Inkjet printers have become common place. Desktop sized machines are now found in many offices and homes. However inkjet printers are also now the printer of choice for many wide format printing applications. Such printers are typically used to print onto rolls of paper or some other substrate whose width can often exceed 1.5 metres. These printers are often used for specialist printing jobs such as the creation of art work on a one-off or limited print run basis.

Whilst inkjet printing technology is reliable and affordable, it does however suffer some drawbacks. The most significant of these is a restricted colour gamut. The inkjet printing process requires the deposition of carefully controlled amounts of ink through an exceedingly small orifice. The ink comprises coloured particles dissolved or suspended in a solvent or aqueous solution. The requirement that the coloured particles can repeatably be ejected through the orifice places restrictions on the size of particle which can be used.

This size restriction has in turn limited the choice of materials that can be suspended within the ink. This in turn has restricted the depth of colour of the inks. In practice this means that inkjet printers cannot reproduce some colours accurately, or alternatively there are colours which can be perceived by the human eye but which simply cannot be reproduced on an inkjet printer.

According to a first aspect of the present invention there is provided a printer including paper registration means for accurately positioning a substrate such that the substrate can be passed a plurality of times through the printer for a multipass printing process.

The applicant has realised that the depth of colour reproducible by an inkjet printer can be enhanced by multipass printing. In effect, multipass printing allows the density of the colour to be increased simply because the density of the chromogen (ie the compound that gives ink its colour) deposited per unit area on the substrate is increased. However, merely reversing the direction of paper travel within the printer to wind back to the beginning of the image and then reprinting it is unlikely to result in sufficiently accurate registration of the two images given that manufacturing tolerances of the rollers within the printer may cause the paper to become skewed or twisted, even by minute amounts, as it passes through the printer. Thus in order to achieve an increased colour gamut by virtue of multiple pass printing, it becomes necessary to accurately position the paper at the commencement of each pass.

Preferably the paper registration means comprises a mechanical alignment device for accurately determining the starting position of the substrate prior to each printing pass.

Advantageously the mechanical alignment device is a registration bar attached or attachable to the body of the printer.

Advantageously a matching registration punch is provided with the printer in order that holes corresponding to the size and position of registration pegs on the registration bar can be formed in the substrate. Thus the holes in the substrate can be brought into alignment with the pegs on the registration bar in order to position the substrate accurately. After such positioning has been achieved the paper guide mechanism within the printer, which typically comprises pinch rollers, can be activated to hold the substrate within the printer and then the substrate can be removed from mechanical engagement with the registration bar. The printing process can then be commenced and a first image laid down on the substrate. Following completion of a first pass of the printing process, the substrate can be repositioned using the registration bar in order to ensure that it is in precisely the same position and orientation with respect to the printer that it was during the first pass of the printing process. A second printing pass can then be performed. Further printing passes could be made if necessary or desirable.

As an alternative to the use of a punch to punch holes into the substrate, a registration plate or section having holes arranged to mate with the pegs on the registration bar may be removably attached to the substrate, for example by a clamp or by adhesive. Such a clamp may be a spring loaded clamp operating in a similar mode to a bulldog clip.

Preferably the substrate is in the form of a sheet of material, for example paper or card.

Preferably the substrate is supported in a plane substantially including the path of the print head. Thus if the substrate is supported prior to reaching the print head and after having passed through the print head bending of the substrate is inhibited and this in turn helps ensure that the distance between the substrate and the print head remains substantially constant. This can be advantageous in maintaining accurate registration between the first and second printed images.

The inventor has noted that simply double printing the same image onto the same substrate at the same position does not necessarily give rise to visually acceptable results. The increase in chromogen density on the substrate can result in the image looking brighter but does not necessarily give rise to an enhancement of the colour gamut. In order to fully realise the potential of multipass printing the original image needs to be processed in order to derive first and second full (and optionally subsequent) colour images, where the first and second (and optionally subsequent) images are non-identical with respect to their colour content.

Advantageously the printer includes image processing means for processing an input image in order to derive the first and second (and optionally subsequent) images for use in multipass printing.

According to a second aspect of the invention there is provided a method of controlling a printer for multipass printing, comprising the steps of : processing the target image to form first and second images that are to be printed with the second image superimposed on the first image; printing the first image on a substrate; repositioning the substrate; and printing the second image over the first image.

In a preferred embodiment of the present invention there is provided a method of controlling a printer for multipass printing, the method comprising the steps of receiving in electronic form an image to be printed; processing the image into first and second full colour images that are to be printed with the second image superimposed on the first image; printing the first full colour image; repositioning the substrate; and printing the second full colour image.

Advantageously the substrate passes in the same direction with respect to the printer during each image printing.

According to a third aspect of the present invention there is provided a computer program for controlling a printer to print in accordance with the method according to the second aspect.

The printer may be managed from a control application which has instructions for performing multipass printing. The modified printing may be controlled by the printer drivers or by way of software modules than co-operate with existing applications. Such modules are known as"plug-ins".

According to a further aspect of the present invention there is provided a printer wherein a paper transport mechanism therein includes position control means for positioning a substrate such that first and second images in spatial alignment can be printed with the second image superimposed on the first image.

The present invention will further be described, by way of example, with reference to the accompanying drawings, in which: Figure 1 is a schematic cross section through a known wide format inkjet printer; Figure 2 is a perspective view of a registration plate secured to a printer just prior to registration with the substrate; Figure 3 is a schematic representation of an ink dot pattern typically produced by an inkjet printer; Figure 4 is a cross-sectional view showing the pattern of ink dots of Figure 3 when deposited on the surface of a substrate using a conventional inkjet printing process; Figure 5 shows the same view as Figure 4 for an image printed using a multipass print in accordance with the present invention; Figure 6 schematically illustrates a modified printer; and Figure 7 is a flow diagram of the image processing used to derive dissimilar images to take full advantage of the potential of multipass printing.

A wide format inkjet printer 1 is schematically illustrated in Figure 1. Such printers are commercially available from manufacturers such as Hewlett Packard. Typically paper 2 in the form of a roll 4 is held within the printer 1. The paper 2 is directed over a series of rollers in order to present it at the correct configuration for printing. In the example shown the paper 2 passes over a first roller 6 before being delivered to a pinch roller configuration comprising rollers 8 and 10. One of the rollers 8 and 10 is driven by a motor within the printer and the other one of the rollers 8 and 10 is urged against the other roller so as to frictionally engage the paper 2 therebetween such that it can be advanced towards a print head 12. The inkjet print head 12 receives coloured ink from an ink reservoir and deposits it on the paper in a known manner. Full colour printing can be achieved by the use of three inks which combine in a subtractive manner, and the inks have the colours yellow, magenta and cyan. However in order to get a better black, the print head also receives a supply of black ink. The print head 12 is mounted to a carriage in order that it can traverse the width of the paper 2 (that is move into and out of the plane of Figure 1) in order to deposit a swathe of ink on the paper in accordance with instructions received from a controller 14. After one swathe of the image has been printed the paper is advanced by the width of the swathe and then the print head traverses the paper once again. In this way an image is built up onto the paper in a raster scan manner. The same technique is used on desktop printers, although rather than receiving paper from a roll individual sheets are presented for printing via a paper handling mechanism.

It might be supposed that in order double print an image the direction of rotation of the rollers 8 and 10 is simply reversed to retrieve the paper back into the printer, and then printing can be recommenced. However in practise this is not easy. Any gearing system which is to have tolerable amounts of friction therein exhibits some degree of backlash or play. This means that if the paper is advanced by a fixed amount, then reversed by the same distance the effect of the backlash within the gears means that the paper is not accurately positioned at its initial starting point. Furthermore imperfections or tensions within the paper and imperfections within the rollers, such as being slightly conical, which may occur because the rollers are manufactured to a tolerance, also means that the paper may tend to seek to twist or skew with respect to the rollers as it passes through the printer and that the difference in forces exerted on paper as it moves in the forward and reverse directions means that the amount of twisting may not be completely reversible. The inventor has realised that registration means, for example a registration plate engaging with co-operating holes made in the paper can be used to accurately position the paper each time. Figure 2 shows an example of a registration plate which may be used with the printer shown in Figure 1 or with desktop style inkjet printers. The registration plate 20 comprises a base plate 22 which is securely affixed to a portion of the printer 24. Advantageously the portion 24 of the printer is itself rigid and immovable, at least with respect to the carriage mechanism for the print head 12, such that its position with respect to a start of print position of the print head 12 is well determined and repeatable. The registration plate 20 carries a plurality of engagement elements, in this case three pins 26 which stand proud from the plate. More pins could be provided.

A punch (not shown) provided with a printer matches the registration plate and can be used to make holes 30 in a sheet of paper (or other substrate) which is to be printed. Thus at the commencement of the printing process the paper 32 is fed through the printer, with the printer mechanism positioned such that the rollers 8 and 10 are not in engagement with one another, in order that the holes 30 can be placed over the pegs 26. This accurately defines a starting position for the paper. Once this position has been defined the printer mechanism is adjusted in order to bring the rollers 8 and 10 into contact with one another thereby pinching the paper 32 therebetween. Once this has been achieved the paper is lifted clear of engagement with the registration plate and printing process is commenced.

The paper is then fed through the printer in the normal manner. Once the first print is completed, the paper is removed from the printer and advantageously is allowed to dry.

The paper is then repositioned within the printer by introducing it along the normal paper feed path, between the rollers 8 and 10 (which have been moved apart) and brought into registration with the registration plate once more. Once registration is achieved the rollers 8 and 10 are moved into inter-engagement again, the paper is removed from engagement with the registration plate and a second printing pass is performed. The process can be repeated further if further printing passes are required.

Merely printing the same image twice without modification of the instructions to the printer does not realise the full potential of this process and does not give rise to the enhancement of the colour gamut that this process can achieve. This can be explained by considering the standard inkjet printing process. Although three coloured inks are provided, each ink has, by definition, only one colour. However it is necessary to be able to print a full tonal range in each of the ink colours. Suppose, for example, that we consider printing a magenta image. Every time the inkjet print head is operated it deposits the same amount of ink onto the paper as a substantially circular spot 40 as shown in Figure 3. The coloured spot 40 is surrounded by the substrate. Thus if the substrate is white and magenta spots 40 are only deposited infrequently such that the majority of the light reflected from the surface is white then only a light magenta colour is perceived.

However, if the spots 40 are deposited more densely then the proportion of magenta to the proportion of white increases and hence a deeper magenta is perceived. The white of a substrate can be blacked out by the printing of black spots as well and this allows the printing of a whole range of"grey scales"in a single colour, in this case from a very very light magenta, that is nearly white, to a very very dark magenta that is nearly black. The same holds true for each of the other colours.

In order to improve the perceived quality of the printing the spots of ink 40 are deposited in a dither pattern. This is well known in the field of printing.

Figure 4 shows a side view of the spots 40 of Figure 3. Each spot 40 effectively sits in a unitary layer on the surface of the substrate 32. Each spot when delivered from the printer contains a quantity of the chromogen, that is the substance which gives the ink its colour, conveyed within a solvent. The solvent evaporates leaving the chromogen, ie the colour, sat on the surface of the substrate. If double printing is allowed for, as schematically shown in Figure 5, it becomes apparent that one spot of ink, for example spot 42 can be printed over a pre-existing spot of ink 40. If the spots 40 and 42 are of the same colour then the density of that chromogen on that particular area of the substrate is increased.

However the spots 40 and 42 do not need to be of the same colour and hence subtractive mixing of the inks can occur. This process therefore allows any single colour to be more vibrant than can be achieved within a single pass print alone, and also allows the subtractive mixing to achieve different ink combinations which simply are not possible with a single pass.

Figure 6 shows a modified printer wherein a linear paper path from a paper storage or input region 50 to a paper delivery tray 52 is provided, with the paper passing between the pinch rollers 8 and 10. This paper path can be included as an additional path to the path shown in Figure 1 for a wide format printer. However the linear path is more typical of that already found in desktop inkjet printers. The printer is modified so as to include the registration plate 20 downstream of the print head 12 thereby allowing for accurate positioning of the paper prior to each pass. Once the paper has been fed from region 50 to 52, the paper can be returned to the region 50 and fed through the rollers 8 and 10 again, subject to re- registration having been performed.

Figure 7 is a flow diagram showing the image processing steps which may be performed within the printer or within a computer or other device arranged to send an image file to the printer for printing. Thus either the printer or the computing device initially receives an image at step 60. This corresponds to a target image that is to be printed. However it is also possible that the image 60 can be computer generated. From step 60 control is passed to step 62 whereas the image is processed in order to form first pass and second pass image files for printing wherein the first and second images are not identical. Typically (but not necessarily) the first image might form the bulk of the printed image whereas the second image might be thought as perturbation to the first image, wherein the perturbation provides the enhanced colour gamut. Depending on the implementation, the conversion of a first image to first and second images may be provided by pixel by pixel lookup in order to map colour values within the original image to appropriate colour values within the first and second images. Following generation of the first and second images the substrate is then positioned within the printer at step 64 and once positioning has been completed control is passed to step 66 where the printer is instructed to print the first image.

Following completion of printing of the first image control is passed to step 68 where the substrate is repositioned and from then on the second image is printed at step 70. Step 64 and step 68 require manual intervention. It also follows that step 66 could be performed before step 62. It is possible to extend this process to further passes through the printer.

The present invention can also be applied to other printer technologies such as wax printing and dye sublimation printing.

In modifications of this process, it is possible to place the registration plate so that it extends in the direction of travel of the paper. In this configuration the registration plate can be positioned in part of the paper inlet region 50. Additionally, where the printer includes an optical scanner the optical scanning function can be invoked to scan the image after the first print pass, to compare this to the original input image and thereby to calculate a difference image. The difference image can then be used to derive the image that should be printed on the second print pass in order to minimise the difference between the image that is printed and the image that was provided to the printer. Furthermore, the use of an optical scanner, or optical paper detection means may also be invoked in order to implement electronic registration of the paper. However such electronic registration requires the ability to detect the position of two or more points along a leading edge of the sheet of paper in order to check whether the paper is skewed and also one point along the side edge of the paper in order to check that the paper has not been laterally offset. These issues do not arise when mechanical registration is invoked.

In a further modification, the printer may include registration means as part of its mechanism. Thus the printer may accept sheets of unmodified paper and may form its own markers in the printer as the paper passes through the printer. This may, for example, be by the use of an electrically operable punch which can punch holes along the side of the sheet of paper. A toothed wheel within the printer may then engage with those holes in order to ensure accurate re-registration of the paper. Additionally and/or alternatively means may be included within the printer in order to reposition the paper to correct for skewing or lateral offsets. Such repositioning means may include pins movable to engage with the edges of the sheet of paper in order to push the paper into contact with an alignment guide.

Alternatively, if the printer manufacturer chooses to engineer the paper transport and feed mechanism to sufficiently high tolerances, then simple reversal of the print mechanism may still ensure correct positioning, or at least substantially correct positioning of the paper.

A further, and perhaps unexpected, benefit of the present invention is that it enables high quality images to be printed on lower quality papers. Typically photo quality inkjet paper is treated such that the ink does not seep into the paper but rather sits on the surface.

However with the multiple pass approach to printing the printing of the first image to some extent seals the surface of the paper thereby allowing the pixels of the second image to be more tightly defined spatially. This also gives rise to the possibility of printing the first and second images at different printing resolutions.

Thus an inexpensive improvement to inkjet printing is provided. Furthermore the inkjet printer operating in accordance with the present invention can give performance comparable to that of a flat bed printer and hence a significant space saving can be achieved.