JOHN F. KENEFICK

Photogrammetric Consultant, Inc.Indialantic, FL 32903

Applications of Photogrammetryin Shipbuilding

Dimensioning of a midship section, development of pipingdrawings from a design model, survey of a section of a largecylinder, and a check on the shape of a ship's cross sectionare described.

INTRODUCTION

E ARLY IN 1975 our firm ("JFK") wasselected to conduct a 14-month program

to assess the potential usefulness of photo­grammetry within the U.S. Shipbuilding In­dustry. As one of many investigations grow­ing out of the National Shipbuilding Re­search Program, the project was cost sharedby the U.S. Maritime Administration andTodd Shipyards Corporation. Todd assumedresponsibility for execution of the project

DEMONSTRATED ApPLICATIONS

MEASUREME T OF A MIDSHIP SECTIO

In order to increase the payload of existingships it has become commonplace to cut aship in half (fore and aft sections), insert amidship section, and then weld the threeunits together. The midship section is almostalways constructed in advance of cutting the"mother" ship. Although the shipbuilderexercises quality conh'ols to reasonably as-

ABSTRACT: In conjunction with the National Shipbuilding ResearchProgram, the potential for applying photogrammetry within the U.S.shipbuilding industry has been assessed. Several actual applicationsof photogrammetry were undertaken for demonstration purposes.These included dimensioning of a midship section, development ofpiping drawings from a design model, survey of a section of a largecylinder, and a check on the shape of a ship's cross section boundedby highly curved shell plating. Photogrammetric procedures and keyresults are summarized for each of these demonstrations.

and selected the photogrammetric firm towork under the technical direction of theirR&D Program Manager, Mr. L. D. Chirillo.The basic guiding principles in performingthe entire project were to assess the useful­ness of photogrammetry from a practicalpoint of view but, above all, with emphasisupon the potential for increased productivitywithin our shipbuilding industry.

Of primary interest to photogrammetristsare several demonstration photogrammetricsurveys which were conducted for this pro­ject.* Each of these is summarized in thisarticle.

sure that the midship section will be dimen­sionally equivalent to the mother ship, themidship section is a large three-dimensionalbody which is difficult to survey as such byusing conventional shipyard procedures. Ofparticular concern is whether the two ex-

• For complete details, see Photogrammetry inShipbuilding, report of the National ShipbuildingResearch Program, sponsored by the U. S.Maritime Administration in cooperation withTodd Shipyards Corporation. Available throughthe National Technical Information Service, 5285Port Royal Road, Springfield, VA 22161. Docu­ment No. PB-262-130NS

PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING,Vol. 43, TO. 9, September 1977, pp. 1169-1175.

1169

1170 PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1977

posed cross sections of the midship areparallel to each other and perpendicular tothe centerline. If they are not, the joinedship will not be straight and may developserious steering and power loss problems.

Whereas it was desirable to demonstrate aphotogrammetric survey of a midship sec­tion, this work also served to illustrate howphotogrammetry can be used to dimensionmating faces of very large ships built inhalves.t In either case, as in most large steeldimensioning tasks within shipbuilding, theaccuracy requirements are such that allpoints must be located within 1/16 inch (onestandard deviation) in each of the three car­dinal directions.

About 100 targets were attached to themidship section at points of particular in­terest to the shipbuilder. Typically, these lo­cations were intersections of main longitud­inal girders with the decks and decks withthe shell plating. The targets were attachedby small but very strong magnets. Severaladditional targets were attached (pressuresensitive adhesive) to the main deck; dis­tances measured by steel tape betweenthese served to scale the photogrammetricmodel.

"".~...

STMION") STATlON .. 1VERTICAL SECTION A-A

FIG. 1. Camera locations for survey of midshipsection. A single camera was moved betweencamera stations. Overhead pictures were takenfrom a personnel carrier slung from the boom of acrane.

t An analytic fit-up prediction of a tanker builtin halves was successfully completed in early1977.

Figure 1 illustrates the photographicscheme. All exposures were secured withJFK's Wild P31 which was moved from onecamera station to the next. Overhead photo­graphs were taken from a personnel carrierslung from the boom of a "cat" crane. Figure2 illustrates a typical photograph taken fromthe ground.

All eight exposures were measured on aKern MK2 monocomparator* and the result­ing measurements were processed throughthe JFK-developed CRABSf software pack­age. Upon convergence of the 8-station so-Iution, the nns of the photographic measure­ment residuals was 2.2 micrometers and thetypical accuracy of a triangulated target was0.03 inch (standard deviation) in each of thethree cardinal directions.

Once the photogrammetric solution wascompleted, a variety of dimensional analyseswere performed for the shipbuilder. Theseinvolved calculations of distances, best­fitting planes, best-fitting planes perpen­dicular to other determined planes, transla­tion of planes parallel to themselves, best­fitting circles to turns of the bilges, and com­paring as-built dimensions to design data.All of these analyses were designed to pro­vide a complete three dimensional "picture"of the midship as-built.

PIPING DRAWINGS FROM A SCALE MODEL

In the design of distributive systems,especially within engine rooms, a scalemodel is often constructed to assure that in­terferences do not occur. One school ofthought holds that design should take placeat the model and, once completed, the modelis the design. This concept is in contrast tothe more conventional approach wherein thedesign takes place on paper first and if amodel is built at all it is usually done only asa final test for interferences.

With the design modeling concept it isnecessary to transfer the design inherent inthe model to paper for dissemination to allpersonnel requiring details of the design. Inorder to illustrate how photogrammetry canaccomplish this quite readily (as opposed tomanual measurement of the model) a sampledrawing was prepared on the Wild AIO**

* Mensuration was performed by Analytic PhotoControl, Inc., of Indian Harbour Beach, Florida.

t CRABS is the acronoym for Close RangeAnalytical Bundle Solution which is a totallyrigorous weighted least-squares simultaneoussolution with error propagation.

** Stereo plotter work was performed by Hen­derson Aerial Surveys, Inc., of Columbus, Ohio.

APPLICATIONS OF PHOTOGRAMMETRY IN SHIPBUILDING

FIG. 2. Typical photograph of the midship section taken from the ground.

1171

from a series of stereo pairs taken with theWild P31 camera of a portion of a 3/4" = l'model of an engine room. Figure 3 is typicalof the photographs used to compile the pip­ing drawing shown in Figure 4. It should benoted that the final product is a dimensioned

drawing; i.e., locations of the pipe center­lines are indicated by distances from over­head decks and bulkheads.

In performing this demonstration it wasrecognized that the direct graphical productof analog photogrammetry was merely

-L'2 ---- 2/-0 1/ -( .. <,,~"" .. A I I---- '-'--

FIG. 3. Portion of a photograph of the machinery space model showing the extent ofthe model mapped.

1172 PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1977

ou

~ ! ~~ , .

FIG. 4. Elevation drawing of the steam systems in a portion of the machinery spacemodel. The drawing is reproduced from the original compilation manuscript and notredrafted for aesthetic value. Compilation scale was %" = 1'. ote that locations ofcenterlines of pipes are dimensioned.

containment of the cargo. Attached to theequator of each tank is a full section of a cyl­inder which hangs from the equator in thefashion of an apron. A similar section of cyl-

PLAN VIEW

o

o

o

I

+,

o

o

~.----------:-.o.... 0,,- CAMERA STATIONS(TYPICAL)

>-----<10'

EDGE OF WEATHERDECK OPENING

touching upon the full potential of photo­grammetry as an aid to design modeling.Should a model be prepared with an eye to­ward photogrammetric dimensioning (themodel photographed was not), a digital rep­resentation would be most desirable so thatpipe subassemblies, arrangements of indi­vidual piping systems, composites of severalsystems, and even isometrics could be plot­ted by computer drafting at any scale and atany time.

This demonstration was not maximized foraccuracy. A comparison of 20 distances de­tem1ined by photogrammetry to the corre­sponding design distances' revealed an aver­age difference of l-Y8 inch at the scale of theship. This includes photogrammetric errorand error in the model itself. In true designmodeling, however, the model is the designand only the photogrammetric error is ofconsequence. Nonetheless, it is felt that thephotogrammetric elTor in this particular dem­onstration is about twice that which wouldbe acceptable to shipbuilders. But, with themodel built to suit, there is little doubt thatphotogrammetry can supply the necessarydata with sufficient accuracy.

MEASUREMENT OF A SECTION OF CYLINDER

One series of liquid natural gas (LNG)tankers under construction within the U.S.utilizes 120-foot diameter spherical tanks for

• This particular model was first designed onpaper.

FIG. 5. Target and camera locations for the sur­vey of a section of a cylinder within the hold for aspherical LNG tank. Dots indicate locations oftargets. At each camera station the camera ispointed across the hold with the axis depressed orelevated depell :ling upon whether the camera ison the weathe deck or within the hold. A singlecamera was me ved between camera stations.

1173APPLICATIONS OF PHOTOGRAMMETRY IN SHIPBUILDING

FIG. 6. A typical photograph looking into the hold from the weather deck. The section ofcylinder is seen as an arc across the center of the photograph.

SURVEY OF A SHIP'S CROSS SECTION

This particular demonstration involved anL G tanker which was to receive its bow

FIG. 7. Deviation of the top edge of the cylinderfrom the best fit circle. Vectors point from the bestfit circle to the actual steel. Data also were pro­vided in tabular form.

plane) of the top edge of the cylinder and the"out of roundness" of the top edge. The lat­ter was determined by a best-fit circle calcu­lation, the results of which are presented inFigure 7.

SleD

SCALE FOR DEVIATIONS-D ,.AFT

FWD

PORT

inder is constructed within each hold of theships which transport the spheres. When asphere is lowered into a hold, the two sec­tions of cylinder mate so as to support thetank in the same manner as a slightly over­sized ball set on top of an open coffee can.

Because the shipbuilder desired detailedinformation regarding the as-built configura­tion of the first-constructed cylinder withinthe first ship of the series, a photogrammet­ric solution was devised. Targets wereplaced about the top edge of the cylinderand at other locations of interest to the ship­builder. Twelve photographs were takenwith the P31 from the locations illustrated inFigure 5. Figure 6 is a typical photograph.Each plate was measured on the MK2 andthe resulting measurements were processedthrough CRABS. An rms of 1.7 micrometerswas achieved for the photographic mea­surement residuals and the typical resultantaccuracy of a triangulated target location was0.04 inch (one standard deviation) in each ofthe three cardinal directions. Taped dis­tances between selected pairs of targetswere used to scale the solution.

A number of different analyses were per­formed for the shipbuilder. The most in­teresting of these was a determination of theplanarity (Le., deviation from a best-fit

1174 PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING, 1977

/

BUILDING DOCKBASIN GATE

3

PLIN

CAMERA STATIONS(TYPICAL)

~".

PORT

­o ,.SCALE FOR DEPARTURES

ElEVATION

FIG. 8. Target and camera locations for the sur­vey of a ship's cross section. Dots indicate loca­tions of targets. Targets interior to the shell andnot on the centerline served mainly as passpoints.A single camera was moved between camera sta­tions. Numbers below the camera stations indicatethe number of exposures taken; each exposureviewed somewhat different portions of the ship'scross section.

FIG. 9. Typical photograph for the survey of theship's cross section.

FIG. 10. Deviation of the shell plating of theship's cross section from the design. Vectors pointfrom the design surface to the actual steel. Datawere also provided in tabular form.

section in two large pieces. The master buttwhere the join was to be made was boundedby shell plates having high degrees of curva­ture. To obtain a representative sample ofthe as-built configuration of the plating,targets were attached to the shell (in theplane of the cross section) as illustrated inFigure 8.

Due to physical restrictions of the job site,it was only possible to secure photographsfrom the walkway along the top of the gate tothe ship's building basin (see Figure 8). Ac­cordingly, nine slightly convergent andhighly overlapping photographs were takenwith the P31 from the locations indicated inFigure 8. Figure 9 is a typical photograph.The photographs were measured on theMK2 and the measurements were then pro­cessed through CRABS with a resultant rmsof the photographic measurement residualsof 1.8 micrometers. The typical accuracy of atriangulated target was 0.03 inch (standarddeviation) in the vertical plane and 0.08 inchto and from the observer. Finally, the as­built shape of the shell plating, as deter­mined by photogrammetry, was compared tothe original design in order to obtain the dif­ferences shown in Figure 10.

CONCLUDING REMARKS

Both fully analytical and analog photo­grammetry can be used to the benefit of ourshipbuilders. Applications for both methodswere found scattered throughout the variousphases of shipbuilding. Principal applica­tions of fully analytical photogrammetry,however, lie in dimensioning large steelunits with an eye toward predicting poten­tial (and very costly) fit-up problems. Re­quired photogrammetric tools andtechniques are state-of-the-art, but it isnecessary to abandon the concepts of stereo

PHOTODE:>JSITY CONTROL SYSTEM FOR ORTHOPHOTO PRODUCTS 1175

pairs and surveyed control in order toachieve sufficient accuracies and to produceresults economically within an acceptableperiod of time.

Analog photogrammetry shows great po­tential for use in combination with designmodeling of distributive systems. It may alsobe applied to the task of developing as-builtpiping drawings. For this latter application itis necessary to develop a simple, fast, andyet accurate solution which can be utilizedwithin a shipbuilder's or naval architect'sfacility.

It is very clear that photogrammetry ismerely a tool for developing basic data.Once the photogrammetric solution isachieved, it is usually necessary to performadditional analyses as was done for the ap­plications just described. Thus, the photo­grammetrist must be adept at performingsuch analyses for the shipbuilder in an un­compromising manner. Generally speaking,a fairly extensive inventory of "data manipu­lation" programs is needed in order to com­plement photogrammetric instrumentationand software.

APPLICATION FORM

PLEASE USE BLOCK LETTERS

THE PHOTOGRAMMETRIC SOCIETY, LONDONMembership of the Society entitles you to The Photogrammetric Record which is published

twice yearly and is an internationally respected journal of great value to the practicing photo­grammetrist. The Photogrammetric Society now offers a simplified form of membership tothose who are already members of the American Society.

To. The Hon. Secretary,The Photogrammetric Society,Dept. of Photogrammetry & SurveyingUniversity College LondonGower StreetLondon WCIE 6BT, England

I apply for membership of the Photogrammetric Society as,o Member - Annual Subscription - $12.50 (Due on' applicationo Junior (under 25) Member - Annual Subscription - $6.25 and thereafter ono Corporate Member - Annual Subscription - $75.00 July 1 of each year.)

(The first subscription of members elected after the 1st of January in any year is reducedby half.)I confirm my wish to further the objects and interests of the Society and to abide by theConstitution and By-Laws. I enclose my subscription.Surname, First NamesAge next birthday (if under 25)Professional or OccupationEducational StatusPresent EmploymentAddress

ASP MembershipCard No .

Signature ofDate Applicant .Applications for Corporate Membership, which is open to Universities, Manufacturers andOperating Companies, should be made by separate letter giv:ng brief information of theOrganisation's interest in photogrammetry.