Liang-Chia Chen

Viewpoint-independent 3D object segmentation for randomly stacked objects using optical object detection

This work proposes a novel approach to segmenting randomly stacked objects in unstructured 3D point clouds, which are acquired by a random-speckle 3D imaging system for the purpose of automated object detection and reconstruction. An innovative algorithm is proposed; it is based on a novel concept of 3D watershed segmentation and the strategies for resolving over-segmentation and under-segmentation problems. Acquired 3D point clouds are first transformed into a corresponding orthogonally projected depth map along the optical imaging axis of the 3D sensor. A 3D watershed algorithm based on the process of distance transformation is then performed to detect the boundary, called the edge dam, between stacked objects and thereby to segment point clouds individually belonging to two stacked objects. Most importantly, an object-matching algorithm is developed to solve the over- and under-segmentation problems that may arise during the watershed segmentation. The feasibility and effectiveness of the method are confirmed experimentally. The results reveal that the proposed method is a fast and effective scheme for the detection and reconstruction of a 3D object in a random stack of such objects. In the experiments, the precision of the segmentation exceeds 95% and the recall exceeds 80%.

Novel 3-D Object Recognition Methodology Employing a Curvature-Based Histogram

In this paper, a new object recognition algorithm employing a curvature-based histogram is presented. Recognition of three-dimensional (3-D) objects using range images remains one of the most challenging problems in 3-D computer vision due to its noisy and cluttered scene characteristics. The key breakthroughs for this problem mainly lie in defining unique features that distinguish the similarity among various 3-D objects. In our approach, an object detection scheme is developed to identify targets underlining an automated search in the range images using an initial process of object segmentation to subdivide all possible objects in the scenes and then applying a process of object recognition based on geometric constraints and a curvature-based histogram for object recognition. The developed method has been verified through experimental tests for its feasibility confirmation.

Automatic object detection employing viewing angle histogram for range images

In this paper, a general scheme for automatic object detection is presented. Classification of three dimensional (3-D) objects using range images remains to be one of the most challenging problems in 3-D computer vision due to its noisy and cluttered scene characteristics. The key breakthroughs for this problem lie mainly in defining unique features that distinguish the similarity among various 3-D objects and developing robust segmentation algorithms that can effectively utilize these defined similarity features. In our approach, the object detection scheme can identify inspecting targets automatically in the range images using an initial process of object segmentation to subdivide all possible objects in the scenes and then applying a process of object classification based on geometric constrains (dimension, point density and surface types) and viewing angle histogram for object classification. The methodology computes the surface normal vector distribution of object model at each viewing angle and aggregates the features into histograms over mesh neighborhoods. These histograms are stored in the database for object searching. The classified objects are finally labeled with the consistent labels by finding the highest histogram matching coefficient according to the object list. The method was verified through some experimental tests for its feasibility confirmation.

A 3-D point clouds scanning and registration methodology for automatic object digitization

Abstract The article presents a robot 3-D scanning system for generation of 3-D point clouds of an object by using multi-view 3-D scanning and novel data registration. Our approach mainly comprises two important elements in the determination of next best probe pose and multiple-view point clouds registration. A novel technique is proposed to register 3-D object scene with overlapped or stacked condition. Under this scenario, conventional registration methods such as the iterative closest point algorithm usually fail to converge to a global minimum when a good initial estimate for image registration does not exist. Our proposed technique uses a 3-D scanner to be mounted on a six degree of freedom-articulated industrial robot. It keeps moving probe continuously in the working range against the object and autonomously varying the probe with various gestures required for complete object scanning and for achieving best 3-D sensing accuracy. The robot scanning path is determined through a proposed algorithm using information from the latest scanning data and registered result of the object. The developed method has been verified through experimental tests for its feasibility test. It confirms that the registration accuracy with one standard deviation can be controlled within 0.323 mm when the objects underlying reconstruction are in range of hundreds of millimeters.

In-situ volumetric topography of IC chips for defect detection using infrared confocal measurement with active structured light

The article presents the development of in-situ integrated circuit (IC) chip defect detection techniques for automated clipping detection by proposing infrared imaging and full-field volumetric topography. IC chip inspection, especially held during or post IC packaging, has become an extremely critical procedure in IC fabrication to assure manufacturing quality and reduce production costs. To address this, in the article, microscopic infrared imaging using an electromagnetic light spectrum that ranges from 0.9 to 1.7 µm is developed to perform volumetric inspection of IC chips, in order to identify important defects such as silicon clipping, cracking or peeling. The main difficulty of infrared (IR) volumetric imaging lies in its poor image contrast, which makes it incapable of achieving reliable inspection, as infrared imaging is sensitive to temperature difference but insensitive to geometric variance of materials, resulting in difficulty detecting and quantifying defects precisely. To overcome this, 3D volumetric topography based on 3D infrared confocal measurement with active structured light, as well as light refractive matching principles, is developed to detect defects the size, shape and position of defects in ICs. The experimental results show that the algorithm is effective and suitable for in-situ defect detection of IC semiconductor packaging. The quality of defect detection, such as measurement repeatability and accuracy, is addressed. Confirmed by the experimental results, the depth measurement resolution can reach up to 0.3 µm, and the depth measurement uncertainty with one standard deviation was verified to be less than 1.0% of the full-scale depth-measuring range.