CENE L. BROTHERS ERNEST B. FISH Texas Tech University Lubbock, TX 79409 Image Enhancement for Vegetative Pattern Change Analysis

Photographic enhancement-overlay processing of aerial imagery gave improved results when compared with traditional change detection techniques.

INTRODUCTION location of change, (3) the magnitude of ATURAL RESOURCES have always been of change, (4) the direction of change, and (5) N vital importance to man, and recent in­ variations in the rates of change (Estes et al., creasing demands on them have progres­ 1972). Time spans between sequential data sively intensified the need for improved collection dictate the degree of accuracy to management (Carneggie and DeCloria, 1973; which a manager can determine these as­ Shepard, 1964). Monitoring changes in ele­ pects ofchange. An effective monitoring sys­ ments of concern to resource managers con­ tem would be one providing relevant, non­ situtes an important phase in the manage- redundant information in an efficient and

ABSTRACT: Valid management decision making in natural resource planning is dependent upon an accurate data base and the ability to detect changes in resource status. Photographic enhancement­ overlay processing of aerial imagery for change detection was de­ veloped and tested. Compared with traditional change detection techniques, the new method gave improved results in terms ofgen­ eral accuracy and the specific nature ofchanges occurring where the objects of interest could be isolated to a specific optical density range. Efficiency of the technique is directly affected by imagery scale and size ofarea to be monitored. ment process. Changes reflect how the envi­ cost effective manner which can be readily ronment is reacting to management practices used with a sustained management infor­ and, thus, provide input for an evaluation of mation system (Paul, 1976; Shepard, 1964). the practices. Change detection of resources has been Most sequential monitoring techniques carried out in various subfields of resource contain the major disadvantage that, unless management in attempts to obtain broad the data are collected while the change is ideas of resource conditions. Studies have actually occurring, the exact time of the shown aerial photography to be superior to change cannot be documented but only other data collection methods because it al­ speculated about within the time span ofthe lows assessment of a complete area and pro­ sequential collection of data (EI-Ashry and vides an accurate recording of surface situa­ Wanless, 1967). However, sequential tions (Collins, 1972; Driscoll, n.d.; Shaklee, monitoring can provide managers with in­ 1976). Studies reviewed here were based on formation on (1) the nature ofchange, (2) the aerial photographic data collection; how-

PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING, 607 Vol. 44, No.5, May 1978, pp. 607-616. 608 PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1978 ever, various analysis and change detection tronic density scanner. Values computed re­ procedures were employed. suit in an "index" which increases as differ­ ences in vegetation on sequential photog­ LITERATURE REVIEW raphy increase (Rouse et al., 1973). Traditional interpretation and mapping Frequently, resource managers are over­ techniques seem to be most frequently used whelmed by the expense and sophistication in studies involving change detection. of some of these change detection tech­ Baseline conditions are interpreted and niques and conclude that they cannot be mapped from imagery which is then com­ of assistance to a particular management pared with information drawn from sequen­ problem. However, demand for rapid collec­ tial imagery of the same area. Measures of tion of information exists, and aerial photog­ the total area covered by a resource of con­ raphy must be considered as a means to cern are taken from both sets of aerial pho­ provide administrators with the required tography. This information is then compared understanding or with the capacity to effi­ to determine change in total area (Avery, ciently acquire an understanding of re­ 1965; Bubenik, 1976; Collins, 1972; Estes et sources (Carneggie, 1970; Poulton, 1970). al., 1972; Richter, 1969; Wray, 1971). Loca­ One of the main limitations of interpreta­ tions of change have been determined from tion is the restricted ability ofthe human eye sequential data by overlaying maps drawn to readily recognize subtle differences in the from the imagery at the same scale or by pro­ diagnostic characteristics of images jection of an image onto a baseline map (Nielsen, 1974). Image enhancement (Milazzo and Lins, 1972; Olson and Meyer, techniques have been developed which re­ 1976; Shaklee, 1976; Wagner, 1963). These duce this restriction through alteration ofthe traditional interpretive techniques are image, accentuating slight differences which time-consuming, wearisome, and subject to could be of interest to the observer. There omission enors (Shepard, 1964). are primarily two enhancement processes; Techniques have been developed which electronic, which converts image densities consider the problems of traditional in­ to voltage, and photographic, which alters an terpretation and attempt to reduce amounts image by re-exposing it on photographic of interpretation needed to determine film. change. Sequential images can be subtracted Electronic processes which can change or averaged with one another by overlaying image densities to voltage display densities, positive and negative transparencies of an on a cathode-ray tube, allow for various en­ area in register. This results in an image in hancements of an image. Density-slicing or which areas that have not changed appear the separation of density ranges is the basis medium gray while areas of change appear upon which this process works. Density as light or dark tones (Jones et al., 1975; ranges are imaged as separate colors, dis­ Reeves, 1975; Simonett, 1974). A "blink" played all at once or separately on a televi­ process permits viewing of sequential im­ sion screen (Sayn-Wittgenstein, 1973; ages alternately at about ten frames per sec­ Schlosser, 1974). ond. Areas on the image which "blink" as Several photographic processes which the pictures are alternated represent areas have been employed to assist the interpreter which have changed (Masry et al., 1975; in identifying subtle image differences in­ Simonett, 1974). These techniques are lim­ clude: tone contrast stretching, posteriza­ ited, in that changes of interest to policy tion, edge isolation, Sabattier effect, Ag­ makers may be embedded in a "larger facontour density-slicing, and litho density­ amount of extraneous change, or clutter" slicing (Clarke, 1967; Kodak, 1973; Nielsen, (Simonett, 1974: 64). 1972, 1974; Rodriguez-Berjarano, 1975; Electronic scanning equipment also has Ross, 1976; Simonett, 1974). been employed to reduce traditional in­ Image enhancement provides a technique terpretation needed to detect change. A which can be used to isolate images of peripheral processor scanner was utilized as selected resources. Isolation ofthese images a "quick-look" change detection technique. can be a means ofelimination ofthe problem The scanner records edges of objects on se­ of "clutter" effecting interpretation of quential photographs that are subsequently changes discussed by Simonett (1974). Thus, compared by using a computer (Humiston a combination of an enhancement process to and Tisdale, 1973). Density changes of isolate images and an overlay process of natural vegetation on sequential imagery positive and negative transparencies of se­ have been detected by employing an elec- quential imagery to detect change should IMAGE ENHANCEMENT FOR VEGETATIVE PATTERN CHANGE 609 provide resource managers with a method this process was tested using model situa­ which gives information rapidly and can be tions and sequential aerial photography. utilized easily. A density step chart was employed to Many changes ofinterest to land managers model objects of various tones.* Exposures have a direct effect on vegetation and result of the density chaIt were made onto ortho in alteration of vegetative patterns. Plant type film with exposures beginning at 2.5 communities integrate separate growth fac­ seconds and at 2.5 second intervals up to a tors which affect the community, and pro­ final exposure of 10 seconds. These nega­ vide a biological expression of these fac­ tives were then contact printed on ortho film tors through the structure and components of to produce positives. Resultant positives and the community (Poulton, 1970). Vegetation negatives of different exposures were over­ has been used in numerous studies to laid in order to mask specific density ranges. monitor ana detect significant resource con­ Simulation of aerial photography was ditions including urba.n, wildlife, geologic, achieved through use of near vertical photo­ and ecologic analyses (Collins, 1972; Dill, graphs taken of a three-dimensional land­ 1963; Driscoll, n.d.; Ducker and Horton, scape model. An initial picture was taken of 1971; Estes et ai., 1972; Hanson and Smith, the model to represent a baseline condition. 1970; Jones et ai., 1975; Larson, 1967; A second photograph was taken after several Lewis, 1974; Morrison, 1972). changes had been made in the simulated Presently, the gap which exists between trees. Exposures ofthe initial negatives were capabilities for gathering extensive amounts made on ortho film beginning at a 5 second ofaerial imagery and the limited capabilities exposure and continuing at 5 second inter­ for handling and analyzing it restricts re­ vals terminating with a 60 second exposure. source managers in their ability to utilize Contact prints were made from these posi­ this source of data in their management de­ tives to produce negatives. In order to de­ cision making process. A monitoring system termine which density-slice resulted in the utilizing a combination of enhancement best image of trees, positive and negative techniques and change detection processes images of differing exposures were com­ could provide resource managers with rel­ bined and visually analyzed. To fmther re­ evant and easily applied information (Car­ fine this density-slice, additional exposures neggie, 1970; Harrell et ai., 1966). were made between the lower and upper The objectives of this study were limits of the selected density range. (1) To test various image enhancement Application of this enhancement process techniques for their applicability to detec­ to sequential aerial photographs was tion of change on sequential photographs; achieved by following procedures used for (2) To determine the accuracy of a enhancement of photographs taken of the selected enhancement technique for its abil­ three-dimensional simulation model. Se­ ity to detect change in sequential photo­ quential photographs utilized were taken in graphs; April of 1951 (CVC-5H-81) and in December (3) To test the developed technique using of 1962 (CVC-200-269) of an area adjacent to image enhancement and overlay transparen­ the Double Mountain Fork of the Brazos cies for the detection ofchange in vegetative River in Garza and Kent counties, Texas. patterns on aerial photography; and Negative transparencies were utilized from (4) To develop conclusions based on ac­ Western Laboratory, Aerial Photography Di­ curacy and expense of the developed vision, ASCS-USDA, Salt Lake City, Utah. technique, concerning its practical applica­ Negatives were photographed with 35 mm tion as a change detection procedure for re­ Plus X Pan film. Exposure settings for these source management. copies were Vz second and! stop varied from ! 5.6 to! 16 at V2-Stop intervals to insure ob­ METHODOLOGY taining images of appropriate exposure for A review of enhancment techniques both photographs. suggested that density-sliced images would Resultant positives were visually analyzed be best suited to change detection. A study in order to determine two with similar deri- of litho-photographic masking, Agfacontour, Sabattier effect, and electronic density­ * Density in this case refers to optical density slicing led to a choice of using photographic which is a function oftransmitted light. Density is masking as a study process, because of its determined by using the following fommla: applicability to a wide range of users and Density = !og,o [intensity of incident light! availability of materials. The application of intensity of transmitted light] (Nielson, 1974) 610 PHOTOGRAMMETRIC E GI EERI G & REMOTE 5E 51 G. 1978 sities and contrast. Density-slices were pro­ s = percentage subtractive change in duced employing techniques described for tree cover. enhancement of the simulated aerial photo­ graphs. In order to evaluate these estimates of A specific study region of 2.25 square percentage tree cover change, the sum ofnet miles was selected on the aerial photographs change and 1951 tree cover estimates was to encompass areas in which coverage of compared to the estimates of cover from the woody vegetation seemed to have changed. 1962 density-slice. These two estimates A coordinate grid was established within the theoretically should have been equal. Esti­ universe to divide it into 576. 2.5-acre sam­ mates which differed by more than ten per­ ple sites. Ten sample sites were then ran­ cent were investigated in order to determine domly selected for detailed study. the cause of the difference. Density-slices which resulted from en­ Change in percentage tree cover from hancement of sequential photographs were 1951 to 1962 on the ten sample sites was also contact printed into Oltho film to produce estimated using traditional interpretive positive and negative images. Positive and techniques as described by Avery (1977). negative density-slices of differing sequen­ Unenhanced sequential transparencies of tial photographs were then sandwiched to­ the area were individually projected onto a gether to produce an image of changes (Fig­ 24 by 24 grid at a scale of 1:3960, on which ure 1). sample sites were located and percentage Analysis of changes was done through vi­ tree cover estimated. A difference of these sual comparisons. Negative aerial density­ two estimates gave net change. slices were projected onto a 24 by 24 grid at a In order to make a valid quantitative com­ scale of 1:3960 for the visual comparison. parison, results obtained and time involved Sample sites were located and percentage for both methods were considered. Accuracy tree cover was visually estimated from the was tested through a comparison of net per­ 1951 and 1962 enhanced images separately. centage change estimated using both A positive density-slice of the 1962 photo­ techniques. Estimates which were within a graph was then overlaid with the 1951 nega­ 10 percent range ofone another were consid­ tive density-slice to depict subtractive ered acceptable values. Those which were changes (Figure 1). Secondly, the 1951 posi­ not within this 10 percent range were re­ tive density-slice was overlaid with the 1962 examined to determine why they were not. negative density-slice to show additive Time involved in each process tested was changes (Figure 1). Net change was com­ compared in order to determine which pro­ puted for each sample site using the formula: vided the most economic method of detec­ tion of changes in vegetative pattern. C = a - s Where C = percentage of net change in tree cover. RESULTS a = percentage additive change in Simulation of image enhancement illus­ tree cover, and trated which positive and negative trans-

01 der Scene Newer Scene type of element of visual type of element of visual transparency interest tone transparency interest tone present 1ight present light negati ve A C negat1 ve absent dark absent dark

present dark present dark positive .X. pos iti ve absent light absent light ~ t area of high light transmittance through area of high light transmittance through combination B=C indicates additive change combination A-O indicates subtractive in element of interest change in element of interest

FIG. I. Example of change detection procedure for enhanced sequential imagery (Jones et al., 1975). IMAGE ENHANCEMENT FOR VEGETATIVE PATTERN CHANGE 611 parencies should be combined to image a inadveitent loss of a small portion of a simu­ specific type of change. Additive change lated tree was detected. such as an increase in numbers or size of Enhancement of the sequential aerial individuals was detected by a combination photography resulted in images of density of the new scene negative and the old scene ranges in which trees appeared. A 12.5 sec­ positive (Figure 1). Subtractive change such ond negative and a 23 second positive pro­ as a decrease in numbers or size of individu­ duced the final density-slice of trees in the als was imaged by a combination of the old 1951 imagery. An enhanced image oftrees in scene negative and the new scene positive the 1962 photography resulted from combin­ (Figure 1). ing a positive of a 16 second exposure and a Positive h'ansparencies, resulting from in­ negative of a 9 second exposure. egative cremental exposures of the initial landscape density-slices which imaged vegetation of model photograph, produced the full density interest from both photographs are illus­ range of that image. An exposure of 5 sec­ trated in Figure 2. Processing of these onds resulted in an almost transparent image density-slices and contacts from them re­ while a 60 second exposure was opaque. An quired 3.83 hours. initial image of simulated trees was pro­ Estimates of percentage tree cover in duced from the density-slice using the 5 and sample sites were made from negative 10 second exposures. Within the density density-slices of the 1951 and 1962 photog­ range some objects other than vegetation raphy. Positive density-slices were then appeared; therefore, the range was further overlaid which resulted in images from reduced by combining the 10 second posi­ which percentage cover change was esti­ tive with a 6 second negative transparency. mated. Figure 3 illustrates these subtractive Reproduction of this density-slice from the and additive changes. An estimate of net second sequential negative of the landscape change resulted hom a difference ofadditive model was obtained by systematic adjust­ and subtractive changes (Table 2). These es­ ment ofexposure and scale settings. Table 1 timates from enhanced photographs re­ lists enlarger settings, resultant scale com­ quired 35 minutes. parisons, exposure times, and resultant tone Estimation of percentage tree cover from comparisons ofthese incremental exposures. the original 9- by 9-inch h'ansparencies for Settings resulting in desired images were each sample site provided data with which then used to produce 4- by 5-inch positive to determine percentage net change. Percent­ and negative transparencies. age cover estimates from both dates and net Changes which were detected by overlay­ change are listed in Table 3 for each sample ing positive and negative density-slices of site. These interpretations took 55 minutes these simulated aerial photographs were to complete. compared to those known changes which A comparison of net percentage cover had been imposed on the landscape model. change, using both change detection Changes imposed included two additions techniques, revealed that estimates were and three subtractions of simulated trees. significantly different for a majority of the These imposed changes were detected. In sample sites (Table 4). Four categories ofdif­ addition, a slight change resulting from an ferences were apparent; those from 75 to 85

TABLE 1 REPRODUCTION OF INITIAL DENSITY-SLICE

Enlarger Setting Exposure Time Density Comparison With (inches) Scale Comparison (seconds) Previous 10 Second Exposure

11 19/32 larger 10.0 darker 11 11132 smaller 5.0 lighter 11 15/32 smaller 7.5 matches 11 17/32 smaller 7.5 matches 11 9/16 matches 7.5 matches Density Comparison With Previous 6 Second Exposure 11 9/16 matches 5.0 darker 11 9/16 matches 4.5 lighter 11 9/16 matches 4.7 lighter 11 9/16 matches 4.8 matches 612 PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1978

1962 Density-slices imaging vegetation of interest. percent, from 30 to 35 percent, from 15 to 17 guishable from grass because of only a slight percent, and from zero to 10 percent. Sample density variation. Enhancement of the 1962 sites which fell into each of these categories photograph imaged deciduous as well as were re-examined in order to determine coniferous vegetation, both of which were reasons for such a wide range of results. elements of interest. Thus, a more accurate Sample sites 1, 7, and 10 had the greatest estimate of net change in woody vegetative difference between the two techniques. cover was made using the enhancement­ After an examination of the aerial photo­ overlay change detection process on these graphs, it was determined that these sample three sample sites when compared to tradi­ areas all fell within the same general vegeta­ tional interpretation techniques. tion type. Estimates of cover made in this Differences of net change for sample sites area were similar for the 1951 images using 5 and 6 were from 30 to 35 percent. An both techniques because the original image examination of the original photographs re­ was of high contrast, and trees were easily vealed that these sites were also in areas interpreted. However, 1962 estimates varied which were similar in vegetation and topog­ considerably because of low contrast in the raphy. Differences between net changes es­ original image. Evergreens were visible on timated were a result of two factors for these the projected original transparency; how­ sites. Shadows and low contrast of the origi­ ever, deciduous vegetation was indistin- nal 1962 transparency limited interpretive

Subtractive Additive FIG. 3. Images of changes produced by positive and negative overlays. IMAGE ENHANCEMENT FOR VEGETATIVE PATTERN CHANGE 613

TABLE 2 TREE COVER ESTIMATES FROM DENSITY-SLICE, ENHANCEMENT-OVERLA YS

1951 1962 Sample 1951 % Subtractive 1962 % Additive % Remaining Site % Cover Changes % Cover Changes % et Change Constant 1 60 25 70 45 +20 35 2 10 0 25 15 +15 10 3 15 10 15 10 0 5 4 40 25 10 3 -22 15 5 10 5 40 30 +25 5 6 30 25 15 10 -15 5 7 20 5 70 60 +55 15 8 20 10 10 5 -5 10 9 20 5 40 15 +10 15 10 85 20 85 25 +5 65 capability in these areas. Enhancement of centage cover hom each original transpar­ the 1962 photograph imaged vegetation of ency led to net change. However, with the interest and eliminated shadows which im­ enhancement-overlay technique the percent­ peded traditional interpretation. age cover removed, added, and/or remain­ Sites in which problems with the ing constant, could also be estimated. enhancement-overlay process were encoun­ A comparison of time involved with each tered are 2 and 4. Within these sites were technique showed that preprocessing of im­ features which were not elements ofinterest ages restricts efficient use of enhancement­ but which fell within the same density range overlay techniques to large areas. Preproces­ as woody vegetation. Differences in net sing involved 89 percent ofthe time spent in change reflect these added elements which the analysis ofthe 25 acres sampled. The dif­ were not interpreted as woody vegetation on ference in time spent for the two techniques the projected original transparencies. was 3.5 hours less for traditional interpreta­ Sample areas which were within a 10 per­ tion. However, the actual estimation of per­ cent range for net changes were considered centage tree cover took 37 percent less time reasonable for visual estimation of vegeta­ using the enhanced images. tive cover. Sample sites which fell within Time involved for preprocessing of the this limit were 3, 8, and 9. These were in study region resulted in enhancement ofthe areas of varying topography and vegetation; total 1440 acres. An estimate oftree cover by however, percentage tree cover in these interpretive techniques for 1440 acres would areas was low. Vegetation of interest was im­ have taken 53 hours, while it would have aged well in these three areas, which led to taken only 38 hours to do the same evalua­ acceptable estimates of cover using both tion using preprocessing and overlays. This techniques. would have represented a 28 percent savings Data which resulted from the two change in time involved for analysis ofthe total area. detection techniques (Tables 2 and 3) varied also in total information provided. Through SUMMARY AND CONCLUSIONS traditional interpretation, estimates of per- The best possible understanding of the

TABLE 3 TREE COVER ESTIMATES FROM TRADITIONAL INTERPRETATIVE TECHNIQUES

Sample Site 1951 Percent Cover 1962 Percent Cover Percent Net Change 1 70 5 -65 2 5 5 0 3 15 15 0 4 15 10 -5 5 10 5 -5 6 30 50 +20 7 25 5 -20 8 20 5 -15 9 10 10 0 10 90 10 -80 614 PHOTOGRAMMETRIC E GINEERING & REMOTE SENSING, 1978

TABLE 4 COMPARISON OF NET CHANGE VALUES

Net Change Net Change Sample Site Enhancement-Overlay Interpretation Difference 1 +20% -65% 85 2 +15% 0% 15 3 0% 0% 0 4 -22% -5% 17 5 +25% -5% 30 6 -15% +20% 35 7 +55% -20% 75 8 -5% -15% 10 9 +10% 0% 10 10 +5% -80% 85 natural environment is vital to adminis­ processing is developed into a continuous trators as a basis for planning and manage­ monitoring system, the cost ofpreprocessing ment decision making. Aerial photographic will be reduced because baseline data al­ interpretation has been proven to be a valu­ ready will be available for subsequent able technique in collection of basic data; analyses. however, monitoring of changes through Photographic variables which in theory traditional interpretive techniques is a tedi­ can affect the results of image enhance­ ous and time-consuming process. Photo­ ment-overlay processing are focal length, graphic enhance-overlay processing of im­ film, flying altitude, time of day, time of agery for change detection, developed and year, and camera systems. Problems which tested in this research program, appears to are caused by these variables are perspec­ offer natural resource managers an effective tive differences, scale changes, seasonal alternative for detection of changes in re­ vegetative changes, shadows, and optical sources of concern. densities which differ from one camera sys­ Accuracy of enhancement-overlay change tem to another and from center to the edge detection was evaluated through a compara­ of a particular image. The most critical of tive analysis of change data obtained from these is photographic scale correlation, the enhancement-overlay technique and which can be manipulated through image from traditional photographic interpretive processing. procedures. Results of this comparison indi­ cate 30 percent of the sampled units were REFERENCES equally evaluated using both techniques. Avery, Eugene T. 1977. Interpretation of Aerial Evaluations of50 percent ofthe sample units Photographs. 3rd ed. Minneapolis, Minn.: were improved using the enhancement­ Burgess Publishing Co., p. 392. overlay technique because of typical prob­ - __. 1965. "Measuring Land Use Changes on lems encountered with traditional inter­ U.S.D.A. Photographs." Photogrammetric pretative processes. Traditional photo­ Engineering and Remote Sensing 31 (April): graphic interpretation apparently evaluated 620-24. 20 percent of the sample units more accu­ Bubenik, Ernie R. 1976. "A Comprehensive Study rately than the enhancement-overlay of Land Uses and Land Capability in Lub­ technique because of extraneous objects bock County, Texas." Master's Thesis, Texas within the density range of interest. This Tech University, p. 82. analysis suggests an overall improvement Carneggie, David M., and Stephen D. DeGloria. using the enhancement-overlay process for 1973. "Monitoring California's Forage Re­ change detection. In addition to improved source Using ERTS-1 and Supporting Aircraft results, enhancement-overlay processing Data." In Symposium on Significant Results provides more thorough information on the Obtained from the Earth Resources Technol­ ogy Satellite-I vol. 1, Sec. A, pp. 91-5. Edited specific nature of changes which have oc­ by Stanley C. Freden, Enrico P. Mercanti, curred. and Margaret A. Becker. Washington, D.C.: Preprocessing of imagery for employment U.S. Government Printing Office. ofthis technique represents a limiting factor. Carneggie, David M. 1970. "Remote Sensing: Re­ Preprocessing time should be considered as view ofPrinciples and Research in Range and an added cost of providing more thorough Wildlife Management." In Range and and accurate data. As enhancement-overlay Wildlife Habitat Evaluation-a Research IMAGE E HANCEMENT FOR VEGETATIVE PATTERN CHANGE 615

Symposium, pp. 165-78. U.S. Department of David Murphy, Robert K. Holz. 1975. "De­ Agriculture, Forest Service. Washington, velopment and Application of Operational D.C.: U.S. Government Printing Office. Techniques for the Inventory and Monitoring Clarke, Alva B. 1967. "A Photographic Edge­ of Resources and Uses for the Texas Coastal Isolation Technique." Photogrammetric En­ Zone." Austin, Texas, p. 52. gineering and Remote Sensing 28 (July): Kodak. 1973. Creative Darkroom Techniques. 393-99. n.p.: Eastman Kodak, Co., p. 292. Collins, W. Gordon. 1972. "The Acquisition and Larson, J. S. 1967. Forest, Wildlife and Habitat Monitoring of Derelict Land Information in Management-A Critical Examination of Great Britain Using Photogrammetric Practice and Need. U.S. Department of Ag­ Methods." In Proceedings ofAmerican Soci­ riculture, Forest Service. Washington, D.C.: ety of Photogrammetry Fall Convention, pp. U.S. Government Printing Office, p. 28. 398-405. Falls Church, Virginia: American Lewis, Anthony J. 1974. "Geomorphic-Geologic Society of Photogrammetry. Mapping from Remote Sensors." In Remote Dill, Henry W., Jr. 1963. "Airphoto Analysis in Sensing, pp_ 105-26. Edited by John E. Estes Outdoor Recreation Site Inventory Plan­ and Leslie W. Senger. Santa Barbara, Califor­ ning." Photogrammetric Engineering and nia: Hamilton Publishing Co. Remote Sensing 29 (January): 67-70. Masry, S. E., B. G. Crawley, and W. H. Hilborn. Driscoll, Richard S. n.d. "Use of Remote Sensing 1975. "Difference Detection." Photogram­ in Range and Forest Management." Fort Col­ metric Engineering and Remote Sensing 41 lins, Colorado. (Mimeographed.) (September): 1145-48. Ducker, K. J., and Frank E. Horton. 1971. "Urban Milazzo, Valerie A., and Harry F. Lins, Jr. 1972. Change Detection Systems: Status and Pros­ "The Use of Small-Scale Photography for De­ pects." In Proceedings of the Seventh Inter­ tecting Land Use Change." In Proceedings of national Symposium on Remote Sensing of American Society of Photogrammetry Fall Environment vol. 2, pp. 1523-36. Center for Convention, pp. 270-92. Falls Church, Vir­ Remote Sensing Information and Analysis. ginia: American Society of Photogrammetry. Ann Arbor, Michigan: Willow Run Labora­ Morrison, Roger B. 1972. "Evaluation of ERTS-l tOlies. Imagery for an Inventory of Post 1890 Accel­ El-Ashry, M. R., and H. R. Wanless. 1967. erated Erosion." In Proceedings ofthe Third "Shoreline Features and Their Changes." Annual Conference on Remote Sensing in Photogrammetric Engineering and Remote Arid Lands, pp. 46-51. Tucson, Arizona: Uni­ Sensing 33 (February): 184-89. versity of Arizona. Estes, John E., Leslie W. Senger, and Randolph R. Nielson, U. 1972. "Agfacontour Film for the In­ Thaman. 1972. "Monitoring the Effects of a terpreter." Photogrammetric Engineering Changing Resource Base in Arid Lands Area: and Remote Sensing 38 (November): 1099­ The West Side of the San Joaquin Valley, 1105. California." In Proceedings of the Third An­ ___. 1974. Photographic Image-Enhancement. nual Conference on Remote Sensing in Arid Canada, Department of the Environment, Lands, pp. 308-16. Tucson, Arizona: Univer­ Forestry Service. Ottawa: Information sity of Arizona. Canada, p. 28. Hanson, W.O., and Justin G. Smith. 1970. "Sig­ Olson, Kurt ., and Merle P. Meyer. 1976. "As­ nificance of Forage Quality as a Tool in sessment of Upper Mississippi River Flood­ Wildlife Management." In Range and plain Changes with Sequential Aerial Photog­ Wildlife Habitat Evaluation-A Research raphy." In Proceedings of the 42nd Annual Symposium, pp. 25-31. U.S. Department of Meeting of the American Society of Photo­ Agriculture, Forest Service. Washington, grammetry, pp. 167-71. Falls Church, Vir­ D.C.: U.S. Government Printing Office. ginia: American Society of Photogrammetry. Harrell, David E., Nicholas C. Crawford, and Poulton, Charles E. 1970. "Practical Applications Robert W. Peplies. 1966. "Remote Sensor Re­ of Remote Sensing in Range Resources De­ turns as Historical Documents." In Proceed­ velopment and Management." In Range and ings of the Fourth Symposium on Remote Wildlife Habitat Evaluation-A Research Sensing ofEnvironment, pp. 599-604. Center Symposium, pp. 179-89. U.S. Department of for Remote Sensing Information and Analysis. Agriculture, Forest Service. Washington, Ann Arbor, Michigan: Willow Run Labora­ D.C.: U.S. Government Printing Office. tories. Reeves, Robert G., ed. 1975. Manual of Remote Humiston, Homer A., and Glenn E. Tisdale. 1973. Sensing, vol. 1. Falls Church, Virgin ia: "A Peripheral Change Detection Process." In American Society of Photogrammetry. Symposium Proceedings on Management and Utilization ofRemote Sensing Data, pp. 413­ Richter, Dennis M. 1969. "Sequential Urban 25. Edited by Abraham Anson. Falls Church, Change." Photogrammetric Engineering and Virginia: American Society of Photogram­ Remote Sensing 35 (August): 764-70. metry. Rodriguez-Bejarano, Dario. 1975. "Density Slic­ Jones, Ron, Peggy Harwood, Robert Finley, ing Applied to Forest Type Delineation." 616 PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1978

Photogrammetric Engineering and Remote Kansas City." In Proceedings ofthe 42nd An­ Sensing 41 (August): 1029-37. nual Meeting of the American Society of Ross, Donald S. 1976. "Precision Equidensity Photogrammetry, pp. 373-78. Falls Church, Slicing With Conventional Photographic Virginia: American Society of Photogram­ Materials." In Proceedings of the 42nd An­ metry. nual Meeting of the American Society of Shepard, James R. 1964. "A Concept of Change Photogrammetry, pp. 239-49. Falls Church, Determination." Photogrammetric Engineer­ Virginia: American Society of Photogram­ ing and Remote Sensing 30 (July): 648-51. metry. Simonett, David S. 1974. "Quantitative Data Ex­ Rouse, J. w., R. H. Haus, J. A. Schell, and D. W. traction and Analysis of Remote Sensor Im­ Deering. 1973. "Progress Report RSC-1978­ ages." In Remote Sensing, pp. 51-81. Edited 1." College Station, Texas. (Mimeographed.) by John E. Estes and Leslie W. Senger. Santa Sayn-Wittgenstein, L. 1973. "A First Look at Barbara, California: Hamilton Publishing Co. Canadian ERTS Experiments in Forestry." In Wagner, Robert R. 1963. "Using Aerial Photos to Earth Resources Technology Satellite-I Sym­ Measure Changes in Land Use Around posium Proceedings, pp. 48-57. Compiled by Highway Interchanges." Photogrammetric Williams A. Finch, Jr. Greenbelt, Maryland: Engineering and Remote Sensing 29 (July): Goddard Space Flight Center. 645-49. Schlosser, Stanley M. 1974. "Television Scanning Wray, James R. 1971. "Census Cities Project and Densitometer." Photogrammetric Engineer­ Atlas of Urban and Regional Change." In Pro­ ing and Remote Sensing 40 (February): 199­ ceedings of the International Workshop of 202. Earth Resources Survey System, vol. 2, pp. Shaklee, Ronald V. 1976. "Conversion of Prime 243-59. Washington, D.C.: U.S. Government Agricultural Land to Urban Land Uses in Printing Office.

Forum

A Correction on Light Diffusion Underwater.

During the first half of this century the source at a stipulated distance underwater understanding of underwater light attenua­ will effectively cover only 75 percent of the tion was at such an infant stage that diffusion surface area experienced for that same dis­ was not even considered in the attenuation tance through air. This fluctuates somewhat equation.* It was in the 1950s that material since the degree ofrefraction, and the differ­ published about this area began to account ence in the photon's speed, changes very for diffusion as an aspect of oceanography. slightly with differences in the temperature Even then, and still, the correct diffusion and pressure of the water and also with the rate in water is not employed. It is presumed individual wavelength being transmitted. that the diffusion rate in water would be the Simple tests at close distances confirm this same as encountered through air, i.e., the in­ and indicate that its effects should be ex­ verse square of the distance, but in actuality perienced when observing and photograph­ the diffusion rate is approximately 0.75 the ing underwater. inverse square of the distance. When the light photon enters the water medium it ex­ periences a change in speed, a proportional * Le-ad change in direction, and also an equal ad­ E=-­2 justment to its diffusion rate, described by d Albert Einstein's theory ofrelativity as "time -Kenneth S. Hoffman dilation". Therefore, using an illumination Miami Beach, FL 33139