Area-based image matching and sub-pixel displacement estimation using similarity measures are common methods that are used in various fields. Sub-pixel estimation using parabola fitting over three points with their similarity measures is also a common method to increase the matching resolution. However, few investigations or studies have explored the characteristics of this estimation.

This study analyzed sub-pixel estimation error using two different types of matching model. Our analysis demonstrates that the estimation contains a systematic error depending on image characteristics, the similarity function, and the fitting function. This error causes some inherently problematic phenomena such as the so-called pixel-locking effect, by which the estimated positions tend to be biased toward integer values. We also show that there are good combinations of the similarity functions and fitting functions.

In addition, we propose a new algorithm to greatly reduce sub-pixel estimation error. This method is independent of the similarity measure and the fitting function. Moreover, it is quite simple to implement. The advantage of our novel method is confirmed through experiments using different types of images.

Sub-Pixel Estimation Error Cancellation on Area-Based Matching

A system for the automatic reconstruction of real-world objects from multiple uncalibrated camera views is presented. The camera position and orientation for all views, the 3-D shape of the rigid object, as well as the associated color information, are recovered from the image sequence. The system proceeds in four steps. First, the internal camera parameters describing the imaging geometry are calibrated using a reference object. Second, an initial 3-D description of the object is computed from two views. This model information is then used in a third step to estimate the camera positions for all available views using a novel linear 3-D motion and shape estimation algorithm. The main feature of this third step is the simultaneous estimation of 3-D camera-motion parameters and object shape refinement with respect to the initial 3-D model. The initial 3-D shape model exhibits only a few degrees of freedom and the object shape refinement is defined as flexible deformation of the initial shape model. Our formulation of the shape deformation allows the object texture to slide on the surface, which differs from traditional flexible body modeling. This novel combined shape and motion estimation using sliding texture considerably improves the calibration data of the individual views in comparison to fixed-shape model based camera-motion estimation. Since the shape model used for model based camera-motion estimation is only approximate, a volumetric 3-D reconstruction process is initiated in the fourth step that combines the information from ail views simultaneously. The recovered object consists of a set of voxels with associated color information that describes even fine structures and details of the object. New views of the object can be rendered from the recovered 3-D model, which has potential applications in virtual reality or multimedia systems and the emerging field of video coding using 3-D scene models.

/pdf/automatic-reconstruction-of-stationary-3-d-objects-from-jvewu59hjq.pdf

Automatic reconstruction of stationary 3-D objects from multiple uncalibrated camera views

The paper presents an Image-based Method for Crack Analysis (IMCA) which is capable of processing a sequence of digital imagery to perform a twofold task: (i) the extraction of crack borders and the evaluation of its width across the longitudinal profile; (ii) the measurement of crack deformations (width, sliding and rotation). Here both problems are solved in 2-D, but an extension to 3-D is also addressed. The equipment needed to apply the method is made up of a digital camera (or a still video-camera in case a high frequency in data acquisition is necessary), an orientation frame which establishes the object reference system, a pair of signalized supports to be placed in a permanent way on both sides of the crack to compute deformations; however, permanent targets are mandatory only for case (ii). The measurement process is carried out in a fully automatic way, a fact also that makes this technique highly operational for unskilled people in engineering surveying or photogrammetry. The accuracy of the proposed method, evaluated in experimental tests adopting different consumer digital cameras, is about ± 5–20 μm, like the accuracy of most deformometers, but with the advantage of automation and of augmented achievable information; moreover, the image sequence can be archived and off-line measurements could be performed at any time.

Crack measurement: Development, testing and applications of an automatic image-based algorithm

In this paper we address external calibration of distributed multi-camera system intended for tracking and observing. We present a robust and efficient method for wide area calibration using virtual calibration object created by two LED markers. Our algorithm does not require for all the cameras to share common volume; only pairwise overlap is required. We assume the cameras are internally calibrated prior to deployment. Calibration is performed by waiving the calibration bar over the camera coverage area. The initial pose of the cameras is calculated using essential matrix decompositions. Global calibration is solved by automatically constructing weighted vision graph and finding optimal transformation paths between the cameras. In the optimization process, we introduce novel parametrization for two-point calibration using direction normal. The results are increased accuracy and robustness of the method under the presence of noise. In the paper, we present experimental results on a synthetic and real camera setup. We have performed image noise analysis on a synthetic wide-area setup of 5 cameras. Finally, we present the results obtained on a real setup with 12 cameras. The results obtained on the real camera setup show that our approach compensates for error propagation when the path transformation includes two to three nodes. No significant difference in reprojection error was found between the cameras on non-direct and direct path of the vision graph. The mean reprojection error for the real cameras was below 0.4 pixels.

Wide-area external multi-camera calibration using vision graphs and virtual calibration object

The star tracker is one of the most promising attitude measurement devices used in spacecraft due to its extremely high accuracy. However, high dynamic performance is still one of its constraints. Smearing appears, making it more difficult to distinguish the energy dispersive star point from the noise. An effective star acquisition approach for motion-blurred star image is proposed in this work. The correlation filter and mathematical morphology algorithm is combined to enhance the signal energy and evaluate slowly varying background noise. The star point can be separated from most types of noise in this manner, making extraction and recognition easier. Partial image differentiation is then utilized to obtain the motion parameters from only one image of the star tracker based on the above process. Considering the motion model, the reference window is adopted to perform centroid determination. Star acquisition results of real on-orbit star images and laboratory validation experiments demonstrate that the method described in this work is effective and the dynamic performance of the star tracker could be improved along with more identified stars and guaranteed position accuracy of the star point.

Motion-blurred star acquisition method of the star tracker under high dynamic conditions

Signalizing points of interest on the object to be measured is a reliable and common method of achieving optimum target location accuracy for many high precision measurement tasks. In photogrammetric metrology, images of the targets originate from photographs and CCD cameras. Regardless of whether the photographs are scanned or the digital images are captured directly, the overall accuracy of the technique is partly dependent on the precise and accurate location of the target images. However, it is often not clear which technique to choose for a particular task, or what are the significant sources of error. The research described in this paper describes aspects of target recognition, thresholding, and location. The results of a series of simulation experiments are used to analyze the performance of subpixel target location techniques such as: centroiding; Gaussian shape fitting; and ellipse fitting, under varying conditions.© (1994) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

/pdf/comparison-of-some-techniques-for-the-subpixel-location-of-3qf3w8c6d9.pdf

Tim Short

Papers

Comparison of some techniques for the subpixel location of discrete target images