Method for manufacturing pictures with depth and pictures manufactured with this method

Method for producing images with depth and images made with this method.

The invention relates to a method for producing images with depth. The invention also relates to images made with such a method.

Various systems are known for displaying images with depth. For example, there are holograms in which a three-dimensional image can be projected into space using laser equipment. However, the technology for producing such spatial images with holograms is complicated and expensive.

Another well-known technology for producing stereoscopic images is based on how the human eye sees depth. The human eye sees depth because the left eye and the right eye perceive an object from a slightly different angle. These two images of the same object are perceived in the brain as an image with depth. A conventional technique for displaying images of an object with depth based on this is based on two images of the object, each taken from a slightly different position, the difference corresponding to the difference between the positions of a left and right a human's right eye. When these two images are presented separately to the left and right eyes respectively, this is perceived as a real perceived image and not from a flat-plane image, such as a print of a photo, including the depth that is customary in perception an image is experienced is also seen when observing these two images. This concerns images in a flat surface that together give the experience of a real perceived image. The presentation of the image for the left eye on the left eye and the image for the right eye on the right eye can take place in various ways. The so-called stereoscopic viewer has been known for a long time, wherein in a viewer comprising two eyepieces an image is placed for each eyepiece which is received from a field of view corresponding to the field of view of the eye in question. By placing a partition between the two images, the left eye can only see the image for the left eye and the right eye can only see the image for the right eye. Because these images are original images taken from mutual positions corresponding to the mutual positions of the left and right eyes, the image thus generated gives the same experience of depth as if the original image were observed. Another way of separating the two images for the left eye and the right eye, respectively, can be by providing the images with different polarization filters, after which viewing the images with glasses with differently polarized glasses can only see the image for the left eye for the left eye and the right eye can only see the image intended for the right eye, whereby again the image is experienced as an image with depth. In both cases, however, the observer can only observe the image if he uses special glasses or special tools for this, which entails a clear limitation.

Another known method of generating images with depth based on the above-described depth effect of the human eye consists of pasting the image in small strips next to each other for the left eye and for the right eye, and to arrange a grid for these small strips. place. This grid is opaque with a transparent slit above each strip, such that this grid largely covers the strips of photos below, but each of the strips is always visible through the slit, so that when the image is perceived at the correct distance the left eye see only the strips of the photo corresponding to the position of the left eye and the right eye only see the strips of the photo corresponding to the position of the right eye, so that here too the image is experienced as an image with depth.

A disadvantage of this latter method is that with the current state of the art of printing it is very difficult to prevent the strips of the photo belonging to the left eye from overlapping with the strips of the photo belonging to the right eye. A solution for this would be to increase the resolution of the printing device so that such an overlap will no longer occur, or only on such a small scale that it is no longer visible to the human eye. The printing technique is based on the maximum visibility of an eye of 300 points / inch (dpi). Although this in itself is correct, it appears that with differences perceptible, its order of magnitude is smaller, a printing unit of 4000 dpi appears to be a difference in the perception of differences, for the human eye having a maximum visibility. In a detachable way, the line thickness of 1/4000 inch is made by the human eye. Due to this sharp perception, attention was drawn to overlaps between the left and right eyes, as a result of which the depth of the perception of differences in images is lost. A solution for this could be the use of printing units with a considerably larger resolution. These are currently unavailable and even printing units with a resolution of 4000 dpi are very expensive.

It is an object of the present invention to provide a method for producing images with depth that can be produced with a printing unit with a resolution that does not have to be exceptionally large.

This object is achieved by a method according to claim 1. The division of the surface of the representation to be imaged into a predetermined number of equal first quadrangles, according to step c) of claim 1, takes place with the aid of simple arithmetic operations. As a result, the position of each rectangle is precisely known and thus with basically unlimited accuracy. Because the pixel representation of each photo is projected onto the series of second rectangles associated with this photo, removing pixels and parts of pixels that fall outside the relevant rectangles, this basically retains unlimited accuracy of the boundaries of each rectangle. As a result, the undesired overlaps can be reliably avoided and a printing unit with an exceptionally large resolution is not necessary.

By generating more than two photos in pixel representation of the representation to be displayed, wherein the mutual recording positions of two photographs each with neighboring recording positions correspond to the mutual positions of the two eyes of an observer, the method according to the invention can produce an image obtained with depth, wherein with a varying observation position of the image also the different visible parts of the object to be displayed become visible in the image. In addition to the depth effect that occurs at every position of observation, this gives a three-dimensional rendering effect.

The method is particularly easy to implement when the third and fourth quadrilaterals are rectangles.

However, it may occur that a straight line that forms a boundary between a third and a fourth rectangle jumps once during printing. In view of the aforementioned sensitivity of the human eye to abnormalities, such a one-off offset of a straight line forming boundaries between the permeable and the impermeable part of the grid will be perceptible to the human eye. By designing the third and fourth quadrilaterals as trapezia or diamonds with angles that have a small deviation of 90 °, the boundary lines of the grid will frequently jump when printed. Because this offset does not occur once, but regularly and regularly, it is not experienced as abnormal and will therefore not be noticed with the resolution of the printing unit that is above the maximum view of the eye.

Because the third quadrilaterals of the grid are light-impermeable or poorly transmissible, the amount of light that can fall on the image of the first print is limited. To this end, it is possible with an external light source, which is not further described, to supply light to the image on the first print. This exposure can take place from the side, but with a certain transparency of the image also from the rear.

If the third quadrilaterals contain identical areas at mutually regular distances which are left transparent during step h), non-transparent areas at regular distances create deviations which, due to their mutual regularity, will not be noticed by the human eye, but which will cause greater access of light to the print of the image and thereby increase visibility.

The light output for the print of the image can also be increased if the third quadrilaterals on the sides which after mounting according to the step i) have been made reflective. Light that would otherwise be absorbed by this back is reflected to the first print of the image, thereby increasing its illumination and hence visibility.

This increase in visibility can also be achieved by providing a separate reflection screen between the grid and the image, as described in claim 7. This reflection screen can comprise reflecting parts which are strips or equal to the covering quadrangles (third quadrangles ) of the grid, but may also be slightly smaller in width. With a smaller width, for example, being strictly in register is a less strict requirement, while an increased light output nevertheless results.

With the method of claim 8, it is possible to apply the entire image with depth to a transparent sheet. The image consisting of the photo segments is then printed on the back and the grid on the front. This printing of both prints takes place in register. The thickness of the transparent sheet determines the mutual distance of both prints. This achieves the advantage that the image as a whole is easily transportable. For example, images on relatively thin transparent sheets can be rolled up. Examples of useful material for these sheets can be polyvinyl and polyester.

It may happen that the illumination of the photo segments on the first print is insufficient. This can happen, for example, when there is insufficient daylight and no additional lighting from the side or from behind, or when this lighting from the side or from the rear has fallen out or fails in comparison with daylight. With the method according to claim 11, the image is placed in parts on the viewing side of the third quadrangles, the opaque part of the grid. In these cases, the segments on the grid serve to ensure that the image remains visible, although not with depth. If there is sufficient light for the segments of the first print, then the photo segments on the grid can contribute to the depth effect of the image.

By basing the first and the third number of the quadrilaterals on the resolution of the printing unit, for example such that the width of a quadrilateral always corresponds to a whole number of dots of the printing unit, the number of undesired offsets can be kept to a minimum.

The invention will now be further described with reference to examples of embodiments. Partly based on the drawings where

FIG. 1 is a schematic representation of the principle of an image with depth using a grid.

FIG. 2 is a schematic representation of the surface of the representation to be displayed divided into first and second rectangles;

FIG. 3 is a schematic representation of the photograph of the representation to be imaged from the position of the right eye of an observer;

FIG. 4 is a schematic representation of a photograph of the representation to be imaged from the position of the left eye of the observer of FIG. 3;

FIG. 5 is a schematic representation similar to FIG.

1;

Figs. 6a-f schematic representations of non-transmissive quadrilaterals;

FIG. 7 is a schematic representation of an image according to the invention provided with a reflection screen;

FIG. 8 is a diagrammatic representation of an image according to the invention executed with photo segments on the grid.

First, the principle of an image with depth will be explained with the help of a grid. This is followed by a discussion of an embodiment of the invention with some variants thereof.

Fig. 1 schematically shows the principle of an image with depth of an unspecified object. Two photographs of this object were taken from two different positions that approximately correspond to the positions of a left eye and a person's right eye. From each of these photos, narrow vertical consecutive strips are cut out at regular intervals and these strips are stuck next to each other in order on a surface, with the strip of the photograph corresponding to the position of the right eye to the left of the strip of the photo corresponding to the position of the left eye. Thus, a photo has been created of substantially the same size as each of the two original photos, with approximately half of the original photos being represented, alternating strips. In Fig. 1 this composite photo is indicated by numeral 1. Only a part of Fig. 1 is shown in Fig. 1, namely a part with two strips of each of the two original photos. From the photograph corresponding to the position of the right eye, the strips R1, R2 are shown, with to the right the strips L1, L2 that originate from the photograph corresponding to the position of the left eye, where R1 and L1 are strips of the same position on their original photograph and R2 and L2 are also strips of the same position on their original photograph.

Grid 2 is mounted at a distance v from photo 1. Grid 2 is divided into strips 3 that are not permeable to light and slits 4 that are permeable to light. The strips 3 which are not transmissive to light should also not be substantially reflective on the side facing the observer, as otherwise these reflections will attract the observer's attention too much and thereby distract from the depth effect. It will be clear that these are gradual concepts. The gaps 4 are located for associated strips. So there are as many slits as there are strips per photo. So there is a strip of every photo in the photo series under each slit of the grid. Strips associated with a slit of the grid are contiguous, but need not be associated with successive slits of the grid. However, in practice this will be the case because this keeps the distance between two consecutive slots in the screen as small as possible, which benefits the light output and also the resolution.

At a distance b from the grid there is an observer with a left eye 6L and a right eye 6R. What this observer sees from the image is represented by two light rays represented for each eye with the number 5. For the left eye 6L the light rays 5L and for the right eye 6R the light rays 5R. It is hereby clarified that due to the presence of the grating the left eye only sees the strips L1, L2 of composite photo 1 originating from the photo corresponding to the position of the left eye, and the right eye sees only the strips R1, R2 of composite photo 1 from the photo corresponding to the position of the right eye. Because the observer sees with his left eye only the image of the photograph corresponding to the position of the left eye and the right eye only sees the image of the photograph corresponding to the position of the left eye, the observer perceives the depicted object as if he were the object in reality, with depth.

It will be appreciated that when, in addition to the photos mentioned here, photos are taken from a different position, e.g. shifted slightly to the left, and these photos are distributed in the same manner as described above and the strips are pasted next to the strips shown in Fig. 1 shows that the observer of Fig. 1, when it moves slightly to the left, then sees the strips from the second pair of photographs. This observer then sees the same object with depth, but from a slightly different angle of view, as a result of which a spatial effect is obtained through a photo montage on a substantially flat surface.

It is advantageous to make the choice of the widths of, for example, the strips 3 and the slits 4 depending on the resolution of the printing unit. For example, with a grid of 40 slits per inch and with a subdivision of 10 photo strips per slit, the image will only come into its own when both print 1 and print 2 are printed on a printer with a resolution of 400 dpi, 800 dpi or

1200 dpi etc.

The principle described above is the principle on which the method according to the description will also know of the photos and clearly the accuracy serves. If there is invention based. From these are that creating the strobe of the grid with large to happen. This is where the inventive irregularities occur, for example, the width of the strips, and in the boundary lines of the strips, the attention of and irregularities than the through the slits of the grid, rather than the observer, is rather determined by these images on the strips. , whereby the depth effect is partially canceled out. An important source of apparent irregularities is that pixels of a photo rarely or never fit on a whole number of lines of a printing unit, causing numerous offset

The starting point is the pixel representation of the mutual recording positions of the photos with the mutual recording positions of occurrence.

a first to be pictured pair of two photographic images, of which a pair of two eyes of an observer matched. When multiple photos are used, they are processed in a similar manner, clear to a person skilled in the art. These photos are stored in a computer's memory. Subsequently, a width and a height of the representation to be displayed are determined and the limits of the representation to be displayed are marked in the photos stored in the computer memory. A length and a width of the image are hereby determined and also of both photos a length and a width of the part that the image will form is determined. For simplicity, it is assumed here that the length and width of the image is the same as the length and width of the photos. This of course does not have to be the case. The invention does not change if there is a certain ratio between the length and width of the image and the length and width of the photos. The following paragraph explains that the pixel representation of a photo is mapped onto the relevant rectangles by projecting the pixel representation onto these rectangles. This projection can take place with a magnification factor or a reduction factor such that the image is given the desired format.

FIG. 2 shows a representation of the surface of the image to be obtained divided into a predetermined first number of equal first rectangles, each rectangle covering the full height of the representation to be displayed. Only the first two of these first rectangles a, b are shown in Fig. 2. Next, each first rectangle is divided into a second number of second rectangles, each second rectangle covering the full height of the representation to be displayed. This second number is equal to the number of photos in step a) and each second rectangle is assigned to a photo in a predetermined manner so that a second rectangle is associated with each photo in each first rectangle. Next, the pixel representation of each photo on the series of second rectangles associated with this photo is mapped by projecting the pixel representation on these second rectangles, removing pixels and parts of pixels that fall outside the relevant rectangles. For pixels from which the parts protruding outside the rectangles have been removed, the pixel value representing a color or a shade of gray is maintained for the remaining part. A sharply defined color or shade of gray is thus obtained for each of the rectangles, with no overlap between the rectangles. The determination of the boundaries of the rectangles is a purely mathematical matter and can be determined by calculation with basically unlimited accuracy. Projecting onto the rectangles and 'cutting' along the edges is also a mathematical operation and can also take place with unlimited accuracy.

FIG. 3 is a representation of the photograph corresponding to the position of the right eye and FIG. 4 is a representation of the photograph corresponding to the position of the left eye. In Fig. 3 the rectangles R1 and R2 are indicated which are projected onto the surface of Fig. 1 as indicated therein.

The generation of a print job for a printer from the collection of grayscale or color values ​​obtained by this image is a standard operation for a person skilled in the art and is not further described here.

The second print, that of the grid, is now produced in a similar manner. This print is made on a transparent surface. Non-light-transmitting quadrilaterals on the side facing the observer are printed on this substrate, which barely or barely reflective quadrilaterals each cover the full height of the image, so that light-transmitting gaps are formed between these quadrilaterals. Through these gaps, the observer's eyes perceive the strips of two photographs behind them, in such a way that the observer's left eye perceives the strips that originate from the photograph whose position corresponds to that of the left eye and the right eye of the observer observes the strips from the photograph whose position corresponds to that of the right eye. In the simplest case, the non-illuminating quadrilaterals are rectangles. The intermediate gaps are therefore rectangular in shape.

FIG. 5 shows a part of FIG. 1. FIG. 5 shows a first print 1, with two photo strips R1, L1, one of each photo, and a second print 2 with a light-transmitting gap 4. Print 1 and Print 2 are placed parallel to each other at mutual distance v. At distance b of print 2 there is an observer whose left eye 6L and right eye 6R are shown. The width of slit 4 is dimensioned such that for the left eye 6L through slit 4 only the sliver L1 that belongs to the photo corresponding to the position of the left eye, and for the right eye 6R through slit 4 only the slit R1 that belongs to the photo corresponding to the position of the right eye is visible. This is illustrated by the light rays 5L and 5R, which bound the field of view of the left eye 6L and the right eye 6R, respectively, through the slit 4. By reducing the width of the gap 4, an area of ​​distances b is obtained within which the depth effect remains perceptible.

If the contours of the non-light-transmitting quadrilaterals are not accurate, the contours may exhibit an irregular jump in the print due to the limited resolution of the printer. When a jump occurs in the contour of a slit in an irregular pattern, the observer's attention is drawn to this staggered slit and derived from the underlying photos, so that the depth effect is canceled out. It is therefore important that these gaps do not show any irregularities. If this nevertheless occurs or may be expected, it is soluble by making the non-translucent quadrangles not rectangles, but trapezia or diamonds, the angles of which differ slightly from 90 °. As a result, staggered lines are generated in the contour, whereby the stagger occurs regularly. An example of a trapezoidal quadrangle 3 is schematically shown in Fig. 6a. A regular offset of the printed line occurs on both sides. Because of this regularity this will not be noticed by an observer, so that his attention remains focused on the image behind the slits of the grid, so that the depth effect is retained.

The non-light-transmitting part of the grid limits the amount of light that reaches from the photos to the observer's eye. It is possible to increase the light output by providing openings in the non-light-transmitting part at regular distances. When these apertures are sufficiently small, they will not drastically influence the image formation, at least if the requirement that these apertures are not noticed by their regularity is met and thereby disturb the depth effect. In Figs. 6b and 6c two different forms of these openings 30, 31 are shown. In fig. 6b it concerns rectangular recesses 30 from the border with the gap 4. In fig. 8 the recesses 30 of fig. 6c still deepened with smaller rectangular recesses 31. Through these recesses 30, 31 the light output of the image on the eye 6L, 6R is increased. Figs. 6d-f show examples of regular small light-transmitting openings arranged in strips 3.

Another way to increase the light output of the image of print 1 is shown in Fig. 7. Fig. 7 shows a print 1 with strips of photos, not further indicated thereon. Furthermore, an impression 2 is visible with non-light-transmitting quadrangles 3 and light-transmitting gaps 4 between them. Placed between image 1 and image 2, preferably close to image 2, is a screen 10 with reflective surfaces 11 on the side facing print 1 and light-transmitting surfaces 12 between surfaces 11. The reflective surfaces 11 increase the light output by that light that would otherwise have been absorbed by non-light-transmitting quadrangles 3 of print 2, is now reflected and can eventually exit through slits 4. Instead of a reflective screen 10, the print 2 on the side facing the print 1 in the mounted state may be printed with reflective material.

The composition of the image by mounting print 1 and print 2 and possibly reflective screen 10 at a fixed distance and in register with respect to each other can take place in ways known to those skilled in the art.

One possibility is to mount the image in a fixed housing which is provided with suitable clamping devices for prints 1 and 2 (and 10), and with an exposure.

In series production of the image according to the invention, raster and photo, resp. imprint 2 and imprint 1 melted between glass. This can be matched to each other with very high accuracy of up to 0.01 mm accuracy.

However, it is also possible to assemble the image according to the invention to some extent autonomously, which can be advantageous for use in display cabinets as can be seen in many cities. On the prints 1 and 3, respectively. the photo and the grid are printed with registration marks. The photo is placed on a vacuum table and the grid sheet over it. The two sheets are precisely aligned on the registration marks. With a magnifying glass it can very accurately match the registration marks.

By clamping a beam over the middle, the sheets can no longer shift relative to each other. The grid sheet is turned over at the top. On the photo a strip of the desired thickness is glued to the top. The grid sheet is folded back and stuck on top of the heating. Below the same procedure. The clamp in the middle can be removed and the sheets stay nicely parallel.

Instead of gluing, a clamping device can also be placed above. Downstairs it is advisable not to pinch but only to hang a weight on each sheet. A replaceable assembly can hereby be obtained in a simple manner so that images according to the invention can be exchanged in a simple manner in public display cabinets.

In a particularly favorable embodiment of the invention, the first print and the second print are both printed on a sheet of transparent material, for example polyester or polyvinyl material. The first print is applied to a surface of the transparent material sheet and the second print is applied to the opposite surface of the transparent material sheet. The thickness of the transparent material sheet is then the distance v. This thickness is related to the application of the image with depth and to the number of strips 3 and slots 4 of the grid. For example, for a large poster, a grid of 40 slits per inch and a sheet thickness of approximately 3 mm may be used. For an A4 size image, a grid of 60 slits per inch may be used and a sheet thickness of approximately 1 mm. For an image of a credit card size, a grid of 120 slits per inch may be used and a sheet thickness of approximately 0.2-0.8 mm, and for an image of stamp size a grid of 200 slits per inch may be used and a thickness of the sheet of approximately 0.05 - 0.5 mm.

A further favorable embodiment of the invention is shown schematically in Fig. 8. For example, when there is insufficient daylight and there is no additional lighting from the side or from the rear, or when this lighting from the side or from the rear has fallen out or is not falling in relation to of daylight, it is possible to place strips 7 of a photograph of the image on the visible side of the opaque strips 3 of the grid. In these cases, the photo segments 7 on the grid serve to ensure that the image remains visible, although not with depth. If sufficient light is present for the segments of the first print 1, the photo segments 7 on the raster 3 do not contribute to the depth effect of the image, but they do enhance the image of the image.

The invention has been explained above with reference to a few examples of embodiments. The invention is not limited to these exemplary embodiments, but is defined by the appended claims.