WPC 2  BJUurier#|xCourierCourier BoldCG TimesCG Times BoldCourier10cpiterAPPLLASE.PRSx  @hhhhhhX@#|x2 Z<0FHP LaserJet 4M (PCL)HPLA4MPC.PRSx6X@Q\P -X@${X01Í ÍX0Í Íҫx6X@QX@<6X9`("Courier 10cpiX2%:,ZLCourierCourier BoldCourierCourier BoldCG TimesX@`7X@>?xxxx `KXR&HHHH6X@Kh@ !`/ )2V8XX<?xxx,>fx6X@`7X@?xxx,x `7X<R&HHH,D,H6X@`7h@8wC;,=Xw PE37XP<?xxx,>fx6X@`7X@?xxx,x `7XHP LaetHPLASERJ.PRSx  @X@80dddddddddYdddddsdXYXXXYYx|X~Y~Y|XddddddddC8C8C8C8oNd|8|H~8<|8dtddddHHJlLlLlLkN|H|8~8dddddddXXXd~8dJkN~8ddxddCddCCCWxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxNdddCYQQddddddFddddFCChhd44ddzzdddvooChdF"dhd9dCCzCddoddCdYds]zUvdYYCCCCz~ozoY~NYdYC8YooYdYzsdzdd~YYzozzzzNd88YYYzYzzzzCCdddddddzzzzzzzzzzzzzzzzzzzNNNNNNNdddddddddddddddddddd888888888888YYYYYYYYYYYYYYYYYYYzzzzzzzzzzzzzzzzzzzzCs~CzCddYC\   pxtll\tll@\@\`L2 ZZ. t"4|J~ ^;C]ddCCCdCCCCddddddddddCCdxN`xoCCCddCdoYoYFdo8Co8odooYNCodddYdddd4dddddCddddddddo8dddddYYYYYN8N8N8N8oddddooooddpddddxodddXXYYXYYXddddddooL8N8N8N8r`o8P8N8ppoddXXYpLoNpLoNPDCopoopodXYXo8oYoNCdddoCddCCCWxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxNdddCdUUddddddFddddFCCssd44ddzzddd~ooCsdF"dsd9dCCzCddoddCdYds`zUvdddCCCCzozoYNYYYN8YooYdYzzdzddYYzozzzzNY88YYYzYzzzzCCdddddddzzzzzzzzzzzzzzzzzzzNNNNNNNYYYYYYYYYYYYYYYYYYYY888888888888YYYYYYYYYYYYYYYYYYYzzzzzzzzzzzzzzzzzzzzCzNzNddYC\   pxtll\tll@\@\`L<?xxx,>fx6X@`7X@?xxx,x `7X<R&HHH,D,H6X@`7h@58wC;,=Xw PE37XPD7zC;,bXz_ pi7Xtal color image of an origTimes New Roman Bold allowed the use of more versattypeTimes New Roman Bold Italicerging a DRG with a DOQ creates 2Times New 2 ?  3'3'StandardA'LegalA'LegalHPLA4MPC.PRSx6XU  ` `   D MERGING DIGITAL RASTER GRAPHICS AND DIGITAL ORTHOPHOTO QUADRANGLES FOR USE IN DIGITAL MAP REVISION Michael Starbuck and William Helterbrand U.S. Geological Survey \ MidContinent Mapping Center 1400 Independence Road  Rolla, Missouri 65401 &ABSTRACT Revision of digital line graph (DLG) map data has traditionally relied upon digital orthophoto quadrangles (DOQ) as the primary image source. The advent of the digital raster graphic (DRG), a digital color image of an original topographic map, has allowed the use of more versatile types of source images. Digitally merging a DRG with a DOQ creates a unique multipurpose image that can be a very effective tool for map revision. Procedures have been developed to allow the automatic digital merging of DOQ and DRG files into a new digital image that retains nearly all of the original information from both input images. By manipulating the display color map, the user can display the merged image in different ways. The revision operator can view the original DOQ, the original DRG, or any ratio of the combined images. Typically, the operator will view an image with a ratio of 80 percent DOQ to 20 percent DRG. This allows the operator to see all of the DOQ image detail while still seeing cartographic features from the DRG. Recent enhancements to the merging process allow the user to manipulate the percentage of each individual DRG color. The DRG image gives the operator existing information for feature interpretation and amount of feature change, and the DOQ provides accurate feature positional information. Merging the two images in this way gives the revision operator access to both of these source images quickly and at no additional cost in terms of file storage requirements or image loading times. The process of image merging is not limited to DOQ's. Shadedrelief images have been successfully merged with DRG's to provide enhanced backdrop images for other data sets. The U.S. Geological Survey believes this technology is an important component in creating and maintaining the National Spatial Data Infrastructure and desires to promote these concepts and techniques through the use of partnershipsX0Í ÍX1. Í.X$ Any use of trade, product, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the U.S. Government. X1. ÍX1. Í '...` `   ? $ INTRODUCTION The U.S. Geological Survey (USGS) has been creating digital line graphs (DLG) since the mid1970's by digitizing or scanning existing maps. DLG's are vector format digital files that include spatial and attribute information about a specific category of topographic map features. These data can be used to recreate the original graphic map or to perform geographic analysis in a geographic information system (GIS). The early methods for revising DLG's involved thX1. ÍX01Í Íe traditional technique of using stereoscopic or monoscopic compilation to revise the graphic map separates. The revised map separates were then digitized to obtain a revised DLG. By the late 1980's, the USGS had developed methods for generating digital orthophoto quadrangles (DOQ). DOQ's are digital images created by scanning aerial photographs and removing feature displacement caused by topographic relief and the camera. Techniques were developed to perform DLG revision using DOQ's as the source. Specifically, a Revision and Product Generation (RevPG) software system (based on the commercial software package ARC/INFO) was developed. RevPG is essentially a series of ARC Macro Language (AML) routines and C programs that allow an efficient and userfriendly interface to ARC/INFO programs. RevPG operators can view DOQ's on a computer display monitor and trace map features with a cursor. Feature attributes can be applied and various procedures can be performed that lead to the production of revised DLG's. In 1993, for the first time, a standard 7.5 minute topographic map, of Parkridge, Illinois, was revised at the USGS fully by digital means using RevPG. In the early 1990's, the USGS began producing a new geographic data product, the digital raster graphic (DRG). A DRG is a color raster image created by scanning an original topographic map and georeferencing it in the Universal Transverse Mercator coordinate system. The DRG provides lowcost digital map information that can be used by a variety of imaging and mapping software systems, including a GIS. The USGS is working with private companies under innovative partnership agreements to provide DRG's for areas where there is customer demand. It is anticipated that the entire country could have DRG coverage in as little as 3 years. DRG's were recognized at the USGS as a potential new tool in the map revision process. One of the strengths of the DRG is the ease with which it can be combined with other data sets. Of particular interest for map revision is the combination of a DRG and a DOQ. Researchers at the USGS MidContinent Mapping Center have developed procedures and software routines to allow the easy merging of a DRG and DOQ into a single image that effectively retains all the information of both original images. By manipulating a color map, a user can change how a merged image is displayed. The user can view an image showing 100 percent DRG, or 100 percent DOQ, or some percentages in between. The advantage of this type of image is that the user can interpret the DOQ image more easily than without the merged DRG image, given clues by....` `  the DRG symbolization and text. The user has the benefit of using both images without having to keep two large files on a disk.  ?   CREATION OF MERGED IMAGES ă The primary process of revising digital map data at the USGS currently involves the use of Data General Unix workstations running ARC/INFO software and the RevPG interface. Although this paper describes how the merging process works using this particular configuration, it is believed that the general techniques could also be applied to other platforms and software systems. When DOQ's are used as the source for revision, a separate header file must be created by reading the header information in the DOQ itself. ARC/INFO uses this file to georeference the image. DOQ's are created georeferenced to the North American Datum of 1983 (NAD83). Within the header portion of the file, values for the x,y position of the 1,1 pixel (first pixel in the upper left of the image) are available for both the NAD83 and the North American Datum of 1927 (NAD27). To place the image on the NAD27, the NAD27 values for the 1,1 pixel are used in the header file. This results in a simple translation and not a rigorous transformation of the image, but at a scale of 1:24,000 there is little observable error. The pixel in a DRG represents a ground sample distance of 2.4384 meters. Each pixel in a DOQ represents a ground sample distance of 1meter. This difference must be reconciled before the two images are merged. Normally, the operator will choose to subsample the DRG to match the 1 meter resolution of the DOQ. This allows the revision operator to retain the best resolution available for viewing the image and results in a file size equivalent to the DOQ. The distribution format of the DRG is a tagged image file format (TIFF) and that of the DOQ is band interleaved by line. Both images must be converted into ARC/INFO's Grid format using the Imagegrid command. The DRG is an indexed color file, meaning it has an associated color table that is used to define the color to be displayed for each pixel value. A standard DRG has 13 possible index values, or colors. The DOQ is a grayscale image, with 256 possible grayscale values. Because the merging process takes advantage of the image color table, the number of colors in the merged file cannot be greater than 256, the size of an 8bit image color table. ARC/INFO reserves 10 colors for itself, leaving only 246 colors that can be used for images. The idea is to try to account for all the possible combinations when the two images are merged. The DRG had 13 colors and the DOQ has up to 256 colors (grayscale values). The result, 3,328 possible combinations, far exceeds the 246 index values available for use. Because of this, the DOQ is requantized, or "sliced," into a more manageable number of grayscale values. The new number of grayscale values is found by dividing the total number of values available in the color table (246) by the number of DRG colors (13) and rounding down to an integer. (Eq. 1)....` ` Ԍ"n = 246 / 13 = 18` !E(1) The DOQ is then requantized into 18 grayscale values. This does not degrade the image significantly because the human eye cannot discern more than about 18 to 20 grayscale values. The DRG and the requantized DOQ are then merged together. (Eq. 2) !N = DRG * 18 + DOQ` !E(2) 44Where N = index value in the new image 44 DRG = index value in the original DRG 44 18 = number of DOQ grayscale values 44 DOQ = index value in the requantized DOQ The merged file retains all of the information present in both the original DRG and the requantized DOQ. This image can now be displayed in many different ways by creating special color tables. A separate routine creates the output color tables for the merged images. Known values include the number of DRG colors, the number of grayscale values in the requantized DOQ, the red, green, and blue (RGB) values for each DRG color index, and mean and standard deviation of the requantized DOQ gray levels. The values of gray in the requantized DOQ are used to generate a range of saturation/intensity values for each DRG color hue. By maintaining the original DRG RGB ratios, color saturation and intensity are varied but the hue remains unchanged. The user specifies the percentage of each DRG color desired in the displayed image. For example, the user may specify a weighting of 20 percent for each DRG color. The result would be a displayed image with 20 percent DRG and 80 percent DOQ. A weighting in this range is suitable for map revision because it allows the operator to easily discern features in the DOQ and simultaneously to see graphic content in the DRG. For each combination of DRG/DOQ weighting, a new color table must be created; however, the C routine does this in just a few seconds.  ?  | USE OF MERGED IMAGES IN MAP REVISION ă The merged DRG/DOQ image is converted into a TIFF file before being used in the revision process. The color table included in the TIFF is the one the operator uses, so usually an 80/20 ratio is created. Although dynamic changing of the color table, or "color toggling," is possible, in practice a procedure has not yet been devised to do it efficiently in the ARC/INFO environment. Ultimately, it is hoped that a procedure will be devised to allow realtime color toggling or manipulation of individual color percentages by the RevPG operator. Currently the operator must use a static merged image to perform the revision. Use of the 80/20 ratio, or something similar, has proven to be effective for enhancing feature interpretation and collection. Merged images are of special benefit to operators who do not yet have extensive experience in photointerpretation. The hints provided by the....` `  symbolization and text visible in the DRG allow the operator to learn more quickly to interpret certain features in the aerial photographs. In addition to aiding in data revision, merged images provide a new way to collect map features in areas where no existing vector data is available. Rather than having to digitize an existing map and then perform an update from orthophotography, users can now accomplish collection and revision in one operation by using merged images. For some areas and feature categories, this simultaneous collection and revision is faster and cheaper than the twostep process.  ?  !FUTURE DEVELOPMENT ă Another use for merged images is in the creation of shadedrelief maps. Shaded relief images derived from USGS digital elevation models (DEM) can be treated like DOQ's because both are grayscale raster images. The shadedrelief image can be merged with a DRG to produce a visually interesting image that can effectively portray landscape relief. The merging technique can also be used to combine other raster images, such as digital scans of hydrology or geology maps and DOQ's. Combining these images in different ways can provide new and unique perspectives to interpret this information. Hydrologists can overlay results of water basin studies over a merged DRG and shadedrelief image, allowing a better visual interpretation and display of the final data. Ground water maps showing locations of underground contaminants can be scanned and merged with a DOQ to show the relationship of the contaminant and surface features. Investigations are under way to find efficient ways to allow dynamic color manipulation of the displayed image. Realtime adjustment of individual colors and image ratios would further enhance the operator's ability to interpret and collect map features. The primary objective of the USGS is to ensure that the fundamental spatial data requirements of the Nation are met. The use of DRG/DOQ data sets will help meet these requirements. The USGS does not plan to make the merged image available as a separate product. Merging data sets is a service more appropriately left to private industry or the data users themselves. The ARC/INFObased RevPg software used to display the merged DRG/DOQ for revision of DLG's and graphic maps is available to the public on the Internet at or on CDROM by writing to USGS Information Services, Box 25286, Denver Federal Center, Denver CO 80225.