O. M. I. Corp. v. Kelsh Instrument Co.

R. DORSEY WATKINS, District Judge.

This is an action for a declaratory judgment brought by O. M. I. Corporation of America and Ottico Meccanica Italiana E Rivelamenti Aerofotogrammetrici S. p. A., plaintiffs ¹ against Kelsh Instrument Company, Inc. and Harry T. Kelsh.² The suit seeks a declaration that Patent No. 2,492,870, issued December 27, 1949 on application filed January 20, 1948, for a “Stereoscopic Projection Map-Making Instrument” is invalid and not infringed by O. M. I. In its answer, Kelsh denied the allegations of O. M. I. essential for its recovery, and counterclaimed for a declaration that the Kelsh patent was valid, and was infringed by O. M. I.³

Pursuant to stipulation of counsel, only Claim 1 of the Kelsh patent, reading as follows, is involved:

“1. An instrument for making maps by stereophotogrammetric methods, comprising a pair of projection lanterns each having a lens and slide-receiving means and supported side by side for projecting super-imposed images of a pair of consecutive slides, a movable plotting table having a screen on which the images are projected for viewing to give a stereoscopic model, a point light source for each lantern and a light condenser to converge the light rays through a small area of the slide and to a point coincident with the nodal point of the lens and diverge the light rays to cover approximately the screen of the plotting table, each of said light source and light condenser being movably mounted relative to the lens in such manner that substantially the entire area of the slide in each lantern may be traversed by the converging light rays, and means for moving each light source and its light condenser so attached to the plotting table that as the plotting table is moved about, the light source and condenser move in a manner to maintain the image on the screen and to maintain the point of convergence of the light rays coincident with the nodal point of the lens.”

The art of photogrammetry, defined as “the science of measurement from photographs,” ⁴ is old⁵ and various devices were on the market in 1946-1947 when Harry T. Kelsh entered the field and “culled the art. He found the best thing from here and the best thing there and put them all together.” ⁶ *Subsequent to Kelsh’s entry into the business of supplying photogrammetric projection apparatus, Ottico with knowledge of such devices “designed and constructed a photogrammetric projection apparatus that was in a general way similar to the Kelsh apparatus * * * There is no doubt that the Italian plaintiff in this litigation knew of the Kelsh instrument when it designed its instrument and that there is a certain general similarity between the instruments * * * ”1⁷ Plaintiffs claim however, that these “similarities to the Kelsh instrument are only incidental to the fact that it is designed to accomplish the same general purpose and that it, like the Kelsh instrument, has incorporated from the prior art the desirable features.” ⁸

The court is therefore required to choose between the principles that (a) “Courts should scrutinize combination patent claims with a care proportioned to the difficulty and improbability of finding invention in an assembly of old elements”, Great Atlantic & Pacific Tea Co. v. Supermarket Equipment Corp., 1950, 340 U.S. 147,152, 71 S.Ct. 127, 130, 95 L.Ed. 162, and (as claimed by plaintiffs) the questionable process of the patent application through the Patent Office, and (b) the presumption of validity through the issuance of the patent, U.S.C. Title 35, § 282; the difficulty of the problem and the success of its solution; the significance of imitation, Diamond Rubber Company of New York v. Consolidated Rubber Tire Co., 1911, 220 U.S. 428, 441, 31 S.Ct. 444, 55 L.Ed. 527; Ackermans v. General Motors Corp., 4 Cir., 1953, 202 F.2d 642; Florence-Mayo Nuway Co. v. Hardy, 4 Cir., 1948, 168 F.2d 778, 782, and of commercial success, Hutzler Bros. Co. v. Sales Affiliates, 4 Cir., 1947, 164 F.2d 260, 267.

The Problem

In a photograph, a three-dimensional object is brought by rays to a point. This perspective projection (the intersection of a cone of rays with an intervening plane) is essentially a graphical record of a set of angles.⁹ By examination of a single photograph, whether by observing the print or by reprojecting the picture, a planimetric or two-dimensional view can be obtained, by which directions, but not distances or depths, can be determined. For the determination of distances, observations (or photographs) of the same object from two directions would be necessary.

In 1898 Th. Scheimpflug experimented with the projection of two photographic pictures of an object taken from different points of view, in an endeavor to produce maps and plans from photographs by a direct and optical method.¹⁰ Two cameras fastened to a table, in different positions, were used to photograph a model of a house, and later a model of ground relief. The negatives were developed and replaced in the same cameras. Each was illuminated from behind by a light source, the cameras in effect becoming projection lanterns. The rays of light emerging from the cameras took exactly the same path traversed in entering to make the photographs. The corresponding rays would therefore intersect at the same points which they had left to take the photographs. In place of the object photographed, a plane intercepting screen was substituted, so thát two projected pictures could be seen which overlay and cut through one another; that is, the corner of the model that had been photographed was reprojected back along the same projection, the corner of the house being reintersected in space, so that there was created on the screen an “optical model” of the house, which before had been a physical model. Those points above were in coincidence which represented the section of the object by the plane of the screen. The screen could be moved, and from the sections corresponding to the several positions of the screen, plans showing contour lines could be prepared.¹¹

A further refinement was necessary to overcome difficulties in such optical intersection. So far as the photogrammetric imaging in stereophotogrammetric survey of a landscape is concerned, all points lie practically at an infinite distance and are sharply defined on the plate. In the reconstitution of the landscape, the base is diminished in scale, and the intersections of the rays from corresponding plate-points are situated not at an infinite but at a finite distance from the plates. Care must be used to secure sharp definition of these points, which are intersected at diverse unequal distances. Scheimpflug quickly rejected the use of the photographic lens for reconstruction, and substituted telephoto lenses of adjustable focal lengths.

In addition to the creation of the optical or instrument model¹² there is the problem of visually detecting the surface of the instrument model. At least four methods of detecting the surface of the instrument model are recognized:

(1) Moving the screen to the point where two corresponding images coincide, as was taught by Scheimpflug.¹³

(2) Projection of the two slides in complementary colors, or any combination of two colors which when combined produce a different color (e. g., red and green). When the two images so projected coincide, the third color will appear.

(3) The “flicker” system. The right and left projectors are alternately turned on and off, so that only one of the projectors is on at a time. If the screen coincides with the surface of each projection, the images would coincide in their relative position on the screen¹⁴, and the image would appear to stand still. If, however, the intersection of the rays fell in front of or behind the screen, the image would appear to move back and forth; to jump or flicker.

(In the first two of these methods, the effort is to effect exact superimposition. In the third, the effort is to have the alternating images occupy in turn an identical position on, and portion of, the screen).

(4) Stereoscopy — by far the most important method of visual inspection. Although judgment of depth and distance due to factors such as shadows, perspective and overlap can be formed with the use of only one eye, or monocular vision, with two eyes essentially a perspective projection can be seen. Each eye acts somewhat as a lens, bringing the cone of rays to the retina, so that there is the rough equivalent of two cameras taking pictures of the same object at different camera positions, the space between the eyes being the separation between the cameras, so that each eye sees a slightly different picture. For stereo viewing, to duplicate the conditions of natural three-dimensional vision, it is necessary to present to each eye its own separate picture, or onset of images. There are a number of ways of separating the two pictures, so that each eye sees only its own picture.

(a) Two photographs of the same object, taken from slightly different positions, may simply be held in front of the eyes, and professionals, or others who are able to accommodate their eyes upon nearby images and at the same time converge them, can obtain the effect of a three-dimensional model.

(b) The stereoscope, keeping each eye from seeing the other picture, and containing lenses enabling the viewer to look out at a distant image and at the same time accommodate for a nearby subject (photograph) may be used as an optical device for assisting the eyes to achieve accommodation and convergence without strain.

(c) The third system or type of separation of the images is by image suppression, in which the left picture is suppressed as far as the right eye is concerned, and the right picture is suppressed with respect to the left eye. This “blocking out” process can be accomplished in any of at least three ways.

(i) The anaglyphic method using colors. Normally complementary colors, such as red and green, are used. In such case, for example, the right picture would be projected in red and a red filter would be put in frames or spectacles over the right eye, which would let in the red picture but would filter (block out, suppress) the green picture; and the left picture would be projected in green, and a green filter placed over the left eye.

(ii) Polarized light. In this method, for example, the projection of the right picture would be polarized in a vertical plane and a polaroid filter placed over the right eye which would accept (pass) only light polarized in a vertical plane. The left picture would be projected in light polarized in a horizontal plane, and a filter would be placed over the left eye which would accept only light polarized in a horizontal plane.¹⁵

(iii) A flicker system (different from the one heretofore mentioned) identified as a synchronized flicker system. A rotating shutter would be placed on each of two projectors synchronized so that only one picture at a time would be projected. The viewer would wear spectacles fitted with shutters synchronized with those of the projectors, so that when the right picure was on, the right shutter would be open and the left shutter closed; and vice versa. This mechanical and electrical suppression would enable each eye to achieve its own picture; and if a very quick shutter action in excess of sixteen flashes per second were employed ¹⁶, the image would appear to stand still, and there would be produced a three-dimensional model in the mind.¹⁷

In stereo viewing of the surface in three dimensions, there must be some way in which, for measurement purpose, the instrument model can be connected with what the mind is reconstructing. This is done by means of the “floating mark” or “floating dot.”

For purposes of measurement the floating mark is placed in coincidence with the point of the object to be measured. This mark can be introduced into the system in two ways. First, a real physical mark — a small black dot, a small point, source of light, or a cross — is put in the same area (the platen or viewing table) in which the actual instrument model is being formed. If the two images have been properly separated, so that the mind is reconstructing a three-dimensional image, the physical mark will appear to be in contact with the surface. If now the mark is raised substantially and the eyes are focused on the images cast on the platen, the mark will appear to split¹⁸, since the eyes will not sense such depth. If the “floating mark” is brought closer to the surface upon which the eyes are focusing, a point will be reached where vision can converge on the mark and the surface at the same time; the two (apparent) parts of the mark will coalesce into one which will appear just above the surface and may be moved down to and below the surface, where finally it will appear to split again. This is the method of Kelsh and Nistri.

Secondly, a dot may be placed on each photographic plate where the chief ray intersects it. The image of each spot is projected down to where it seems to intersect in space.

The projection of the photographs likewise posed problems. The images of the diapositive (i. e., a positive printed on a transparent medium) will all appear on the projection surface on the same plane. When two of these are combined to project a three dimensional surface, sharp imagery at different levels or depth of focus, is required. This is achieved by reducing the apertures of the projection lenses.¹⁹

An early problem was encountered from the fact that the camera lenses used in taking photographs were subject to substantial distortion from imperfections in manufacture and design. The use in projection of the same lens used in photographing, or an equivalent lens, would automatically compensate for distortion²⁰. However, the object when photographed was practically in optical infinity, and the reprojection through the camera lens would require infinity projection, unless supplemental means were used. Therefore, an auxiliary lens system, sometimes with a movable element, was introduced, by which the rays would be collected and focused on a near point.

In viewing, there is a fundamental choice between viewing with transmitted light, and viewing with diffused light. In transmitted light, the light of the photographic images is transmitted optically into the eye, the eye observing an image formed in air or in glass. This ■observation may be by looking directly at the diapositive, or the light may be “piped” by optical train, by reflection from one or more glass surfaces. The system is highly efficient with respect to light loss.²¹

When diffused light is used, images are projected onto a surface, or screen, which may be translucent, for observation from the rear, or may be opaque.²² This use would be classed as an inefficient use of light from the standpoint of a single viewer.²³ Both the Kelsh and Nistri plotters work on diffused light.

A further problem in the production of photogrammetric plotters relates to the selection of an illumination system. Sufficient light must be provided to produce reasonably bright images. There is a choice between a fixed illumination source and a movable illumination source. A fixed source must produce adequate illumination over the entire plate, negative, or diapositive; if the light source is movable, there is need to illuminate only the portion of the photograph being viewed at the time.

In either case, light should be collected and directed by a condenser (reflector or mirror) or a lens condenser toward the “image” of the lens aperture —the entrance or exit pupil.

These problems received the attention, often undivided, of members of the Zeiss, Bausch and Lomb, Nistri and other organizations, and of independent students, craftsmen and professors. As a result, various photogrammetric plotters were designed, patented, and made. At the final argument, plaintiffs’ counsel agreed that they were basing the claim of invalidity of the Kelsh patent on the Multiplex, Stereoplanigraph and Gallus-Ferber photogrammetric plotting devices, and the Barr Patent No. 1,655,306 for a “Photogrammetric Plotting Apparatus”, issued January 3, 1928 on application filed June 23, 1927, and the Horner Patent No. 2,085,498 for a “Device and Method for Aerial Survey” issued June 29, 1937 on an application filed April 20, 1933.²⁴ To determine the state of the art, and whether Kelsh is truly a combination or only an aggregation, an analysis of these five citations will be necessary.²⁵

The Prior Art and Patents.

Multiplex

A Multiplex plotter was developed by Zeiss in 1932; in 1934 Nistri brought out a very similar device; and Bausch and Lomb also has produced the same type of plotter.

In these devices, there is the dichromatic projection through fixed lenses in complementary colors, usually red and green, of partially overlapping pictures, and observation, through glasses of the same complementary colors, of the spacial model formed by the fusion (superimposition) of the doubly projected area of the overlap. It is distinguished from both Kelsh and Nistri in that the entire field of a reduced diapositive is projected by fixed illumination; and the entire base or screen is illuminated, not simply the platen. Its advantage is that the operator cannot possibly skip or miss in moving across the screen. Its disadvantages are that an expensive reduction from a 9 x 9 negative to a 2 x 2 diapositive is required; that in such reduction some definition is lost, and the projection of the entire diapositive, even in reduced size, does not have the sharpness of the projection of a relatively small portion of a diapositive, particularly at the outer edges; and that a lamp of 30-100 watts intensity, with a very large light condenser, is required for each projector.

Stereoplanigraph

In this “Cadillac” or “Lincoln” class plotter, two images of partially overlapping photographs are separately projected and the transmitted light is brought to the viewer by separate optical trains. Daylight is used where and when available, in which event the entire diapositives are viewed. Where daylight is unreliable or not available, artificial lighting is employed, in which case only the portion of the diapositives so illuminated would be viewed. Such lights would be movable, and perhaps would swing about the entrance nodes of the projection lenses. The Porro-Koppe system of projection is used, so that an auxiliary lens system is necessary on each projector.

These devices are very expensive, prices ranging between $50,000-$100,-000, as compared with approximately $6,-500 for the Kelsh and $8,500 for the Nistri plotter.

Gallus-Ferber

This device, after some thought and hesitation, was stated by plaintiffs’ expert²⁶ to be the closest device to Kelsh in the prior art. Its “modern” form (1938) is attributed to Nistri.²⁷ It is therefore somewhat startling to find that the description of this allegedly closest device is left to a text book and articles,²⁸ the most recent of which is more than twenty years old, and the testimony of plaintiffs’ expert therefrom, when Umberto Nistri, the president of O. M. I., was present in court throughout the trial.

In Gallus-Ferber, two cameras identical with those with which the photographs were taken are used as projectors. As in other Porro-Koppe projections an auxiliary system of (sliding) lenses, in addition to the projection lens, is associated with each projector. Condensers and lights are used, which illuminate only the portion of the negatives visible on the screen. These lights are movable. The literature does not discuss whether they rotate about the image of the lens aperture. The device may be operated in a darkened room.

A flicker system is used, by which shutters alternately blank the images from each camera at the rate of about two per second. Without attachments no spacial model is formed;²⁹ the viewing is monocular,³⁰ allowing only planimetrieal or two-dimensional viewing. Measurement of height is made by noting the change in elevation of the cameras, not of the platen or screen, as in' Kelsh and Nistri.

By adjustment of the auxiliary lenses and without additional attachments, overlapping portions of the negatives may be “rectified” ³¹ on the platen in a circle with a diameter of less than two inches, as compared with a diameter of three and one-half to four inches on the Kelsh platen. A small arrow on the screen or platen, formed by the intersection of two very fine lines, is brought over the point to be restituted.

Plaintiffs’ expert testified that as a “very eyestraining proposition” if the operator of a Gallus-Ferber machine put on a pair of spectacles with rotating shutters synchronized with those on the projectors, the images could be viewed stereoptically.³² To the court, however, to accomplish with Gallus-Ferber what can be done with Kelsh or Nistri, it would be necessary to:

(1) Eliminate the flicker and have both projectors on constantly and simultaneously, and projecting superimposed images.

(2) Insert red and green filters in the projectors.

(3) Increase the illumination.

(4) Equip the operator with glasses having a red and a green filter.

(5) Remove the auxiliary lens systems.

(6) Use shorter length focal lenses for projecting.

(7) Use both eyes in viewing.

If Gallus-Ferber is the closest of the prior art to Kelsh, how close is close?

The Barr Patent

Images from a pair of overlapping photographs are projected upon and kept separate upon a translucent screen. “The partial pictures * * * can be viewed from the front by means of a comparator which may be constructed either on the coincidence or the stereoscopic principle * * * ”³³ When the two separated images are examined on the screen by means of a stereoscopic comparator, floating marks are inserted in the optical system.³⁴

The projection lenses employ the Porro-Koppe principle, so that slideably adjustable auxiliary lens systems are necessary. Projection may be made of part of the diapositives at a time. Apparently the lights may swing so as always to pivot around the aperture of the lens.³⁵

The Horner Patent

The court, all counsel, and both experts are in agreement that the Horner Patent No. 2,085,498 for “Device for Aerial Survey” is an involved document, and one difficult of interpretation. The two experts were in direct conflict as to its disclosures. Plaintiffs’ expert testified that:

“He describes a system of projecting stereoscopically, or projecting optically, I should say, the images on these two auxiliary planes which he makes use of and collects these images and brings them to the eyes by an optical transmission system. He goes to great pains to describe the mechanical achievement of these planes, which leads me to interpret it as diffusion screen of a translucent nature, meaning a rear view nature in this case. The Horn-er instrument would then be using diffused light for viewing. But also he has a very complicated system of optics for convenience purposes to bring them to a fixed eye position. So, as previously described, the operator doesn’t have to physically move around. He can stay still and move this optical system around.”

The patent does use language which in the court’s opinion justifies the above quotation as one proper interpretation, except that the court does not understand how stereoptic, or even helpful, “rear view” observation could take place.

Defendant’s expert said of Horner that “he is describing several types of methods of viewing through optical trains * -x- -x- ip0 ke specific, Figure 9 represents one form and Figures 2 and 3 represent a second form. The thing in common to both is that they view this system through a binocular optical train * * * Figure 9, you view directly through an optical train up into the back, into the diapositive * * * ”

He further testified that in the other case, Figure 2, images were formed on translucent screens, which are viewed by additional optic trains.

Support for the conclusion that Horner’ contemplated, as one of his viewing methods, transmitted light and direct optical viewing, without diffusion or screens, is found in the statement in the specifications that “daylight may be substituted for the lamps.”³⁶ It is highly improbable that daylight would suffice for satisfactory viewing by diffusion.³⁷

A further complication of Horner, which neither expert was able to explain to the court’s understanding or satisfaction, is his reference to having the “main rays” of the images meet on the small plate above the drawing board.³⁸ This plate is variously referred to as a “mark plate”³⁹ or “projection plate.”⁴⁰ The reference to “projection plate”, particularly in connection with Figure 1, suggests to the court the projection of superimposed images on the plate, particularly in view of the language:

“The projection rays fall through the object lenses of the picture projecting cameras directly and without refraction on to the drawing or surveying table or on to the mark plate of a drawing device movable over the table in such a manner that the measuring out of reproduction can take place in the direct ray path.

“It is an important condition that the complementary main rays of a pair of images meet each other on the mark plate of the drawing appliance in any position of the set.” ⁴¹

The experts did not agree with any interpretation that involved image projection onto the mark plate or projection plate. Difficulty was admitted, however, as to how the “main rays” only could be projected onto the mark or projection plate, and all the rest of the pair of images be diverted elsewhere.

Horner claimed that his device afforded the possibility for stereoptic observation of the relief by either chromatic or optic separation of the images; and that with synchronized shutters on the projection cameras and the “stereoptic glass” in the optical train, it would be possible to check the stereoscoptic projections on the drawing table.

Much time was also spent at the trial in an effort to determine how much Horner disclosed with respect to the motion of the moving, rotatable light sources. That such light sources were disclosed is clear. That they should rotate about the image of the aperture is clear, although not stated. That they would not so rotate on a device constructed in accordance with Figure 1 was physically demonstrated at the final argument.

Background of the Kelsh Invention

The inventor, Harry T. Kelsh, after being graduated from the University of Pennsylvania with a Bachelor of Science degree, was commissioned as an officer in the Coast and Geodetic Survey in 1911, where he remained until he became a lieutenant in the Air Force during the first World War. After a period in private business he returned to the Coast and Geodetic Survey in 1933 as a civilian employee. In 1936 he joined the Soil Conservation Service as a mapping engineer. He remained in that service for ten years, and was then transferred to the Geological Survey where he remained until his retirement in 1956. He has been active in the American Society of Photogrammetry since 1936.

Kelsh had had some knowledge of ground photogrammetry since 1914, and then of aerial photogrammetry during the War. This, coupled with his membership in the American Society of Photogrammetry, led to acquaintance with what was being done in this field in Europe and the United States. Consideration of the instruments in practical use caused him to appraise the Stereoplanigraph as a “very wonderful instrument,” but “the first practical, more usable instruments that came to America were the Multiplex about the mid-1930’s.” Zeiss and Nistri had done certain work on direct projection, but had not produced a full scale projection plotter. Their answer had been by miniaturization, leading to the Multiplex.

The general opinion during the late 1930’s and early 1940’s was that a full scale projection plotter, using the 9x9 diapositives, was not practical. A condensing lens to cover all of a 9 x 9 diapositive would have to be at least seventeen inches; of great weight; and its size would prevent the two projectors from being placed close enough together in space to simulate the relations of the “taking” cameras; and the heat generated would be prohibitive.

Depth of focus, based upon the aperture of the lens and the focal length of the lens, was critical; but even in the reduced diapositive size, or 2x2, used in Multiplex, detail “washed out” in miniaturization and was lost, particularly in the corners of the model. Definition at the corners was essential to positioning the pictures so that they would regain the positions of the taking cameras.

Kelsh for some ten years experimented, duplicating most of the previous experiments. He tried the large condenser lens necessary to illuminate the whole of a 9 x 9 diapositive. This was not satisfactory, even when the weight was reduced by making it of plastic — on which a patent was nevertheless obtained. Then it occurred to him that instead of seeing the whole model at one time, as in the Multiplex, it was necessary to see only a small portion at one time. This would permit an increase in light. Focusing on the platen, some four to six inches above the drafting board, reduced the projection distance. By peaking the light cone on the projection lens, enough light was obtained to cut the diaphragm opening, giving a sharp image and depth of focus. The maintenance of this condition regardless of what portion of the photograph was being observed was accomplished by hooking the light system to the lens so that as the light swung, as the platen was moved, the light was automatically positioned through each projector on to the platen.

This eliminated the often complicated optical train and auxiliary lens systems; depth of focus was secured by miniaturizing the lens opening instead of the diapositive; and this permitted the operator to work in more light than the “salt mine” of the Multiplex.

That the problem on which Kelsh worked was an important and difficult one, and that its solution was commercially satisfactory, is clearly established by the evidence.

(a) Defendant’s expert, at a time when he was manager of Bausch and Lomb’s department in charge of developing photogrammetric instruments, learned of the Kelsh plotter in 1946-1947. He knew that Kelsh was trying to cover a large diapositive; he thought of it as a very large Multiplex of the double direct projection type. At that time Kelsh was trying a very large condenser; a fixed light source. Nevertheless the witness was interested, as he thought it might become competitive, or that Bausch and Lomb might buy into it for manufacturing purposes. There was need for a Multiplex type instrument that would not need a reduction printer. In his opinion, the Kelsh instrument as perfected “more than met the need”; it permitted the production of larger scale maps than could be made by the use of the Multiplex because of greater detail and a more highly magnified image. This permitted the Kelsh plotter to be used on much plotting work where Multiplex could not meet the contract scale requirements.

On cross-examination the witness testified as to the prior art. He agreed, as defendant’s counsel had conceded, that every part of the Kelsh device was known somewhere in the prior art. He further testified, however:

“This combination, we believe, was unusual. Let me state from experience, because I had the problem of developing new instruments at Bausch and Lomb, and we did not see this combination as a practical, working combination.”

In his opinion the prior art did not disclose “this particular combination.” ⁴² The parts could be found in the prior art “object for object” but not “function for function.”⁴³ He regarded Kelsh as second only to the Stereoplanigraph.⁴⁴

(b) When plaintiffs’ expert began intensive work in photogrammetry in 1950, the Kelsh plotter, patented December 20, 1949, was already a tool of the trade.

(c) In 1955, when plaintiffs’ expert was asked to expand the photogrammetry courses at Massachusetts Institute of Technology he thought it desirable to survey the field of private use of photogrammetrical instruments. The reports he received indicated that as of January 1956, the following photogrammetrical plotters were privately owned:

(d) The Kelsh machine came on the market commercially in about 1949. As of October 1958, Kelsh had shipped over one hundred machines in the past twelve months. During the preceding twelve months, O. M. I. had sold fifteen Nistri plotters. They had been on the American market for eighteen months.

(The court is familiar with the well established rule in this Circuit that “commercial success is a factor of probative value on the question, in close cases, whether a patent actually involves novelty and invention.” Hutzler Bros. Co. v. Sales Affiliates, 4 Cir., 1947, 164 F.2d 260, 267; Florence-Mayo Nuway Co. v. Hardy, 4 Cir., 1948, 168 F.2d 778, 781; B. F. Goodrich Co. v. United States Rubber Co., D.C.Md.1956,147 F.Supp. 40, 73, affirmed 4 Cir., 1957, 244 F.2d 468. The reference to commercial success is intended as illustrative of one element only of the criteria relative to invention, and is not intended as any indication that the court considers this to be a “close case.”)

(e) In an article in the Photo gram-metric Engineering Journal, June 1957, page 607, the Kelsh plotter was classed as the outstanding development in photogrammetry in the past ten years. For the development of the plotter Mr. Kelsh was in 1949 given the Fairchild Award by the Photogrammetric Society of America.

(f) In the division of this opinion dealing with infringement, the similarities of the Nistri device to Claim 1 of the Kelsh patent are considered. This Circuit has often emphasized the great weight, on the question of validity, that should be accorded the flattery of imitation; that the presumption of validity is further buttressed when the one attacking validity “gives the tribute of its praise to the prior art” but gives to the patent “the tribute of its imitation.” Diamond Rubber Co. of New York v. Consolidated Rubber Tire Co., 1911, 220 U.S. 428, 441, 31 S.Ct. 444, 450, 55 L.Ed. 527; Ackermans v. General Motors Corp., 4 Cir., 1953, 202 F.2d 642, certiorari denied 1953, 345 U.S. 996, 73 S.Ct. 1139, 97 L.Ed. 1403, rehearing denied 1953, 346 U.S. 842, 74 S.Ct. 16, 98 L.Ed. 362; Florence-Mayo Nuway Co. v. Hardy, 4 Cir., 1948, 168 F.2d 778, 782.

(g) The problem, the solution of which in plaintiffs’ 20-20 vision of hindsight (even without the use of ocular trains or auxiliary lenses) is stated as plain and clear from the prior art, was not solved by Bausch and Lomb, Zeiss or Nistri before Kelsh; although all three had produced Multiplex-type plotters ; the Stereoplanigraph was well known; Barr and Horner were available to them, and Nistri (O. M. I.) had produced plotters of the Stereoplanigraph, Multiplex and Gallus-Ferber types. It was only when the Kelsh plotter also became part of the “prior art”, and was admittedly known by Nistri, that the accused device was conceived.

In this connection the court is strongly impressed by the failure of Umberto Nistri, the president of O. M. I., to testify, although he was in court during the entire trial. If the problem were as simple, and so clearly anticipated, as his counsel argued, the court would have expected Nistri to explain how and when he developed the accused device; how it had naturally flowed from his company’s own Stereoplanigraph, Multiplex and Gallus-Ferber machines and his knowledge of Barr and Horner; and how it “just” (or “inevitably”) happened to come out like Kelsh — but later.

(h) “A patent shall be presumed valid. The burden of establishing invalidity of a patent shall rest on a party asserting it.” United States Code, Title 35, § 282.

This burden has been said to be “a heavy one, as it has been held that" ‘every reasonable doubt should be resolved against * * *.’ ” the party asserting invalidity. Mumm v. Jacob E. Decker & Sons, 1937, 301 U.S. 168, 171, 57 S.Ct. 675, 676, 81 L.Ed. 983.

Plaintiffs claim that the presumption of validity from the issuance of the patent is “destroyed” by the fact that in the prosecution of the Kelsh patent the Examiner was misinformed by defendant’s counsel as to the showings of the Barr and Horner patents, and acted on the basis of such misinformation. Plaintiffs do not contend that there was fraud in the obtention of the patent, of a nature that would invalidate it, but that the conduct (argument) of Kelsh’s counsel destroys the prima facie presumption of validity from the issuance of the patent.

There is a seriojis question whether arguments of attorneys in the prosecution of a patent application may be considered in cases involving questions of patent validity. Denominational Envelope Co. v. Duplex Envelope Co., 4 Cir., 1935, 80 F.2d 186, 192-193. Especially would this be true where, as here, plaintiffs claim reliance by the Examiner upon such alleged misinformation, but have offered no evidence of such reliance in fact, or that the Examiner was deceived by any alleged misinformation. However, the court has carefully examined all of the alleged instances of “misinformation”, and finds as facts, and concludes as a matter of law, that the instances cited by plaintiffs in support of their contentions relate to matters which should have been resolved in defendant’s favor, or as to which reasonable doubt exists and which do not exceed the bounds of reasonable argument, particularly in view of the obscurity of the language of the Barr and Horner patents; and that there is no reason to assume that the Examiner was misled if in fact he was in error in his interpretation of any ambiguities.

For example, the Examiner initially held that the light sources of Horner were pivoted at the center of the lenses in Figures 1 and 2. Defendant’s counsel contended that in Figure 1 the light source would not pivot about the lens aperture (nodal point). Plaintiffs’ counsel argue that as a competent optical man would know that the light “should” so pivot, defendant’s argument was improper. The court has previously stated that in its opinion the physical demonstration at the final argument showed that the light sources would not so pivot (or rotate) on a device constructed as shown in Figure 1.

Similarly, plaintiffs criticize the characterization of Barr and Horner as optical train instruments. There can be no question, either from the patents themselves, or from the testimony, that at least certain applications of these patents require optical trains.

Likewise, criticism is made of the contention by defendant’s counsel in the Patent Office that Horner and Barr disclose viewing similar to two transparencies hung in a window; as long as there is any light outside, the images will be visible to the eyes, against a generally illuminated background. The court would be inclined to agree that this was an inept description of Barr; but would assume that the Examiner could equally detect the “transparency” of such claim. As to Horner, the patentee’s reference to the substitution of daylight, for lamps, in viewing certainly justifies the argument of defendant’s counsel, even if it does not validate it.

The court is entirely satisfied and finds that the Kelsh device is an embodiment of a true combination involving invention. Although the parts constituting the combination were old, they were not old in combination, and particularly not old in this combination. Traitel Marble Co. v. U. T. Hungerford Brass & Copper Co., 2 Cir., 1927, 18 F.2d 66, 68, certiorari denied 1927, 274 U.S. 753, 47 S.Ct. 765, 71 L.Ed. 1333; Metropolitan Device Corporation v. Cleveland Electric Illuminating Co., 6 Cir., 1929, 36 F.2d 477, 479; Loom Co. v. Higgins, 1882, 15 Otto. 580, 105 U.S. 580, 591-592, 26 L.Ed. 1177; Hoeltke v. C. M. Kemp Mfg. Co., 4 Cir., 1935, 80 F.2d 912, 917, certiorari denied 1936, 298 U.S. 673, 56 S.Ct. 938, 80 L.Ed. 1395; Samuel M. Langston Co. v. F. X. Hooper Co., D.C.Md.1934, 8 F.Supp. 613, 617, affirmed 4 Cir., 1935, 79 F.2d 992; Black & Decker Mfg. Co. v. Baltimore Truck Tire Service Corp., 4 Cir., 1930, 40 F.2d 910, 912, 914; Kendall v. Trico Products Corp., 6 Cir., 1929, 31 F.2d 522, 524-525.

Now that the Kelsh combination has succeeded⁴⁵ it is argued that it is obvious. But it was not obvious to Zeiss, Baush and Lomb, or Nistri, leaders in the optical and photogrammetric field. The Kelsh combination far transcends the art of a skilled mechanic, and in the court’s opinion, and it so finds, amounts to a genuine invention.

Infringement.

Plaintiffs contend that in a suit brought to have a patent declared invalid and not infringed, the burden of proof of infringement rests upon the defendantpatentee. Illogical as this seems to the court, there is some support for it. Philip A. Hunt Co. v. Mallinckrodt Chemical Works, D.C.N.Y.1947, 72 F.Supp. 865, 875, affirmed 2 Cir., 1949, 177 F.2d 583.

In this case the question of burden of proof is academic. If it rests on the defendant, that burden has adequately been met.

In advance of the pre-trial conference held in this case, defendant’s counsel submitted a memorandum, in which he analyzed the elements of Claim 1 as follows:

“Elements of claim

“1. An instrument for making maps by stereoscopic photogrammetric methods comprising

“2. a pair of projection lanterns

“3. each having a lens

“4. and slide receiving means

“5. and supported side by side for projecting superimposed images of a pair of consecutive slides

“6. a movable plotting table

“7. having a screen on which the images are projected for viewing to give a stereoscopic model

“8. a point source of light for each lantern

“9. and a light condenser to converge the light through a small area of the slide and to a point coincident with the nodal point of the lens and diverge the light rays to cover approximately the screen of the plotting table

“10. each of said light source and light condenser being movably mounted relative to the lens in such manner that substantially the entire area of the slide in each lantern may be traversed by the converging light rays, and means for moving each light source and its light condenser so attached to the plotting table that as the plotting table is moved about, the light source and condenser move in a manner to maintain the image on the screen and to maintain the point of convergence of the light rays coincident with the nodal point of the lens.”

At the pre-trial conference, plaintiffs’ counsel agreed that elements 1, 2, 3, 5, 6 and 7 above are found in the accused device. The court is of the opinion and finds, from the testimony and its examination of the accused device, that the remaining four elements, or their exact equivalents, are also found in the accused device. Briefly considered: i

4. Plaintiffs contend that “slide receiving means” are claimed as a separate element in Kelsh but are not a separate element in Nistri; that in Nistri the glass plate or diapositive rests directly on the projection cone. The court finds' as a fact that the diapositive does not rest directly upon the cone in Nistri, but is supported by four members. The point in any event seems captious. Neither Nistri nor any other plotter could work unless it “receives” a slide; and the means by which a slide is received, are “slide receiving means.”

8. and 9. Kelsh calls for “a point source of light for each lantern and a light condenser.” Kelsh uses a light bulb with a condensing lens. Nistri uses a light bulb with a condensing mirror.⁴⁶ A condensing lens and a condensing mirror are merely different forms of “a light condenser.”

9. and 10. Kelsh calls for the convergence of rays to a point coincidental “with the nodal point of the lens.”

In their brief after the conclusion of testimony and before final argument, plaintiffs’ counsel said that in Nistri the light is converged at the image of the aperture. “The image of the aperture is almost, but not quite the same physical place as the nodal point.”

In his opening statement at the final argument, plaintiffs’ counsel said:

“I have since talking the last time over here and more discussions with the experts, come to the belief that this distinction is not as important as I once thought it to be, so that this point will not be too important * * * ’> ⁴⁷

The purported distinction between this element of Claim 1 and the Nistri device was not further pressed; and the Court finds that this is not a basis for avoiding infringement.

10. Kelsh claims means for moving each light source and its light condenser so attached to the plotting table that as the plotting table is moved about, the light source and condenser move so as to maintain the image on the screen and in focus. Kelsh and Nistri both employ means to accomplish this purpose. Kelsh uses two space rods; Nistri, one. Practical advantages are claimed for the Nistri device over the Kelsh device in this respect but each is obviously a “means” for the same end; and Claim 1 clearly reads on both.

The court finds as a fact and concludes as a matter of law that the accused device infringes Claim 1 of the Kelsh patent.

Summary

Claim 1 of United States Patent 2,-492,870 is valid and infringed; plaintiffs’ complaint is dismissed; defendant’s counterclaim for a declaration of validity and infringement is granted as to Claim 1.

The foregoing opinion embodies the court’s findings of fact and conclusions of law under F.R.Civ.P. 52(a) 28 U.S.C. but either side may submit requests for other or more detailed findings.

An appropriate decree will be entered upon submission.

1 . O.M.I. is an American corporation engaged in the importation and sale in the United States of the products of Ottico, an Italian corporation engaged in the manufacture and distribution of a number of products including aeronautical instruments, diesel engine parts, cryptographs, and photogrammetric projection apparatus, one of which is the source of this litigation. For convenience the two plaintiff will usually be referred to collectively as O.H.I., or Nistri (from the name of an Italian firm long prominent from at least as early as 1919 in the photogrammetric field), and whose present president, Umberto Nistri, was present in court throughout the trial, but did not testify.

2 . Harry T. Kelsh is the patentee of the the patent in suit, and Kelsh Instrument Company, Inc. is the owner of the patent rights by assignment from Harry T. Kelsh. Service was not obtained upon Mr. Kelsh, and he was dropped as a defendant. He testified in the trial. The defendant will ordinarily be referred to as Kelsh.

3 . The instant litigation arose out of an investigation by the Customs Bureau of the Treasury Department under U.S.C.A. Title 19, § 1337, instigated by Kelsh, into the importation by O.M.I. into the United States of instruments aEegedly infringing the Kelsh patent. This investigation may have some significance on the burden of proof as to vahdity of the patent, but in the view taken by the court, it is not a controlling consideration.

4 . B. B. Talley, Engineering Applications of Aerial and Terrestrial Photogrammetry, p. 1, 1988, Pitman Publishing Corporation — see also, O. von Gruber, Photogrammetry, p. 1, 1942, American Photographic Publishing Co., being a reprint of the 1932 volume printed in England.

5 . Talley, op. eit., credits Aristotle with the first recorded reference to the optical projection of images about 350 B.O. In 1832 Wheatstone experimented with stereoscopy and constructed the “first of the present type of mirror stereoscope in 1838 * * * Shortly thereafter came the first prismatic stereoscope of Brewster (1844) and that of Helmholtz, (about 1852) and in 1858 d’Almieda in Paris demonstrated the principle of dichromatic projection.” (pp. 1-2).

6 . Statement by counsel for Kelsh, Transcript 42.

7 . Plaintiffs’ brief, pp. 4-5.

8 . Plaintiffs’ brief, p. 5.

9 . Transcript 51-52.

10 . The Scheimpflug experiment was discussed by plaintiffs’ expert, Transcript 61-67, largely from an article by W. Sander of the Zeiss firm, found in von Gruber, op. cit., particularly pp. 159-103. The paragraph in the opinion to which this footnote is attached and the following one are summaries of these sources.

11 . Movement of the screen would disclose where the rays intersected. ' If the location of the screen did not coincide with, for example, tire intersection of the rays from the right corner of the house model, the viewer would see two corners, one coming from the right and one from the left projector. By moving the screen, the viewer could ascertain where the intersection actually occurred; and thus trace off an orthographic projection of the optical model of the house.

Presumably measurement of the distance the screen had to be moved to produce intersection of any other portion of the optical model would allow a determination of the relative distances between such portions.

12 . An “instrument model” was defined by plaintiffs’ expert as “the aggregate of the intersecting directions” or “the aggregate of all the ray intersections” which reconstruct a model of the original object (Transcript 70-71).

13 . Electronic scansion, by matching or comparing images, is now being investigated.

14 . Plaintiffs’ expert states the images “would fall on top of each other so it would appear to stand still.” Transcript 77. Apparently the witness, and certainly plaintiffs’ counsel, construed this as superimposition. Webster’s Unabridged International Dictionary defines “superimpose” as: “To lay or impose (one thing) over or above; —usually with on or upon * * *"

It would seem to the court that one image could not be superimposed upon a preceding or following image. Since plaintiffs’ expert testified that an application of the flicker system (GallusFerber) was the closest in the prior art to the Kelsh plotter, the correct interpretation, while not dispositive, is of more than semantic significance, and will be further discussed.

15 . The Kelsh patent, col. 1, 1. 29, refers to the use of polarizers. Kelsh’s expert testified that polarization was practical on simple devices for direct viewing, but not in projection instruments of the Kelsh type (Transcript 442-444).

The Kelsh instrument and the accused device both use the anaglyphic system. There was no testimony in the case as to any photogrammetric plotting device presently in use that employs polarization.

16 . Motion picture projection is at approximately this rate.

17 . This model in the mind is a subjective model of the real object, and is different from the physical ray intersection, the instrument model.

18 . Just as, when staring into space, if one finger is held in front of the eyes at close range, there will appear to be two fingers — one the image that the right eye is seeing, and one that the left eye is seeing.

19 . As in the Multiplex type projector, later to be discussed, developed by Zeiss in 1932 and Nistri in 1934.

20 . The so-called Porro-Koppe system.

21 . The Zeiss Stereoplanigraph of 1921 is an example.

22 . One looking directly into the lens of a projector -would be viewing by transmitted light; observation of the projected image on the screen would be by diffused light.

23 . von Gruber, op. eit. 304, states that “On observing a screen image at a distance of 25 cms. with unaided eyes, for example, there reaches the eyes one-twenty thousandth part only of the amount receivable by transmitted light.” The diffusion screen viewing is relatively simple and inexpensive. Nistri has a device making use of a pure optical transmission, which sells for approximately ten times the price of the accused device. However: “The difference in accuracy between these instruments is very slight and depends perhaps as much on the operator as any other factor.” (Transcript 123; Nistri’s expert).

24 . Barr and Horner are cited as references at the end of the Kelsh patent.

25 . The admission by Kelsh’s counsel that all the elements in Kelsh’s patent can be found somewhere in the prior art (without concession that all can be found in the died art) perhaps increases the burden of analysis, to determine whether these elements are found in isolation; and whether the art would suggest, to one skilled in it, the alleged invention of Kelsh.

At the conclusion of the trial, the court suggested that plaintiffs’ counsel in their brief should include short statements descriptive of the Multiplex, Stereoplanigraph and Gallus-Ferber devices, and the Barr and Horner patents, insofar as they were deemed by counsel to be relevant. This invitation was not accepted. The court accordingly prepared notes of its gleanings with respect thereto from the testimony and exhibits. From these notes the court at the final argument stated its understanding on the matters, and gave counsel for both sides an opportunity to concur or disagree.

The following summaries are a synthesis of the foregoing.

26 . Transcript 244-248.

27 . Talley, op. cit. 517.

28 . Talley, op. cit. 516-522, which in turn is based upon a translation of “The Problem of Aerial Photography” in Revue Des Forces Adriennes, March 1933, by Commandant E. Robini of the French Air Force; “The Obtaining of a Topographic Map Purely by Photography” in Science et Industries PhotograpMques, Second Series, Volume VII, pp. 153-160, May 1936, by E. Ferber; and “Notions de Photogrammetrie Terrestrie Adrienne”, J. B. Bailliere et Fils, Paris 1934, pp. 247-254 by Ch. Abdullah.

29 . Talley, op. cit. 517.

30 . A one-eyed person could function perfectly in its use.

31 . Talley, op. cit. 518, uses language which became the source of considerable discussion in the case. He refers to the necessity for an auxiliary system of sliding lenses in order to focus for projection on the screen. He then states;

“As this system has a certain distortion it is necessary to project only the ray which corresponds to the point to be traced along the axis in which the distortion is zero.”

Reading this literally, defendant’s counsel insists that the picture was so distorted that only one ray from each projector could be used. (The same argument is made by defendant’s counsel as to the Horner patent). However, from the testimony and exhibits the court has the definite impression that an image is seen. The reference to the central ray would seem primarily to relate to focusing. If the central ray of each projector met at the arrow, the other rays, constituting the base of cones projected not at right angles to the plane of observation, would not be exactly centered mathematically or geometrically; but for practical purposes the portion of the picture immediately adjacent to the intersection of the two rays woulé. be in focus.

32 . The witness admitted that it was extremely doubtful if stereoptic viewing could be obtained unless the flicker rate was increased from two per second to sixteen per second.

33 . Barr, Patent No. 1,655,306 p. 2, Is. 109-112.

34 . Talley, op. cit. 532. Because of its bulk, the Barr and Stroud device is often referred to as the “Big Bertha.”

35 . Barr, Patent No. 1,655,306, p. 2, Is. 40-52; p. 4, 1. 128; p. 5, 1. 3.

36 . Homer, Patent No. 2,085,498, p. 3, col. 1, Is. 3-4.

37 . The Stereoplanigraph, which can be operated by daylight, views by transmitted light through optical trains.

The Kelsh and Nistri machines were set up near a blank wall in the court room, about forty feet from the windows. With no artificial illumination it was possible, by bending down and looking up through one of the projection lenses, to see the image of the diapositive, by transmitted light. No visible projection could be observed on the platen.

38 . Horner, Patent No. 2,085,498, p. 2, col. 2, Is. 3,13, 65; p. 6, col. 2,1. 3.

39 . Page 2, col. 2, 1. 62; p. 4, col. 2, 1. 48; p. 5, col. 1, 1. 3; p. 5, col. 2, 1. 67.

40 . Page 4, col. 1, Is. 63, 69; p. 4, col. 2, Is. 39, 50; p. 5, col. 1, Is. 16, 22.

41 . Page 2, col. 2, Is. 5-15.

42 . Transcript 454.

43 . Transcript 457.

44 . Transcript 504.

45 . That the Kelsh device is useful and good was conceded at the final argument.

46 . The vice president of O. M. I. testified: “We do not use lenses to condense the light rays. We use a section of an ellipsoid mirror.” (Transcript 590).

47 . Plaintiffs’ expert, referring to The Principles of Optics, Hardy and Perrin, McGraw-Hill Book Company, 1932, p. 62, testified as follows (Transcript 140):

“Now, if we consider a ray entering the lens, having an angle Theta with respect to the axis of the lens, the same ray leaving the lens with the same angle with respect to the axis, the nodal points are tlie intersection with the lens axis of the prolongation of the entering ray and the emerging ray, so this point where I have extended that ray or until it intersects the axis is the nodal point on object side, where the extension of the ray back to the axis on the other side is the second nodal point.”

He further testified (Transcz-ipt 142) : “The nodal point is primarily a cardinal point of the lens used by lens designers and people in optics.”

He further testified (Transcript 148), that the light does not actually pass through the nodal points, but between them.