• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.11
• /
• pp.1088-1095
• /
• 2009

In this paper, we describe the reorientation method of distorted road sign by using projection transformation for improving recognition rate of road sign. RSR (Road Sign Recognition) is one of the most important topics for implementing driver assistance in intelligent transportation systems using pattern recognition and vision technology. The RS (Road Sign) includes direction of road or place name, and intersection for obtaining the road information. We acquire input images from mounted camera on vehicle. However, the road signs are often appeared with rotation, skew, and distortion by perspective camera. In order to obtain the correct road sign overcoming these problems, projection transformation is used to transform from 4 points of image coordinate to 4 points of world coordinate. The 4 vertices points are obtained using the trajectory as the distance from the mass center to the boundary of the object. Then, the candidate areas of road sign are transformed from distorted image by using homography transformation matrix. Internal information of reoriented road signs is segmented with arrow and the corresponding indicated place name. Arrow area is the largest labeled one. Also, the number of group of place names equals to that of arrow heads. Characters of the road sign are segmented by using vertical and horizontal histograms, and each character is recognized by using SAD (Sum of Absolute Difference). From the experiments, the proposed method has shown the higher recognition results than the image without reorientation.

• Journal of Biomedical Engineering Research
• /
• v.15 no.3
• /
• pp.275-280
• /
• 1994

In the case of the image reconstruction from unknown projection data such as imaging the object with opaque obstructions, conventional reconstruction algorithms may reconstruct a degraded image. In this paper, a new method for the estimation of the unknown projection data based on known projection data and the bandwidth of projection data is proposed. The proposed method successfully estimates the unknown projection data through iterative transformation between projection space and frequency space using the known projection data and the bandwidth of the projection data. Computer simulation shows that the proposed method significantly improves image quality and convergence behavior over conventional algorithms. In addition, the proposed method is successfully applied to ultrasound attenuation CT using a sponge phantom.

• Asia-pacific Journal of Multimedia Services Convergent with Art, Humanities, and Sociology
• /
• v.7 no.6
• /
• pp.897-906
• /
• 2017

Projection Mapping, also known as Spatial Augmented Reality(SAR) has attracted much attention recently and used for many division, which can augment physical objects with projected various virtual replications. However, conventional approaches towards projection mapping have faced some limitations. Target objects' geometric transformation property does not considered, and movements of flexible objects-like paper are hard to handle, such as folding and bending as natural interaction. Also, precise registration and tracking has been a cumbersome process in the past. While there have been many researches on Projection Mapping on static objects, dynamic projection mapping that can keep tracking of a moving flexible target and aligning the projection at interactive level is still a challenge. Therefore, this paper propose a new method using Unity3D and ARToolkit for high-speed robust tracking and dynamic projection mapping onto non-rigid deforming objects rapidly and interactively. The method consists of four stages, forming cubic bezier surface, process of rendering transformation values, multiple marker recognition and tracking, and webcam real time-lapse imaging. Users can fold, curve, bend and twist to make interaction. This method can achieve three high-quality results. First, the system can detect the strong deformation of objects. Second, it reduces the occlusion error which reduces the misalignment between the target object and the projected video. Lastly, the accuracy and the robustness of this method can make result values to be projected exactly onto the target object in real-time with high-speed and precise transformation tracking.

• Journal of Sensor Science and Technology
• /
• v.28 no.3
• /
• pp.182-186
• /
• 2019

When a camera is employed for 3D sensing, accurate camera calibration is vital as it is a prerequisite for the subsequent steps of the sensing process. Camera calibration is usually performed by complex mathematical modeling and geometric analysis. On the other contrary, data learning using an artificial neural network can establish a transformation relation between the 3D space and the 2D camera image without explicit camera modeling. However, a neural network requires a large amount of accurate data for its learning. A significantly large amount of time and work using a precise system setup is needed to collect extensive data accurately in practice. In this study, we propose a two-step neural calibration method that is effective when only a small amount of learning data is available. In the first step, the camera projection transformation matrix is determined using the limited available data. In the second step, the transformation matrix is used for generating a large amount of synthetic data, and the neural network is trained using the generated data. Results of simulation study have shown that the proposed method as valid and effective.

• Journal of Korean Society for Geospatial Information Science
• /
• v.19 no.4
• /
• pp.55-63
• /
• 2011

Having a wide angle of view, a fisheye lens is useful for obtaining images of the inside wall of a tunnel. A circular fisheye tunnel image can be transformed into a familiar rectangular image by applying the concept of cylindrical projection. This projection transformation causes several types of distortions in the projected image. This paper discusses the distortion on the boundary lines between smoothly curved wall and flat ground. We analyzed the cause of this boundary distortion, developed transformation model, and derived a correction formular. A distortion correction software programmed in Visual C++ applied to projected image. Consequently, boundary-corrected image could be obtained. Research into other distortions of projected image will helpful in obtaining tunnel image that resembles real tunnel from fisheye tunnel image.

• Journal of ILASS-Korea
• /
• v.7 no.4
• /
• pp.9-15
• /
• 2002

Convolution Fourier transformation tomographically reconstructs the spatially resolved spray injection rate from direct measuring cells. Asymmetric sprays generated from a twin-hole air shroud injector are tested with 12 equiangular projections of measurements. For each projection angle, line of sight integrated injection rate was measured at 35 positions with equal spacing measuring cells of 3 mm in width, 100 mm in length, 55 mm in depth and 0.5 mm thickness of separating wall. Interpolated data between the projection angles effectively increase the number of projections, which significantly enhances the signal-to-noise level in the reconstructed data. This modified convolution Fourier transformation scheme predicts well the structure of asymmetric sprays. Comparative study has been made between sprays with and without air shrouding. Tomograhpic reconstruction of injection rate from direct measuring cells obtained can be used to estimate the accuracy of volume fraction of spray from the LDPA tomographic reconstruction.

• Journal of Sensor Science and Technology
• /
• v.11 no.4
• /
• pp.239-246
• /
• 2002

In this paper a new design of multi-layer perceptron(MLP) for camera calibration is proposed. Most existing techniques determine a transformation from 3D spatial points to their image points and camera parameters are tried to be estimated from the transformation. The technique proposed here, on the other hand, determines rays of sight uniquely from image points and parameters are estimated from the relationship using an MLP. By this approach projection and back-projection can be done more straightforwardly. Being based on a geometric model, a network design process becomes less ambiguous, which is a clear merit compared to other neural net based techniques. An MLP designed according to the technique proposed showed fast and stable learning in tests under various conditions.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.34S no.2
• /
• pp.52-62
• /
• 1997

To reconstruct the complete 3-D shape of an object, seveal range images form different viewpoints should be merged into a single model. The process of extraction of the transformation parameters between multiple range views is calle dregistration. In this paper, we propose a new algorithm to find the transformation parameters between multiple range views. Th eproposed algorithm consists of two step: initial estimation and iteratively update the transformation. To guess the initial transformation, we modify the principal axes by considering the projection effect, due to the difference fo viewpoints. Then, the following process is iterated: in order to extract the exact transformation parameters between the range views: For every point of the common region, find the nearest point among the neighborhood of the current corresponding point whose correspondency is defined by the reverse calibration of the range finder. Then, update the transformation to satisfy the new correspondencies. In order to evaluate the performance the proposed registration algorithm, some experiments are performed on real range data, acquired by space encoding range finder. The experimental results show that the proposed initial estimation accelerate the following iterative registration step.

• Communications for Statistical Applications and Methods
• /
• v.15 no.5
• /
• pp.775-781
• /
• 2008

It is developed that a family of the semicircular Laplace distributions for modeling semicircular data by simple projection method. Mathematically it is simple to simulate observations from a semicircular Laplace distribution. We extend it to the l-axial Laplace distribution by a simple transformation for modeling any arc of arbitrary length. Similarly we develop the l-axial log-Laplace distribution based on the log-Laplace distribution. A bivariate version of l-axial Laplace distribution is also developed.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2005.07a
• /
• pp.744-747
• /
• 2005

The exploitable efficiency of UHP is very important in a projection system. We always use lens array for light shaping and energy distribution transformation in conventional projection system. Because of the limitation of arc gap and F/# of elements, the utilize efficiency is always dismal. In this paper we try to bring out a new method to reduce the sparkle formed on the PS converter and enhance the system efficiency.