RFID locating system that implemented the newly created BConTri algorithm. A comprehensive assessment of the positional accuracy was applied to field experiment results and a high accuracy of the algorithm was observed. A measure of the spatial dilution of reader distribution was formulated and a linear relationship between this measure and the locating accuracy was observed, forming the base for using this measure as a reliable indicator for the resulting locating accuracy. This accuracy indicator helps in estimating the accuracy and quality control of RFID-based locating and tracking systems in construction.

Refinement of human silhouette segmentation in omni-directional indoor videos

In this paper, we present a methodology for refining the segmentation of human silhouettes in indoor videos acquired by fisheye cameras. This methodology is based on a fisheye camera model that employs a spherical optical element and central projection. The parameters of the camera model are determined only once (during calibration), using the correspondence of a number of user-defined landmarks, both in real world coordinates and on a captured video frame. Subsequently, each pixel of the video frame is inversely mapped to the direction of view in the real world and the relevant data are stored in look-up tables for fast utilization in real-time video processing. The proposed fisheye camera model enables the inference of possible real world positions and conditionally the height and width of a segmented cluster of pixels in the video frame. In this work we utilize the proposed calibrated camera model to achieve a simple geometric reasoning that corrects gaps and mistakes of the human figure segmentation, detects segmented human silhouettes inside and outside the room and rejects segmentation that corresponds to non-human activity. Unique labels are assigned to each refined silhouette, according to their estimated real world position and appearance and the trajectory of each silhouette in real world coordinates is estimated. Experimental results are presented for a number of video sequences, in which the number of false positive pixels (regarding human silhouette segmentation) is substantially reduced as a result of the application of the proposed geometry-based segmentation refinement.

Contact-free and pose-invariant hand-biometric-based personal identification system using RGB and depth data

Hand-biometric-based personal identification is considered to be an effective method for automatic recognition. However, existing systems require strict constraints during data acquisition, such as costly devices, specified postures, simple background, and stable illumination. In this paper, a contactless personal identification system is proposed based on matching hand geometry features and color features. An inexpensive Kinect sensor is used to acquire depth and color images of the hand. During image acquisition, no pegs or surfaces are used to constrain hand position or posture. We segment the hand from the background through depth images through a process which is insensitive to illumination and background. Then finger orientations and landmark points, like finger tips or finger valleys, are obtained by geodesic hand contour analysis. Geometric features are extracted from depth images and palmprint features from intensity images. In previous systems, hand features like finger length and width are normalized, which results in the loss of the original geometric features. In our system, we transform 2D image points into real world coordinates, so that the geometric features remain invariant to distance and perspective effects. Extensive experiments demonstrate that the proposed hand-biometric-based personal identification system is effective and robust in various practical situations.

키넥트 센서를 이용한 인공표식 기반의 위치결정 시스템

In this paper, a landmark based localization system using a Kinect sensor is proposed and evaluated with the implemented system for precise and autonomous navigation of low cost robots. The proposed localization method finds the positions of landmark on the image plane and the depth value using color and depth images. The coordinates transforms are defined using the depth value. Using coordinate transformation, the position in the image plane is transformed to the position in the body frame. The ranges between the landmarks and the Kinect sensor are the norm of the landmark positions in body frame. The Kinect sensor position is computed using the tri-lateral whose inputs are the ranges and the known landmark positions. In addition, a new matching method using the pin hole model is proposed to reduce the mismatch between depth and color images. Furthermore, a height error compensation method using the relationship between the body frame and real world coordinates is proposed to reduce the effect of wrong leveling. The error analysis are also given to find out the effect of focal length, principal point and depth value to the range. The experiments using 2D bar code with the implemented system show that the position with less than 3cm error is obtained in enclosed space(3,500mm×3,000mm×2,500mm)

Biomechanical evolution of the Tkachev on uneven bars in female gymnastics

The development of joint kinematics and kinetics is fundamental to the successful performance of complex flight skills in gymnastics bar routines. Biomechanical understanding of these skills can provide coaches and scientists with key information to make training safe and effective. The Tkachev is a complex and popular gymnastics skill with many different variations. Recently, a new version has been performed, which has become popular with elite female performers. This study examined the key biomechanical characteristics of this skill and contrasted these to the earlier versions reported. Elite female gymnasts (n = 5) were recorded and manually digitised using twin video cameras (50 Hz) at the 2007 World Gymnastics Championships. Three-dimensional (3D) DLT was used to reconstruct the real world coordinates. Individualised inertia characteristics were calculated and used to determine mass centre kinematics. Inverse dynamics analysis was used to calculate joint kinetics at the hips and shoulders from the known values at the toes. The results of this study showed an increased flight time and rotational capacity during the aerial phase for the toe-on Tkachev, as well as a more simple movement pattern and joint kinetic demand with single power impulses at the hips and shoulders compared with previous versions. The key finding of this study was that the toe-on version appeared to be less physically demanding than that the inward and outward techniques, and provide the opportunity to perform more complex aerial phase body positions. These results can help coaches to physically prepare their gymnasts and biomechanists in terms of understanding the demands of these skills.

Development and optimization of a DSP-based real-time lane detection algorithm on a mobile platform

In this study, image processing-based real-time lane detection, which is one of the significant problems in autonomous vehicle control, is explored. A mobile robot platform is developed for that purpose. The motion of the mobile robot is provided by 4 direct current motors, which are independently controlled. An image processing code is developed in a Visual DSP 5.0 environment and run on a BF-561 processor embedded in the ADSP BF-561 EZ-KIT LITE evaluation board (Analog Devices). In the image processing algorithm, Hough lines obtained from the Hough transform of the captured images are called candidate lane marks. Various elimination methods are implemented on these candidate lane marks to detect the actual lane marks. Once the actual lane marks are determined, the real-world coordinates of these lane marks are computed using inverse perspective mapping. The heading angle of the mobile robot is then determined based on the position of the lane marks and the center of the mobile robot. The developed mobile robot platform and the lane detection algorithm are tested under various conditions, including dashed lane marks, varying lightening conditions, and the presence of one lane mark only. It is observed that the algorithm performs successfully and detects the lane marks even in the presence of various disturbance effects under these conditions. After the optimization of the developed image processing code, the number of frames processed by the algorithm increases from 3/s to 30/s, which should be satisfactory for real-time applications.

Semiautomated Tool for Extraction of Microlevel Traffic Data from Videographic Survey

Real-world traffic data collection, extraction, and analysis at a microscopic level is necessary for modeling various traffic and transportation phenomena. In developing countries such as India, where the traffic is characterized by different vehicle types and non-lane-based vehicle movement, an effective tool for vehicle data extraction and analysis is necessary. Development of a fully automated software tool for vehicle trajectory extraction is practically impossible because of the weak lane-based movement of vehicles and occlusion. For addressing these issues, a semiautomated tool, a traffic data extractor, was developed to extract the details of vehicle movements with good accuracy (0.1 s). A videographic survey followed by a camera calibration technique was used to obtain the real-world coordinates from two-dimensional image coordinates. The concept of a vanishing point–based camera calibration technique was used in this tool. The steps involved in handling the developed tool and its applications in microscopic traffic data extraction and analysis are presented.

A New Nasopharyngeal Dynamic Reference Frame Improves Accuracy in Navigated Skull Base Targets

We questioned whether the position of the dynamic reference frame (DRF) influences the application accuracy in electromagnetically navigated cranial procedures. A carrier for an electromagnetic DRF was developed, which could be fixed at the posterior edge of the vomer near the center of the head. This nasopharyngeal DRF was compared with a standard DRF fixed to the surface of the forehead. Image coordinates and real-world coordinates were co-registered and the total target error (TTE) was measured in the frontal and the lateral skull base of formalin fixed human head. At each anatomical site, 10 targets served for TTE determinations and 5 different fiducial combinations were used for registration. With the nasopharyngeal DRF, lower TTE values (2.8 ± 1.4 mm; mean ± SD) were observed when compared with the forehead DRF (3.7 ± 2.8 mm; P = .004). TTEs of both anatomical sites investigated were significantly lower when using the nasopharyngeal DRF (frontal skull base 3.4 vs 2.1 mm, P = .005 and lateral skull base 3.9 vs 3.5 mm, P = .013) than with the standard forehead mounted one. Positioning the DRF in the center of the head significantly improved the application accuracy of targets in the skull base with electromagnetic navigation by 25%.

UAV-BASED PHOTOGRAMMETRIC POINT CLOUDS – TREE STEM MAPPING IN OPEN STANDS IN COMPARISON TO TERRESTRIAL LASER SCANNER POINT CLOUDS

Abstract. In both ecology and forestry, there is a high demand for structural information of forest stands. Forest structures, due to their heterogeneity and density, are often difficult to assess. Hence, a variety of technologies are being applied to account for this "difficult to come by" information. Common techniques are aerial images or ground- and airborne-Lidar. In the present study we evaluate the potential use of unmanned aerial vehicles (UAVs) as a platform for tree stem detection in open stands. A flight campaign over a test site near Freiburg, Germany covering a target area of 120 × 75 [m2] was conducted. The dominant tree species of the site is oak (quercus robur) with almost no understory growth. Over 1000 images with a tilt angle of 45° were shot. The flight pattern applied consisted of two antipodal staggered flight routes at a height of 55 [m] above the ground. We used a Panasonic G3 consumer camera equipped with a 14–42 [mm] standard lens and a 16.6 megapixel sensor. The data collection took place in leaf-off state in April 2013. The area was prepared with artificial ground control points for transformation of the structure-from-motion (SFM) point cloud into real world coordinates. After processing, the results were compared with a terrestrial laser scanner (TLS) point cloud of the same area. In the 0.9 [ha] test area, 102 individual trees above 7 [cm] diameter at breast height were located on in the TLS-cloud. We chose the software CMVS/PMVS-2 since its algorithms are developed with focus on dense reconstruction. The processing chain for the UAV-acquired images consists of six steps: a. cleaning the data: removing of blurry, under- or over exposed and off-site images; b. applying the SIFT operator [Lowe, 2004]; c. image matching; d. bundle adjustment; e. clustering; and f. dense reconstruction. In total, 73 stems were considered as reconstructed and located within one meter of the reference trees. In general stems were far less accurate and complete as in the TLS-point cloud. Only few stems were considered to be fully reconstructed. From the comparison of reconstruction achievement with respect to height above ground, we can state that reconstruction accuracy decreased in the crown layer of the stand. In addition we were cutting 50 [cm] slices in z-direction and applied a robust cylinder fit to the stem slices. Radii of the TLS-cloud and the SFM-cloud surprisingly correlated well with a Pearson's correlation coefficient of r = 0.696. This first study showed promising results for UAV-based forest structure modelling. Yet, there is a demand for additional research with regard to vegetation stages, flight pattern, processing setup and the utilisation of spectral information.

Mapping ideas from cyberspace to realspace: visualizing the spatial context of keywords from web page search results

We introduce a new method for visualizing and analyzing information landscapes of ideas and events posted on public web pages through customized web-search engines and keywords. This research integrates GIScience and web-search engines to track and analyze public web pages and their web contents with associated spatial relationships. Web pages searched by clusters of keywords were mapped with real-world coordinates (by geolocating their Internet Protocol addresses). The resulting maps represent web information landscapes consisting of hundreds of populated web pages searched by selected keywords. By creating a Spatial Web Automatic Reasoning and Mapping System prototype, researchers can visualize the spread of web pages associated with specific keywords, concepts, ideas, or news over time and space. These maps may reveal important spatial relationships and spatial context associated with selected keywords. This approach may provide a new research direction for geographers to study the diffusion of human thought and ideas. A better understanding of the spatial and temporal dynamics of the ‘collective thinking of human beings’ over the Internet may help us understand various innovation diffusion processes, human behaviors, and social movements around the world.

Real-Time Navigation for a Humanoid Robot Using Particle Filter

This paper presents a practical real-time visual navigation system, including a vision system, a particle filter (PF) based localization system, and a path planning system, for humanoid robots in an indoor environment. A neural network (NN) converter system is used to solve the image distortion problem. The monocular vision system detects objects of interest in the scene, calculating their position in the image, and converting the position in the image to real world coordinates. The PF localization system estimates the current position by the robot’s motion model and corrects the estimated position by using feedback from the data gathered by the vision system. The path planning system determines the next motion based on the result of the localization system. This paper uses a tree-like path planning method which not only guides the robot to the destination but also avoids obstacles at the same time. The navigation method allows a user to assign several different target destinations to the robot simultaneously. The proposed method is implemented on a humanoid robot “ROBOTIS DARwIn-OP”, an open platform humanoid robot. The effectiveness of the system is demonstrated in an empirical evaluation.

Remote sensing of wave runup over breakwater slope in field using optical imagery

Yoo, J., Choi, J.-Y., Yoon, J.-J., Jun, K.-C., Shim, J.-S. Kim, S. and Min, I.-K., 2013. Measurements of wave runup over breakwater observed in field using optical imagery.In this paper, wave run-up on the seaward slope of a breakwater in field was measured remotely using optical video imagery and investigated through comparisons with an empirical formula. In order to extract properties of wave run-up and over-topping on the breakwater slope from field videos, individual image frames were sequentially sampled at 3 Hz from a video burst first. Then, the sampled sequential image frames were rectified into real world coordinates. A crossshore image timestack was generated by collecting an image pixel array, which was defined along the cross-shore transect on the seaward breakwater face of interest. Trajectory of the run-up on the seaward face evolving in time was extracted from the image timestack using a line detection method. The properties of the run-up estimated from field video bursts were combined to the measurements of incident waves, in order to derive an empirical relation between incident waves and run-up. The video-based relative wave run-up height (i.e. Ru,2% / Hs) at a typical armored breakwater with tetrapods was found to be about 2.15. This value is about in the range of the general formula, which was derived from laboratory experiments in earlier studies.

The Accuracy of GIS Tools for Transforming Assumed Total Station Surveys to Real World Coordinates

Most surveying works for mapping or GIS applications are performed with total station. Due to the remote nature of many of the sites surveyed, the surveys are often done in unprojected, local, assumed coordinate systems. However, without the survey data projected in real world coordinates, the range of possible analyses is limited and the value of existing imagery, elevation models, and hydrologic layers cannot be exploited. This requires a transformation from the local assumed to the real world coordinate systems. There are various built-in and add-in tools to perform transformations through GIS programs. This paper studies the effect of using Georeferencing tool, Spatial Adjustment tool (Affine and similarity) and CHaMP tool on the precision and relative accuracy of total station survey. This transformation requires real-world coordinates of at least two control points, which can be collected from different sources. This paper also studies the effect of using geodetic GPS, hand-held GPS, Google Earth (GE) and Bing Basemaps as sources for control points on the precision and relative accuracy of total station survey. These effects have been tested by using 111 points covered area of 60,000 m2and the results have shown that the CHaMP tool is the best for preserving the relative accuracy of the transformed points. The Georeferencing and spatial adjustment (similarity) tools give the same results and their accuracy are between 1/1000 and 1/300 depending on the source of control points. The results have also shown that the cornerstone to preserve the precision and relative accuracy of the transformed coordinates is the relative position of the control points despite their source.

Interpretation of Medical Imaging Data with a Mobile Application: A Mobile Digital Imaging Processing Environment

Digital Imaging Processing (DIP) requires data extraction and output from a visualization tool to be consistent. Data handling and transmission between the server and a user is a systematic process in service interpretation. The use of integrated medical services for management and viewing of imaging data in combination with a mobile visualization tool can be greatly facilitated by data analysis and interpretation. This paper presents an integrated mobile application and DIP service, called M-DIP. The objective of the system is to (1) automate the direct data tiling, conversion, pre-tiling of brain images from Medical Imaging NetCDF (MINC), Neuroimaging Informatics Technology Initiative (NIFTI) to RAW formats; (2) speed up querying of imaging measurement; and (3) display high-level of images with three dimensions in real world coordinates. In addition, M-DIP provides the ability to work on a mobile or tablet device without any software installation using web-based protocols. M-DIP implements three levels of architecture with a relational middle-layer database, a stand-alone DIP server, and a mobile application logic middle level realizing user interpretation for direct querying and communication. This imaging software has the ability to display biological imaging data at multiple zoom levels and to increase its quality to meet users’ expectations. Interpretation of bioimaging data is facilitated by an interface analogous to online mapping services using real world coordinate browsing. This allows mobile devices to display multiple datasets simultaneously from a remote site. M-DIP can be used as a measurement repository that can be accessed by any network environment, such as a portable mobile or tablet device. In addition, this system and combination with mobile applications are establishing a virtualization tool in the neuroinformatics field to speed interpretation services.

Scale and Rotation Invariant Two View Microgripper Detection that Uses a Planar Pattern

In automated grasping of microparts or objects with unknown dimensions and orientations, at least two cameras have to be used to acquire the depth information. In addition to recognition and reconstruction of the real-world coordinates of the target objects, the system has to be able to detect also the real-world coordinates of the microgrippers from the images. This paper presents a scale and rotation invariant microgripper detection method that uses a planar pattern. The method is suitable especially for prototyping systems, whose composition might vary between the experiments. The gripper detection is shown to be accurate enough for challenging micromanipulation tasks of small electronic components and individual paper fibers.

A New Monoplotting Tool to Extract Georeferenced Vector Data and Orthorectified Raster Data from Oblique Non-Metric Photographs

Since its invention in the first half of the nineteenth century, photography has assumed a leading role as a means for documenting the real world. With the improvement of technology, photography developed into photogrammetry, enabling the mapping and georeferencing of landscape elements beginning with stereo photographs. With the introduction of aerial photography, terrestrial oblique photography became obsolete for cartographic purposes and was nearly forgotten by most specialists in photogrammetry. In recent times, the improvement of computing power and the production of high resolution Digital Elevation Models has made the spatial georeferincing of single oblique pictures (monoplotting) more approachable. In this paper, we focus on a new monoplotting tool developed by our research group and we illustrate the basic concept, the solutions implemented and options as well as the results of a case study on land-use and vegetation evolution over a 100-year period. The tool has been conceived to georeference ordinary individual photographs in order to orthorectify the visible landscape or to produce and export map layers (e.g. georeferenced vector data) by drawing them directly on these pictures. The basic requirements of the system are the digital version of the historical picture, the DEM of the depicted landscape, the real-world coordinates of a suitable number of control points unambiguously recognizable on the picture, and – suitable but not mandatory – the real-world coordinates of the precise shooting point and of the centre of the picture.

A simple, interactive GIS tool for transforming assumed total station surveys to real world coordinates – the CHaMP transformation tool

Increasingly, geoscientists and biologists are monitoring the natural environment with total station and terrestrial laser scanning surveys. Due to the remote nature of many of the sites monitored (e.g., streams, rivers, glaciers, etc.) the surveys are often done in unprojected, Cartesian, local, assumed coordinate systems. However, without the survey data projected in real world coordinates the range of possible analyses is limited and the contextual power of existing imagery, elevation models, and hydrologic layers can not be exploited. This requires a transformation from the local assumed to the real world coordinate systems. We present a simple interactive interface, as an ArcGIS Add-In, that allows a user to transform unprojected total station data into real-world coordinates using three benchmark coordinates, which can be collected from a hand-held GPS (available at http://ctt.joewheaton.org/). Unlike most transformations built into GIS programs, our tool uses an affine transformation (simple shift and rotate) to preserve the precision and relative accuracy of the total station survey, while leveraging the absolute positional accuracy of the hand-held GPS to place one's data approximately in real world coordinates for GIS overlay purposes. The user can quickly visually inspect between six and twelve transformation options, while comparing the residual error estimates to interactively choose the most reasonable transformation. The tool provides an easy-to-use, cost-effective workflow, which facilitates the sharing and visualization of precise total station survey data in real world coordinates through a webGIS or virtual globes (e.g., Google Earth, NASA Whirlwind). The tool has been tested and was used by 12 crews to transform topographic total station surveys of 364 sites into real world coordinates as part of the Columbian Habitat Monitoring Program (CHaMP).