Tech Briefs

Camera Images Hydrogen Fires in Three Wavelength Bands

The camera filters and processing can be customized for other multispectral imaging applications.

A special-purpose multispectral video camera has been designed to provide an enhanced capability for viewing hydrogen fires. Hydrogen fires do not emit sufficient visible light to be seen by the unaided human eye, but they do emit strongly at other wavelengths — especially in the infrared and near-infrared portions of the spectrum. Therefore, like some other video cameras developed previously for the same purpose, this camera is designed to respond to infrared light emitted by hot water molecules in hydrogen flames. Going beyond previous designs, this camera provides a combination of imaging in three wavelength bands and processing of the three images, all for the purposes of (1) reducing spurious responses to background light and solar radiation, and (2) synthesizing an image of a hydrogen flame overlaid on an ordinary visible-light image of the scene that contains the flame.

The camera includes custom optics similar to the color-separation prisms found in broadcast video cameras, but operating in different spectral regions. Instead of separating the red-green-blue light, this camera separates the incoming light into visible and two infrared channels. The incoming light is first focused by a commercial electronic news gathering (ENG) motorized zoom lens assembly, then passed through the prism to separate it into three spectral images.

Image
Figure 1. A Multispectral Camera represented here by a block diagram uses filtering and thresholding algorithms to produce a visible image of a typically transparent hydrogen fire.
The operation of the camera is illustrated in Figure 1. Incoming light passes through the zoom lens and beam-splitting prism. The visible wavelengths are routed to the color charge-coupled device (CCD) to acquire a standard color video image. The near infrared light is split into two different channels, one to image wavelengths corresponding to the flame emissions and another to image a near-infrared wave band that omits the flame emissions. The spectral content of the light arriving at the CCD sensors is selected by narrow-band filters placed in front of the imaging arrays.

The light arriving at the CCD array sensors is transformed into electronic image data. The outputs of the CCDs are digitized and processed using proprietary algorithms programmed into an Altera programmable logic device. The background infrared (IR) image is then subtracted from the flame IR image. Filtering and thresholding algorithms isolate the flame pixels. The resultant isolated flame image is superimposed on the visible image, producing a red overlay on the visible image, denoting the location and size of the hydrogen flames. A digital-to-video encoder provides a standard NTSC (National Television Systems Committee) and S-video output. Multiple modes provide output of the color, flame IR, background IR, or overlay image. Figure 2 shows the overlay mode image with the flame represented in red. The inset shows a standard color video image for comparison. An additional output mode cycles among color, flame IR, and overlay images at a fixed interval. User control of the output mode is available via remote-control input or a switch on the rear of the camera.

Image
Figure 2. The Processed Image of a Hydrogen Flame is superimposed in red against the background, making it clearly visible, whereas the standard color video image of the same scene in the inset does not reveal the flame.
With an eye towards other types of future application, the camera was designed with a number of features to accommodate customized filtering and image processing. Alignment features built into each array channel allow image registration of the three images to less than one pixel accuracy. This allows the camera to be easily realigned after disassembly for filter replacement. The camera optical mounts are fabricated from low-expansion alloys to minimize temperature sensitivity. The onboard image processing is implemented with a programmable logic device that can be customized for special applications. Future versions of the camera will provide digital output from the three CCD arrays to provide a direct interface to computer systems for additional image processing.

This work was done by David B. Duncan, Greg Leeson, and Sherwood Kantor of Duncan Technologies, Inc., for Stennis Space Center.

In accordance with Public Law 96-517, the contractor has elected to retain title to this invention. Inquiries concerning rights for its commercial use should be addressed to
David B. Duncan
Duncan Technologies, Inc.
11824 Kemper Road
Auburn, CA 95603

Refer to SSC-00056, volume and number of this NASA Tech Briefs issue, and the page number.