Vision Processing for Real-time 3-D Data Acquisition Based on Coded Structured Light

Abstract

Structured light vision systems have been successfullyused for accurate measurement of 3-D surfaces in computer vi-sion. However, their applications are mainly limited to scanningstationary objects so far since tens of images have to be captured forrecovering one 3-D scene. This paper presents an idea for real-timeacquisition of 3-D surface data by a specially coded vision system.To achieve 3-D measurement for a dynamic scene, the data acqui-sition must be performed with only a single image. A principle ofuniquely color-encoded pattern projection is proposed to design acolor matrix for improving the reconstruction efficiency. The ma-trix is produced by a special code sequence and a number of statetransitions. A color projector is controlled by a computer to gen-erate the desired color patterns in the scene. The unique indexingof the light codes is crucial here for color projection since it is es-sential that each light grid be uniquely identified by incorporatinglocal neighborhoods so that 3-D reconstruction can be performedwith only local analysis of a single image. A scheme is presented todescribe such a vision processing method for fast 3-D data acqui-sition. Practical experimental performance is provided to analyzethe efficiency of the proposed methods. Vision Processing for Real-time 3-D Data Acquisition Based on Coded Structured Light

INTRODUCTION:

COMPUTER vision has become a very important means to obtain the 3-D model of an object. A number of 3-D sensing methods have been explored by researchers in the past 30 years. The structured light has made its progress from single light-spot projection to complex coded pattern, and, consequently, the 3-D scanning operation speeds up from several hours per image to dozens of images per second.

The first stage of feasible structured light systems came in early 1980 when the binary coding or gray coding methods were employed. This kind of pattern can achieve high accuracy in the measurements. This is due to the fact that the pattern resolutions are exponentially increasing among the coarse-to-fine light projections and the stripe gap tends to 0, but the stripe locations are easily distinguishable since a small set of primitives is used, and, therefore, the position of a pixel can be encoded precisely. It also takes the advantage of easy implementation, and, thus, this method is still the most widely used in structured light systems. The main drawback is that they cannot be applied to moving surfaces since multiple patterns must be projected. In order to obtain a better resolution, a technique based on the combination of gray code and phase shifting is often used. Its drawback is that a larger number of projection patterns (e.g., >20images) are required.

With the aim to project only one light pattern before capturing a scene image, color stripes are invented for replacing multiple black/white projections. This idea brings a development of “one-shot” 3-D image acquisition and it is possibly applied in measuring moving objects. People have attempted a lot of such systems for practical implementation, in which a phase-shifting method can also be employed. Among them, the De Bruijn sequences are the mostly used technique. Although these promise real-time applications, limitations of this method are still considerable. One is its tradeoff between reliability and accuracy. Since adjacent color stripes should have enough spectral difference, people have to use a limited number of color stripes or apply them periodically, which produces either stripe ambiguity or rough resolution. Another limitation is the flexibility of its system setup. Since it is a 1-D spatial coding method, the baseline between the camera and the projector should be nearly orthogonal with light planes. It is suitable for setting up a fixed system, but not for some applications where dynamic reconfiguration and recalibration if multiple degrees of freedom are required.

Related Post