KEYWORDS: Speckle, Speckle pattern, 3D modeling, Cameras, Clouds, Image fusion, Error analysis, High dynamic range imaging, 3D metrology, Fringe analysis
Speckle projection provides rich textures for correlation, and has been widely used in three-dimensional(3D) reconstruction. However, for objects with high dynamic range(HDR) surface, conventional uniform speckle projection usually causes over-exposure and over-dark regions simultaneously in the captured images, leading to miss-match and errors in 3D result. It is difficult to perform defect elimination via adjustment of speckle intensity globally or the camera exposure time. To tackle this problem, this paper proposed a novel adaptive speckle projection method to distinguish the appropriate projection intensity of specific parts in the speckle pattern, thereby avoiding over-exposure, while the dark regions not being affected. First, uniform intensity patterns of multiple gray-levels are projected onto the surface of testing object, the appropriate projection intensity at each pixel position in the camera coordinate system is calculated, and the saturated area in the captured image is marked. Then, a set of orthogonal fringe patterns are projected onto the testing object to establish the coordinates mapping relationship between the camera and the projection system, and the adaptive speckle pattern under the projection coordinate system is generated. Finally, the generated adaptive speckle pattern is used to scan the testing object, and the spatial-temporal correlation algorithm is used for 3D shape retrieval. Experimental results demonstrate feasibility of 3D shape reconstruction of HDR surfaces with the proposed method, and obvious advantages compared with the traditional methods in terms of reconstruction completion and measurement accuracy. Keywords: adaptive speckle, high dynamic range, coordinates mapping, spatial-temporal correlation, 3D reconstruction
To eliminate traces in the overlap area of images caused by splicing, and reconstruct 3D face models with high-precision and photo-realism, this paper presents a multi-view texture fusion method based on triangular patches in 3D face models. The 3D point cloud is acquired according to the multi-view range images of an object and the parallax principle. By performing free registration method, ICP (Iterative Closest Point) method and a series of post-processing algorithm on the point cloud data, we can obtain a complete and non-redundant 3D geometric model with higher precision than counterpart methods. Using the brightness balance algorithm proposed in this paper, the differences of texture images in brightness and color are eliminated. The compound-weight construction algorithm combining with Gauss-Laplace Pyramid method is proposed to fulfill the 3D model texture fusion. In a variety of modeling experiments with our self-developed 3D face acquisition device, the proposed method is demonstrated to be efficient. It outperforms the counterpart methods and has practical significance in real world applications.
The trade-off between accuracy and computational cost is a central aspect of the problem of real-time vehicle detection, especially for resource-limited platforms. We propose a vehicle-detection method based on template aggregate channel features (TACFs) and soft channel cascade to address the computational requirements of a monocular rearview vehicle detection system. The TACFs is a system of template features designed based on the original aggregate channel features, which is aimed at fully describing the characteristics of the vehicle’s appearance. The proposed soft channel cascade is a pipeline of multiple single-channel classifiers and a full-channel classifier, in which the single-channel classifiers are used to quickly filter the background window and reduce the area of the region of interest, and the full-channel classifier is used to verify the candidate window to improve the detection precision. The proposed method is evaluated on three different datasets with a variety of scenarios involving dense traffic, poor lighting, bad weather conditions, and partial occlusion. Detailed evaluations show that the proposed method enables high-precision detection and high recall rate, and performs better than the existing state-of-the-art vehicle detection methods. In addition, the soft channel cascade can rapidly reject most windows that do not contain any vehicles and enables real-time detection.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.