A scanning optoelectronic instrument generates the digital equivalent of a three-dimensional (X,Y,Z) map of a surface that spans an area with resolution on the order of 0.005 in. (≈0.125mm). Originally intended for characterizing surface flaws (e.g., pits) on space-shuttle thermal-insulation tiles, the instrument could just as well be used for similar purposes in other settings in which there are requirements to inspect the surfaces of many objects. While many commercial instruments can perform this surface-inspection function, the present instrument offers a unique combination of capabilities not available in commercial instruments.

Figure 1. The Camera Observes the Line of Light projected on the surface of a plate by a tilted laser. Any deviation of the line from its nominal position is indicative of a hole or bump on the surface.
This instrument utilizes a laser triangulation method that has been described previously in NASA Tech Briefs in connection with simpler related instruments used for different purposes. The instrument includes a sensor head comprising a monochrome electronic camera and two lasers. The camera is a high-resolution unit with digital output. The sensor head is mounted on a computer-controlled, servomotor-actuated translation stage at a fixed height above the nominal X,Y plane. Scanning is effected by using the translation stage to position the sensor head repeatedly at small, equal increments of Y until the entire surface has been traversed in the Y dimension.

Figure 2. The Electronic System of the Instrument includes an onboard processor that generates 3D data in real time, making it possible to display images of the scanned surface in real time.
Figure 1 depicts the basic optical layout for the laser triangulation. The camera is aimed downward (in the –Z direction). Each laser is equipped with optics to project an X-oriented line onto the nominal X,Y plane at a nominal Y position, and is tilted at a known angle of incidence. At each incremental position along the scan, the camera records the image of the laser-illuminated line on the surface. The camera is oriented so that pixel rows are X-oriented and pixel columns are Y-oriented.

The X coordinate of each surface point in the image of the line is obtained by direct correspondence between X and the pixel-column number. Any deviation of the laser-illuminated line from its nominal Y position (and, hence, its nominal pixel-row number) indicates a deviation of the surface from the nominal X,Y plane. The image is digitized and the depth (Z) of the surface at each point along the line is calculated from the Y (pixel-row) deviation by use of a standard triangulation equation. The Y position of each point along the line is obtained from a combination of (1) the known Y position along the scan, (2) the aforementioned Y deviation of the illuminated line, and (3) another standard triangulation equation to correct for the effect of Z on the apparent Y position. The process as described thus far is repeated at each increment of position along the scan. The data collected at all the increments of position are assembled to produce a three-dimensional (3D) map of the surface.

Two lasers are used (but not simultaneously) in conjunction with a dual-scan scheme to overcome shadowing at overhangs, edges of steep holes, and the like. As depicted in the lower part of Figure 1, a surface is scanned twice: from left to right and from right to left. During the scan toward the right, the left laser is used; during the scan toward the left, the right laser is used. Both lasers illuminate common areas (typically, a central area at the bottom of a hole), and each laser illuminates an edge area that may be shaded from the other laser. Surface points in the hole that may be shaded from the left laser during the rightward scan are illuminated by the right laser during the leftward scan, and vice versa.

Figure 2 is a block diagram of the electronic system of the instrument. The system includes an onboard processor, plus an external personal computer (PC) for further processing of the acquired data and displaying resulting depth maps. The processor is capable of generating 3D data in real time, eliminating the need for both onboard memory and post-processing to generate 3D data. The 3D data output of the onboard processor is sent to the PC via a high-speed serial data-communication link. By reducing the computational burden on the PC, onboard preprocessing enables the PC to create and display 3D images in real time during scanning.

This work was done by Joseph Lavelle and Stefan Schuet of Ames Research Center.

This invention is owned by NASA and a patent application has been filed. Inquiries concerning rights for the commercial use of this invention should be addressed to

the Ames Technology Partnerships Division at (650) 604-2954.

Refer to ARC-14652-1.

Photonics Tech Briefs Magazine

This article first appeared in the November, 2006 issue of Photonics Tech Briefs Magazine.

Read more articles from the archives here.