Summary This paper describes a PC-based imaging workstation that PC-based imagingworkstation that uses a video camera and appropriate computer hardware andsoftware to digitize, display, and store images of fluid flow patterns in realtime. The system includes image-processing software so that the digital imagescan be quantitatively analyzed. Introduction Flow-visualization experiments have been used for many years to study fluidflow phenomena. For example, flow-visualization phenomena. For example, flow-visualization studies in EOR research have greatly increased ourunderstanding of the phenomenon of viscous fingering. In these phenomenon ofviscous fingering. In these studies, the fluid flow fields were made visiblewith transparent models and colored fluids or by exploiting the contrast in therefractive indices of the fluids and the models. An important step in a flow-visualization experiment is the acquisition ofpermanent records of the fluid flow field for subsequent study. In the abovestudies, permanent records of the visualized flow were in the form ofphotographs, videotapes, or movies. Unfortunately, images recorded onphotographs or on film are not easily amenable photographs or on film are noteasily amenable to quantitative analyses. What is needed is animage-acquisition technique that captures the images in digital form. Suchimages can then be easily analyzed with computer image-processing techniques. The availability of low-cost, high-resolution, microcomputer graphics deviceshas now made such a digital image-acquisition system affordable. The objective of this study was to develop a low-cost, high-resolution, microcomputer-based imaging workstation for the acquisition of digital imagesin flowvisualization studies. This paper describes our imaging workstation andpresents example images of displacement phenomena obtained with it. Hardware Fig. 1 shows the major components of the flow-visualization imagingworkstation. The imaging system consists of a video camera, a transparent modelpacked with a porous medium, and a microcomputer outfitted with animage-digitizing board, a high-resolution graphics-adapter board, a colorgraphics monitor, a monochrome monitor, and a graphics printer. Our workstation was configured around an IBM PC-AT TM having 1 megabyterandom access memory and 20 megabytes of hard disk for image storage. The 4-bitimage digitizer is capable of capturing and digitizing images every 0.5 secondsinto 16 gray scales with a resolution of 640 columns by 400 rows of pictureelements (pixels). The 16 gray scales are numbered from 0 (black) to 15(white). The digitizer, which plugs into one of the PC expansion slots, isplugs into one of the PC expansion slots, is an electronic circuit board thataccepts analog signals from a video camera and converts them into digitalimages. To take advantage of the digitizer's high resolution, we used ahigh-resolution graphics board that exactly matches it. The graphics boardaccepts digital data from the computer memory, converts the data to analogsignals, and sends the signals to the graphics monitor for display. We used the IBM color monitor for image display, and to enter commands, weincorporated a monochrome monitor and keyboard into the configuration. Withthis arrangement, we can see the entered commands on the monochrome monitorwhile the processed image is displayed on the color monitor. For hard copies of the images, we used a paint-jet color-graphics printercapable of printing images in 16 colors from a palette printing images in 16colors from a palette of 256 at a resolution of 90 dots per inch. Images can beprinted on paper or on transparency film. Images were captured with a black and white video camera. However, we alsoconfigured the system for a color camera by installing a color-filter board toconvert the color signal into a black-and-white signal to minimize noise. Thevideo camera, which plugs directly into the digitizer board, plugs directlyinto the digitizer board, converts the light intensities from the object to beimaged into analog signals for the digitizer. Our flow-visualization experiments were conducted in a 40 × 40 × 0.32-cm [16x 16 × 0.13-in.] transparent lucite model, packed with 30/35-mesh glass beads. Designed to simulate a quarter five-spot flood pattern, it had a porosity of38% and a permeability of about 20 darcies. The model can be used without theporous medium to study fluid flow in a Hele-Shaw cell. JPT P. 558
Read full abstract