Coloured fabric

Abstract

There is provided herein a method of producing a coloured fabric, the method comprising: providing a plurality of yarns, each yarn having one of a set of primary yarn colours; optionally, identifying a colour recipe for a desired fabric colour, the colour recipe comprising a primary yarn colour warp composition and a primary yarn colour weft composition; and weaving yarns of the plurality of yarns to form the fabric optionally according to the identified colour recipe. Coloured fabric obtained using the method is also provided herein.

Description

[0001] COLOURED FABRIC

[0002] The invention relates generally to coloured fabrics and a method of manufacturing coloured fabrics. More particularly, but not exclusively, the invention relates to coloured fabrics manufactured using specific combinations of coloured yarns.

[0003] Background

[0004] For conventional production of coloured woven fabrics, yarns are spun or extruded and woven together to form a fabric. After the yarns are woven to form a fabric, the resultant fabric is dyed in a water bath at high temperatures to provide a desired colour. However, conventional production of coloured woven fabrics requires vast quantities of fresh water and produces many pollutants, especially during the dyeing process.

[0005] Dope dyeing is a waterless for synthetic and semi-synthetic materials alternative in which colourants are added during the yarn extrusion process. This produces a plurality of yarns which are already dyed to the desired colour, meaning the fabric formed via the weaving process does not need to be subsequently dyed.

[0006] However, whilst dope dyeing significantly reduces the environmental impact during production of coloured fabrics, it is not suitable for all circumstances. The process is often expensive, and fabric manufacturers are often required to obtain large quantities of a single yarn colour in order to meet minimum order requirements and / or to be cost effective. Additionally, many colours are not compatible with dope dyeing and cannot be developed due to the nature of the pigments. Given that there is demand for a wide range of coloured fabrics, it is not feasible to obtain dope-dyed yarns with the required colour for every desired fabric.

[0007] The present invention was devised with the foregoing in mind.

[0008] Summary of Invention

[0009] According to a first aspect of the invention, there is provided a method of producing a coloured fabric. The fabric may be a woven fabric produced by weaving weft and warp yarns together.The method may comprise providing a plurality of yarns. Each of the plurality of yarns may have one (or more) of a set of primary yarn colours. For the avoidance of doubt, the term “primary” used herein does not refer to the traditional meaning of primary with respect to colours, and the “set of primary yarn colours” is not to be construed as exclusively referring to red, yellow, and blue yarns, or cyan, and magenta yarns. The set of primary colours may include any colour such as blue, red, yellow, green, orange, purple, pink, brown, black, white, grey, cyan, magenta, lime, teal, indigo, violet, maroon, navy, olive, turquoise, coral, peach, lavender, mint, beige, charcoal, ivory, taupe, khaki, rust, ochre, sienna, umber, etc, or combinations of one or more thereof.

[0010] In some embodiments, at least one of the colours of the set of primary yarn colours is not selected from red, yellow, blue, cyan and magenta. In some embodiments, at least one of the colours of the set of primary yarn colours is not selected from red, yellow, blue, cyan, magenta, white and black.

[0011] In some embodiments, the set of primary yarn colours does not comprise all three of red, yellow and blue colours (i.e. the set of primary yarn colours may comprise none, one or two, but not all three of red, yellow and blue). In some embodiments, the set of primary yarn colours does not comprise all three of cyan, magenta and yellow colours (i.e. the set of primary yarn colours may comprise none, one or two, but not all three of cyan, magenta and yellow).

[0012] Therefore, in particular embodiments:

[0013] (i) the set of primary yarn colours does not comprise all three of red, yellow and blue; and / or (e.g. and)

[0014] (ii) at least one of the colours of the set of primary yarn colours is not selected from red, yellow, blue, white and black.

[0015] In particular embodiments:

[0016] (i) the set of primary yarn colours does not comprise all three of cyan, magenta and yellow; and / or (e.g. and)

[0017] (ii) at least one of the colours of the set of primary yarn colours is not selected from cyan, magenta and yellow.In more particular embodiments:

[0018] (i) the set of primary yarn colours does not comprise all three of red, yellow and blue; (ii) the set of primary yarn colours does not comprise all three of cyan, magenta and yellow; and / or (e.g. and)

[0019] (ii) at least one of the colours of the set of primary yarn colours is not selected from red, yellow, blue, white, black, cyan, and magenta.

[0020] The skilled person will understand that the coloured fabric may be of a colour which may appear to be different (e.g. under ambient conditions, during normal use of the colour fabric by the end user) from any of the primary colours of the yarn used to weave the coloured fabric. The colour of the fabric may be referred to as a “perceived colour” when the fabric is observed from a distance of at least 10 cm away, under standard ambient conditions. In some embodiments, such as when there is one yarn in the warp and one yarn in the weft, the perceived colour may be observed from a distance of at least 1 cm away, or at least 2 cm away, under standard ambient conditions.

[0021] The resultant colour of the coloured fabric may be a solid colour or it may comprise visual surface effects of different perceived colours. A “solid colour” refers to a single (i.e. perceived single), uniform colour that has no variation in shade, tint or pattern across its surface; a solid colour consistent and uninterrupted, making it distinct from gradients, patterns, or textured colours.

[0022] Therefore, in some embodiments, the coloured fabric is, or comprises, a solid colour.

[0023] In some embodiments, the coloured fabric comprises visual surface effect. “Visual surface effect” includes gradients, patterns (including small blocks of repeating colour such as check patterns, or lenticular patterns), stripes, dotting or dashed lines, textured colours, birefringence or angle dependent colour. Suitably the visual surface effect, such as stripes, is of at least two different solid colours.

[0024] The term “yarn” herein refers to elongated members which are capable of being woven together. Yarns can comprise a plurality of fibres or can be a single continuous piece of material. The fibres may be long continuous fibres, herein known as “filaments”. A filament can be used directly to make yarn or fabrics without the need for spinning. Yarns may be made using a monofilament (i.e., a single, filament strand which can bethick), or a multifilament, (i.e ., many fine filaments grouped together with or without twisting to form a yarn. Twisting typically allows for greater control of the thickness of the resultant yarn.

[0025] As such, each yarn may comprise, or be formed of, one or more filaments which are combined. One or more filaments can be combined via twisting or grouping to form a yarn. The thickness of the yarn can be varied by combining different numbers of filaments, the more filaments combined, the thicker the resulting yarn.

[0026] Suitably, a yarn is formed from, or comprises, up to 100 filaments, such as up to 50 filaments, preferably up to 45 filaments, more preferably up to 40 filaments, and / or a yarn is formed from, or comprises, at least 5 filaments, preferably at least 15 filaments, more preferably at least 25 filaments. The skilled person will understand that any of these upper and lower values may be combined to form combinations of upper and lower values.

[0027] A yarn having a given primary yarn colour can be formed of, or comprise, one or more filaments having the same colour, which is the same colour as the primary colour of the yarn.

[0028] A yarn having a given primary yarn colour can be formed of, or comprise, one or more filaments having colours which are different to the primary colour of the resultant yarn. Different combinations of coloured filaments can be combined to form each of the different primary-coloured yarns.

[0029] A yarn can be formed from at least two filaments having two colours which are different to the primary colour of the resultant yarn. A yarn can be formed from at least three filaments having two or three colours which are different to the primary colour of the resultant yarn. A yarn can be formed from at least four filaments having two, three or four colours which are different to the primary colour of the resultant yarn. A yarn can be formed from at least five filaments having two, three, four or five colours which are different to the primary colour of the resultant yarn.

[0030] Suitably, two or more yarns may be twisted or blended (e.g. twisted) together to form a thicker yarn, wherein the two or more yarns are of different colours and thus theperceived colour of the resultant thicker yarn is a different colour to the two or more yarns. The resultant thicker yarn is known as a “polychromatic yarn” herein. The polychromatic yarn may be used as part of the “plurality of yarns” provided in the methods described herein.

[0031] Suitably the two or more yarns used to make the polychromatic yarn is formed from, or comprises, up to 100 filaments, preferably up to 50 filaments, 20 filaments, preferably up to 15 filaments, more preferably up to 10 filaments, and / or at least 1 filament, preferably at least 3 filaments, more preferably at least 5 filaments. The skilled person will understand that any of these upper and lower values may be combined to form combinations of upper and lower values.

[0032] The method may comprise identifying a colour recipe for a desired fabric colour. The colour recipe may comprise a primary yarn colour warp composition. The colour recipe may comprise a primary yarn colour weft composition.

[0033] The method may comprise weaving yarns of the plurality of yarns to form the coloured fabric according to the identified colour recipe, e.g. by weaving a plurality of warp yarns and a plurality of weft yarns together. As used herein, the terms “a plurality of warp yarns” and “warp yarns” may be used interchangeably. Similarly, the terms “a plurality of weft yarns” and “weft yarns” may be used interchangeably. One or more of the yarns of the plurality of yarns may be the polychromatic yarn.

[0034] Weaving the plurality of yarns may comprise plain weave, twills and satins (such as plain weave).

[0035] Producing a coloured fabric by weaving yarns according to a colour recipe may enable a wide range of textile colours to be achievable using a given set of yarns. This may reduce manufacturing complexity and timescales, as a manufacturer can simply stock yarns with a given set of yarn colours (i.e., the primary yarn colours) which provide the ability to provide a wide array of fabric colours.

[0036] Producing a fabric with a desired resultant colour by combining individual primary yarns may reduce the waste, energy, and pollution associated with manufacture, as there is no need to produce a bespoke dye and dye the entire fabric. The skilled person willunderstand that the coloured fabric obtained using the method described herein may be of a colour which may appear to be different (e.g. under ambient conditions, during normal use of the colour fabric by the end user) from any of the primary colours of the yarn used to weave the fabric.

[0037] In some embodiments, the method comprises weaving warp yarns and weft yarns together, wherein:

[0038] the warp yarns comprise yarns of at least one colour, or at least two colours of the set of primary yarn colours (as defined herein), such as two or three colours (e.g. two colours). The skilled person will understand that each individual yarn has one colour individually selected from the set of primary colours as defined herein.

[0039] In some embodiments, the method comprises weaving warp yarns and weft yarns together, wherein:

[0040] the warp yarns comprise yarns of at least one colour, or at least two colours, each colour individually selected from the set of primary yarn colours (as defined herein), such as two or three colours (e.g. two colours) and

[0041] the weft yarns comprise yarns of at least one colour selected from the set of primary yarn colours (as defined herein), such as one, two, three, four or five colours. The colours of the warp yarn (e.g. the at least one colour, or at least two colours of the warp yarn) may be different from or the same as the colours of the weft yarn (e.g. the at least one colour of the weft yarn).

[0042] In particular such embodiments, the method comprises weaving warp yarns and weft yarns together, wherein:

[0043] the warp yarns comprise yarns of one colour, or two colours, each colour individually selected from the set of primary yarn colours (as defined herein); and

[0044] the weft yarns comprise yarns of one, two, three or four colours, each colour individually selected from the set of primary yarn colours (as defined herein).

[0045] In more particular such embodiments, the method comprises weaving warp yarns and weft yarns together, wherein:

[0046] the warp yarns comprise yarns of one colour or two colours, each colour individually selected from the set of primary yarn colours (as defined herein); andthe weft yarns comprise yarns of two, three or four colours (e.g. two), each colour selected from the set of primary yarn colours (as defined herein).

[0047] The warp yarns of at least one colour, or at least two colours selected from the set of primary yarn colours may comprise warp yarn of a first colour, warp yarn of a second colour, optionally warp yarn of a third colour, etc as appropriate.

[0048] In particular embodiments, the method comprises alternatively weaving (i.e. weaving in an alternative pattern) the warp yarns of the first colour and the warp yarns of the second colour.

[0049] The weft yarns of at least one colour of the set of primary yarn colours may comprise weft yarn of a first colour, optionally weft yarn of a second colour, optionally weft yarn of a third colour, optionally weft yarn of a fourth colour, optionally weft yarn of a fifth colour, etc as appropriate. The first, second, third, fourth and fifth colours etc are selected from the set of primary yarn colours as defined herein, and may be the same or different from the first, second, third etc colours of the warp yarn.

[0050] In particular embodiments, the method comprises alternatively weaving (i.e. weaving in an alternative pattern) the weft yarns of the first colour and the weft yarns of the second colour.

[0051] In more particular embodiments, the method comprises:

[0052] alternatively weaving (i.e. weaving in an alternative pattern) the warp yarns of the first colour and the warp yarns of the second colour; and

[0053] alternatively weaving (i.e. weaving in an alternative pattern) the weft yarns of the first colour and the weft yarns of the second colour.

[0054] As used herein, “alternatively weaving” or “alternative pattern” refer to weaving yarn in a sequence such that yarn of certain colours is intercalated and repeats predictably in the resulting woven fabric. For example, if red warp yarn and yellow warp yarn is woven in an alternative pattern, adjacent warp yarns in the resulting fabric may be in the following order: red, yellow, red, yellow, or in the following order: red, red, yellow, yellow, red, red, yellow, yellow, amongst other possible orders. The “alternatively weaving” or “alternative pattern” may also comprise yarn of a further colour whichintercalates the yarn of the specified colours. For example, if red warp yarn and yellow warp yarn is woven in an alternative pattern, yarn of additional colours may be present e.g. white yarn, so that adjacent warp yarns in the resulting fabric may be in the following order: red, white, yellow, red, white yellow, among other possible orders.

[0055] In particular embodiments, the method comprises weaving warp yarns and weft yarns, such that (e.g. in the coloured fabric):

[0056] every ten adjacent warp yarns comprise at least one warp yarn of a first colour and at least one warp yarn of a second colour; and / or (e.g. and)

[0057] every ten adjacent weft yarns comprise at least one weft yarn of a first colour and at least one weft yarn of a second colour, optionally at least one weft yarn of the third colour and optionally at least one weft yarn of a fourth colour.

[0058] In more particular such embodiments, the method comprises weaving warp yarns and weft yarns, such that (e.g. in the coloured fabric):

[0059] every four adjacent warp yarns comprise at least one warp yarn of a first colour and at least one warp yarn of a second colour; and / or (e.g. and)

[0060] every four adjacent weft yarns comprise at least one weft yarn of a first colour and at least one weft yarn of a second colour, optionally at least one weft yarn of the third colour and optionally at least one weft yarn of a fourth colour.

[0061] In more particular such embodiments, the method comprises weaving warp yarns and weft yarns, such that (e.g. in the coloured fabric):

[0062] every two adjacent warp yarns comprise one warp yarn of a first colour and one warp yarn of a second colour; and / or (e.g. and)

[0063] every four adjacent weft yarns comprise at least one weft yarn of a first colour and at least one weft yarn of a second colour, optionally at least one weft yarn of the third colour and optionally at least one weft yarn of a fourth colour.

[0064] In more particular such embodiments, the method comprises weaving warp yarns and weft yarns, such that (e.g. in the coloured fabric):

[0065] every three adjacent warp yarns comprise one warp yarn of a first colour, one warp yarn of a second colour and one warp yarn of a third colour; and / or (e.g. and)every four adjacent weft yarns comprise at least one weft yarn of a first colour and at least one weft yarn of a second colour, optionally at least one weft yarn of the third colour and optionally at least one weft yarn of a fourth colour.

[0066] The primary yarn colour warp composition may define the proportion of each primary yarn colour in the warp of the fabric. The primary yarn colour weft composition may define the proportion of each primary yarn colour in the weft of the fabric.

[0067] In some embodiments, the primary yarn colour warp composition comprises at least two colours of the set of primary yarn colours (as defined herein), such as at two or three colours of the set of primary colours. The skilled person will understand that in such embodiments, the yarn used in the warp direction is of at least two different primary colours.

[0068] In some embodiments, the primary yarn colour weft composition comprises at least one colour of the set of primary yarn colours (as defined herein), such as one, two, three, four or five colours of the set of primary colours.

[0069] In particular embodiments, the primary yarn colour warp composition comprises at least two colours (as defined herein), such as two colours, and the primary yarn colour weft composition comprises at least one primary colour (as defined herein), such as two, three or four colours.

[0070] The thickness of the yarn is important for obtaining a desired resultant colour by combining individual primary yarns. A suitably thin yarn is advantageous for aiding the optical mixing and thus achieving a resulting colour that is solid and not undesirably stripy.

[0071] In some embodiments, providing the plurality of yarns may comprise providing yarn and / or polychromatic yarn with a thickness of at least ID (i.e. 1 denier), such as at least 2D, 5D, 10D, 15D, 20D or 30D. In some embodiments, providing the plurality of yarns may comprise providing yarn with a thickness of at most 100D, such as at most 90D, 80D, 70D, 60 D or 50 D. The skilled person will understand that any of these upper and lower values may be combined to form combinations of upper and lower values. Measuring denier is a widely recognized standard method done by direct gravimetricanalysis - measuring the mass of a known length of filament or fiber, then calculating the denier using the formula: Denier=(mass (g)length (m))x9000. For example, ASTM D1577.

[0072] For example, in some embodiments, providing the plurality of yarns may comprise providing yarn with a thickness of from ID to 100D, such as from 5D to 80 D, for example from 5D to 60D.

[0073] The density of the coloured (woven) fabric may influence the optical mixing and thus how solid the resultant colour is. The density of woven fabric is typically expressed as EPI (Ends Per Inch) and PPI (Picks Per Inch), which relate to the density of warp and weft threads, respectively, in the woven fabric. Measuring EPI and PPI are widely recognized standard methods done by measuring the number of warp threads (ends) per linear inch of woven fabric. The most commonly used methods are Manual Counting with Pick Glass or Counting Glass or a Microscopic Method (ASTM D3775 or ISO 7211-2: 2024).

[0074] In some embodiments, the method comprises weaving warp yarns and weft yarns together to achieve a density of the coloured fabric of at least 100 EPI, such as at least 110 EPI, at least 120 EPI, at least 130 EPI, at least 140 EPI, or at least 150 EPI. In some embodiments, the method comprises weaving warp yarns and weft yarns together to achieve a density of the coloured fabric of at most 600 EPI, such as at most 450 EPI, at most 300 EPI, such as at most 290 EPI, at most 280 EPI, at most 270 EPI, at most 260 EPI, or at most 250 EPI. The skilled person will understand that any of these upper and lower values may be combined to form combinations of upper and lower values.

[0075] For example, in some embodiments, the method comprises weaving warp yarns and weft yarns together to achieve a density of the coloured fabric from 100 EPI to 300 EPI, such as from 150 EPI to 250 EPI.

[0076] In some embodiments, the method comprises weaving warp yarns and weft yarns together to achieve a density of the coloured fabric at least 90 PPI, such as at least 100 PPI or at least 110 PPI. In some embodiments, the method comprises weaving warp yarns and weft yarns together to achieve a density of the coloured of at most 150 PPI, such as at most 140 PPI, at most 130 PPI, or at most 120 PPI. The skilled person willunderstand that any of these upper and lower values may be combined to form combinations of upper and lower values.

[0077] For example, in some embodiments, the method comprises weaving warp yarns and weft yarns together to achieve a density of the coloured fabric from 100 PPI to 150 PPI, such as from 100 PPI to 130 PPI.

[0078] In some embodiments, the method does not substantially comprise jacquard weaving.

[0079] Providing the plurality of yarns may comprise adding colourant to a material comprising one or more polymers to dye the material and extruding the material to form the plurality of yarns. Providing the plurality of yarns may comprise dope-dyeing the yarns.

[0080] The plurality of yarns may comprise (e.g. consist of) dope-dyed yarns.

[0081] The skilled person will understand that “dope-dyed” is also known as “solution-dyed”. Therefore, “dope-dyed” and the like and “solution-dyed” and the like are used herein interchangeably.

[0082] Defining the warp and weft primary yarn colour compositions separately may provide the advantage of producing output colours which are closer to the desired output colours, as the structure of the fabric is considered. This may provide improved accuracy and precision when compared with a colour recipe which simply defines the overall proportion of each colour within the fabric.

[0083] Introducing colourant to the yarns during the extrusion process may reduce the waste and pollution associated with the dyeing process.

[0084] As used herein, the term “between” comprises the end values, i.e. “between 2-50” comprises 2 and 50.

[0085] The set of primary yarn colours may comprise between 2-50 colours. The set of primary yarn colours may comprise between 2-30 colours. The set of primary yarn colours may comprise between 2-15 colours. The set of primary yarn colours may comprise between 2-10 colours.The set of primary yarn colours may comprise between 5-50 colours. The set of primary yarn colours may comprise between 5-30 colours. The set of primary yarn colours may comprise between 5-15 colours. The set of primary yarn colours may comprise between 5-10 colours.

[0086] Identifying a colour recipe for the desired fabric colour may comprise choosing a colour recipe, such as retrieving a pre-calculated colour recipe from a memory. Identifying a colour recipe for the desired fabric colour may comprise retrieving a colour recipe from a memory which was determined experimentally.

[0087] Storing pre-calculated and / or experimentally determined colour recipes in a memory and retrieving the associated colour recipe when needed may reduce the time required to identify the colour recipe.

[0088] Identifying a colour recipe for the desired fabric colour may comprise calculating the colour recipe. Calculating the colour recipe may be a computer-implemented method.

[0089] Calculating the colour recipe may comprise providing a colour predictor. The colour predictor may be configured to receive a colour recipe as an input and provide a predicted fabric colour as an output.

[0090] The colour recipe may be calculated using an optimisation algorithm to minimise the colour difference between a desired fabric colour and a predicted fabric colour by adjusting the colour recipe provided as an input to the colour predictor. The optimisation algorithm may identify a colour recipe which minimises the colour difference between a desired fabric colour and a predicted fabric colour. The optimisation algorithm may identify a colour recipe for which the colour distance between a desired fabric colour and a predicted fabric colour is below a predetermined threshold. Alternatively, optimisation may be achieved by optimising the minimum error between obtained versus predicted spectrum and calculating the colour difference.

[0091] The colour difference may be defined using a function in CIELAB coordinates. The colour difference may be defined using the CIEDE2000 or CMC (e.g. CIEDE2000) colour distance formula.The optimisation algorithm may be a gradient descent algorithm. The optimisation algorithm may be a brute-force algorithm. The optimisation algorithm may be a genetic algorithm.

[0092] The optimisation algorithm may comprise comparing the predicted fabric colour for each possible colour recipe with the desired fabric colour.

[0093] Using a colour predictor and optimisation algorithm enables colour recipes to be determined for a range of desired colours, rather than requiring every possible colour recipe to be stored in a memory.

[0094] The colour predictor may be a trained Al model. The colour predictor may be a neural network. The neural network may comprise primary yarn colour compositions as inputs. The primary yarn colour weft compositions and primary yarn colour warp compositions may be represented as separate nodes in the input layer of the neural network. The neural network may be a multi-layer perceptron.

[0095] The values of the input nodes may be constrained based on one or more requirements. The values of the input nodes may be constrained based on the total number of yarns to be used to produce the fabric. The values of the input nodes may be constrained based on the number of warp yarns to be used to produce the fabric. The values of the input nodes may be constrained based on the number of warp yarns to be used to produce the fabric.

[0096] Constraining the input nodes, and therefore the colour recipe, may ensure that the identified colour recipe is suitable for the manufacturing requirements.

[0097] The nodes of the output layer of the neural network may represent the spectral reflectance profile of the predicted fabric colour. The spectral reflectance profile may define the spectral reflectance values of the predicted fabric colour for a plurality of wavelengths.

[0098] The method may comprise generating weaving instructions based on the identified colour recipe. Generating weaving instructions based on the identified colour recipemay comprise converting the proportions of the primary yarn colours into the number of each yarn type within a weaving pattern. Generating the weaving pattern may comprise reorganising the order of yarns within the weaving pattern to avoid and / or reduce the number of consecutive yarns of the same type.

[0099] The method may comprise weaving yarns according to the weaving instructions to form the fabric.

[0100] The method may further comprise identifying a colour recipe for the polychromatic yarn (i.e. for a perceived colour of the polychromatic yarn). The identifying a colour recipe for the polychromatic yarn may have any of the features describe above for the identifying a colour recipe for the coloured fabric (including combinations thereof).

[0101] For example, identifying a colour recipe for the polychromatic yarn may comprise calculating the colour recipe. Calculating the colour recipe may comprise providing a colour predictor. The colour predictor may be configured to receive a colour recipe as an input and provide a predicted colour of the polychromatic yarn as an output.

[0102] The colour predictor may be a trained Al model. The colour predictor may be a neural network. The neural network may comprise primary yarn colour compositions and / or primary colour filaments as inputs. The neural network may be a multi-layer perceptron.

[0103] The resulting polychromatic yarn may be an input for the neural network described above in relation to the colour predictor described in relation to identifying the colour recipe for the coloured fabric.

[0104] According to a second aspect of the invention, there is provided a coloured fabric (i.e. coloured woven fabric) obtainable or obtained by the method of the first aspect of the invention (including all embodiments thereof).

[0105] For the avoidance of doubt, the coloured fabric may have any of the features of the first aspect of the invention including combinations thereof.

[0106] Therefore, in some embodiments, the coloured fabric comprises:warp yarns of two colours, each colour individually selected from the set of primary yarn colours (as defined herein); and

[0107] weft yarns of at least two colours, each selected from the set of primary yarn colours (as defined herein), such as at least three colours. The two colours of the warp yarn may be different or the same the at least two colours of the weft yarn.

[0108] In particular such embodiments, the coloured fabric comprises:

[0109] (i) yarn with a thickness of yarn with a thickness of from ID to 100D, such as from 5D to 80 D, for example from 5D to 60D; and / or

[0110] (ii) warp yarns alternatively comprising warp yarns of a first colour and warp yarns of a second colour; and / or

[0111] (iii) weft yarns alternatively comprising weft yarns of a first colour and weft yarns of a second colour; and / or

[0112] (iv) yarn formed of, or comprising, one or more filaments having colours which are different to the primary colour of the resultant yarn, optionally wherein the yarn is formed via twisting and / or interweaving the one or more filaments); and / or

[0113] (v) the coloured fabric has a colour (e.g. a perceived colour) which is different from any of the primary colours of the yarn used to weave the coloured fabric.

[0114] According to a third aspect of the invention, there is provided a coloured fabric comprising:

[0115] warp yarn of at least two colours, each colour individually selected from the set of primary yarn colours (as defined in the first aspect of the invention); and

[0116] weft yarns of at least two colours, each selected from the set of primary yarn colours (as defined in the first aspect of the invention), such as at least three colours. The at least two colours of the warp yarn may be different from or the same as the at least two colours of the weft yarn.

[0117] The coloured fabric of the third aspect may have any of the features of the first and second aspects of the invention including combinations thereof. There is also provided a method of making the coloured fabric of the third aspect comprising the step of weaving together the warp and weft yarns as described herein.

[0118] For example, in particular embodiments, the coloured fabric comprises:(i) warp and weft yarn with a thickness of from ID to 100D, such as from 5D to 80 D, for example from 5D to 60D; and / or

[0119] (ii) warp yarns alternatively comprising warp yarns of a first colour and warp yarns of a second colour; and / or

[0120] (iii) weft yarns alternatively comprising weft yarns of a first colour and weft yarns of a second colour; and / or

[0121] (iv) yarn formed of, or comprising, one or more filaments having colours which are different to the primary colour of the resultant yarn, optionally wherein the yarn is formed via twisting and / or interweaving the one or more filaments); and / or

[0122] (v) the coloured fabric has a colour (e.g. a perceived colour) which is different from any of the primary colours of the yarn used to weave the coloured fabric.

[0123] According to a fourth aspect of the invention, there is provided a method of calculating a colour recipe for producing a coloured fabric. The method may be a computer-implemented method. The method may be used as part of the method of the first aspect, and any features / steps described in the first aspect of the invention are equally applicable to the fourth, and vice versa.

[0124] The method may comprise:

[0125] using an optimisation algorithm to minimise the colour difference between a desired fabric colour and a predicted fabric colour;

[0126] wherein the predicted fabric colour is determined using a colour predictor, and wherein the colour predictor comprises primary yarn colour compositions as inputs; and wherein, for each primary yarn colour, the weft composition and warp compositions are represented as separate inputs to the colour predictor.

[0127] The colour predictor may be a neural network. The primary yarn colour warp compositions and primary yarn colour weft compositions may be represented via separate nodes in the input layer of the neural network.

[0128] The optimisation algorithm may be a gradient descent algorithm. The optimisation algorithm may be a brute-force algorithm. The optimisation algorithm may be a genetic algorithm.

[0129] The neural network may be a multilayer perceptron.The nodes of output layer may represent the spectral reflectance profile of the predicted output colour.

[0130] According to a fifth aspect of the invention, there is provided a system for providing a colour recipe.

[0131] The system may comprise an interface. The interface may be configured to enable a user to input a desired fabric colour. The interface may be configured to enable a user to input one or more constraints.

[0132] The system may comprise a processor. The processor may be configured to receive the desired fabric colour and identify a corresponding colour recipe. The processor may be part of a user device. The processor may be part of an external server to which a user device is connected.

[0133] The system may comprise an output. The output may be configured to provide the colour recipe to the user.

[0134] The interface and output may be part of a user device.

[0135] The processor may be configured to identify the colour recipe corresponding to the desired fabric colour by calculating the colour recipe using the method of the fourth aspect of the invention.

[0136] The processor may be configured to identify the colour recipe corresponding to the desired fabric colour by retrieving the colour recipe from a memory, wherein the colour recipe was pre-calculated using the method of the fourth aspect of the invention, or wherein the colour recipe was determined experimentally.

[0137] According to a sixth aspect of the invention there is provided a colour predictor. The colour predictor may be configured to receive a colour recipe as an input. The colour predictor may be configured to provide a predicted fabric colour as an output. The colour predictor can comprise any of the features described in relation to the first, fourth, or fifth aspects of the invention.The colour predictor may be an Al model. The colour predictor may be a neural network. The colour predictor may be a multilayer perceptron.

[0138] The colour predictor may comprise an input layer, wherein each node represents the relative proportion of each primary yarn colour in the colour recipe. The input layer may comprise nodes representing the relative proportion of each primary yarn colour in the warp of the fabric. The input layer may comprise nodes representing the relative proportion of each primary yarn colour in the weft of the fabric.

[0139] Optional features of any of the above aspects may be combined with the features of any other aspect, in any combination. For example, features described in connection with the locator of the first aspect may have corresponding features definable with respect to the system of the fifth aspect, and vice versa, and these embodiments are specifically envisaged. Features which are described in the context or separate aspects and embodiments of the invention may be used together and / or be interchangeable wherever possible. Similarly, where features are, for brevity, described in the context of a single embodiment, those features may also be provided separately or in any suitable subcombination.

[0140] Brief description of the drawings

[0141] The invention will now be described, by way of example only, with reference to the accompanying drawings, in which:

[0142] Figures 1(a) and (b) show schematics of woven fabrics;

[0143] Figure 2 shows a flowchart representing a method for producing a coloured woven fabric according to the present invention;

[0144] Figure 3 shows a flowchart representing a method for calculating a colour recipe according to the present invention;

[0145] Figure 4 shows a method for generating a weaving pattern according to the present invention;Figure 5 shows a schematic of a colour predictor in accordance with the present invention; and

[0146] Figure 6 shows a flowchart of a method for training a colour predictor according to the present invention.

[0147] Figure 7 shows a schematic of a system 400 for providing a colour recipe.

[0148] Figure 8 shows a fabric (11) made of 3 colours in weft (4), 1 colour in warp (2). Figure 8b shows a fabric (11’) made of 3 colours in weft (4) but 2 colours in warp (2’).

[0149] Figure 9 shows a polychromatic yarn (530) made by twisting yarns (510) and (520) having different colours. Coloured fabric (540) is made of the polychromatic yarn (530) in weft and the polychromatic yarn (530) and other yarn (510) in warp.

[0150] Detailed description

[0151] Figure 1(a) shows a schematic of a woven fabric 10. The woven fabric 10 is formed of warp yarns 2 and weft yarns 4. The warp yarns 2 are shown vertically and the weft yarns 4 are shown horizontally.

[0152] During manufacture of woven fabrics, such as the woven fabric 10 of Figure 1(a), warp yarns 2 are fixed in place and weft yarns 4 are woven between the warp yarns 2 to create the fabric 10. The weaving process can be performed entirely by hand but, in many cases, the warp yarns 2 are held in place using a loom whilst the weft yarns 4 are woven between the warp yarns 2.

[0153] Although the weft yarns 4 in Figure 1(a) are woven between consecutive warp yarns 2, in other examples, the weft yarns 4 can be woven across multiple warp yarns 2 at a time. For example, Figure 1(b) shows a schematic of a woven fabric 10’ wherein each weft yarn 4 passes in front of two or more warp yarns 2 and then passes behind two of more warp yarns 2. In other examples, the number of warp yarns 2 crossed by the weft yarns 4 can be irregular throughout the fabric 10.Using differently coloured yarns to create a woven fabric can produce a fabric with a resultant colour (i.e., a colour observed by a human observer) which is different from any of the constituent yarn colours. In a simple illustrative example, combining red and yellow yarns can create a fabric which appears orange, despite there being no yarns with an orange colour.

[0154] A “colour recipe” refers to a specific combination of coloured yarns which produces a resultant output colour. For a given output colour, a colour recipe can indicate the required yarn colours required to produce that output colour.

[0155] The colour recipe can indicate the yarn colours, as well as the relative proportions of each of the yarn colours, required to produce a given output colour. In another simple illustrative example, a colour recipe can indicate that, in order to produce a dark orange output colour, red and yellow yarns can be combined with a ratio of, for example, 2: 1. Although the colour recipe in the previous sentence is expressed as a ratio, the skilled person will recognise that a colour recipe can be represented in any suitable manner, for example, as yarn colour proportions (i.e., red=0.66, yellow=0.33).

[0156] For a woven fabric, rather than only indicating the required yarn colours and their relative compositions, a colour recipe can also indicate the “yarn colour warp composition” and the “yarn colour weft composition”. The “yarn colour warp composition” and the “yarn colour weft composition” respectively refer to the individual yarn colours and their relative proportions in each of the warp and weft of the fabric required to produce a given fabric colour.

[0157] In some examples, a colour recipe can include other parameters in addition to the proportion of yarn colours.

[0158] For example, a colour recipe can comprise details about the yarns to be used, such as the yarn material, yarn diameter, yarn weight, and yarn finish, as these details can influence the resultant colour provided by the colour recipe. In some examples, a colour recipe can comprise details about the weaving process, such as the weaving tension (e.g., ends per inch, pitch per inch) and weaving pattern (plain weave, twill weave, satinweave, etc), as these details can influence the resultant colour provided by the colour recipe.

[0159] Figure 2 shows a flowchart 100 representing a method for producing a coloured woven fabric.

[0160] The method 100 comprises: providing 110 a plurality of yarns; identifying 120 a colour recipe for a desired fabric colour; and weaving 130 yarns to form the fabric according to the identified colour recipe.

[0161] Providing 110 the plurality of yarns can comprise obtaining a pre-manufactured plurality of yarns. Alternatively, providing 110 the plurality of yarns can comprise forming the plurality of yarns. Forming the plurality of yarns can comprise extruding one or more materials to form the plurality of yarns. Each of the plurality of yarns can be “dope-dyed”, meaning colourants are introduced during the extrusion process to produce coloured yarns.

[0162] Each of the plurality of yarns is coloured. Each of the plurality of yarns has a colour which is one of a set of colours referred to herein as the “primary yarn colours”. In some examples, there are 50 primary yarn colours. In other examples, there can be any number of primary yarn colours, for example, there can be 40, 30, 20, 10, 8, 5, or 4 primary yarn colours.

[0163] Each of the plurality of yarns can be formed of, or comprise, one or more polymers. The polymers can be synthetic polymers or natural polymers. The plurality of yarns can be formed of, or comprise, any material suitable for undergoing extrusion and dope-dyeing. In some examples, the plurality of yarns are formed of, or comprise, materials which can undergo dissolution or melting such that they can be spun into a yarn and colourants can be added during the spinning process. In some examples, the yarns can comprise additives, such as UV absorbers, for example to improve performance and / or longevity.

[0164] In some examples, the yarns are formed of, or comprise, one or more of: polyamide, polypropylene, polyethylene terephthalate, polyethylene, viscose, lyocell, modal, cellulose carbamate, brewed protein, or a plant-based protein. This list is non-exhaustive but proves examples of materials which can be used to produce the yarns.In some examples, each of the yarns are formed of, or comprise one or more filaments which are combined together. The one or more filaments can be combined via twisting and / or interweaving, to form a single larger and / or thicker yarn.

[0165] A yarn having a given primary yarn colour can be formed of, or comprise, one or more filaments having colours which are different to the primary colour of the resultant yarn. Different combinations of coloured filaments can be combined to form each of the different primary-coloured yarns.

[0166] In some examples, each of the filaments are coloured via dope-dyeing. In some examples, filaments are extruded having a plurality of different colours, and various combination of these filaments can be used to form the yarns having the primary yarn colours.

[0167] Identifying 120 a colour recipe for a desired fabric colour can comprise identifying a colour recipe defining a primary yarn colour warp composition and a primary yarn colour weft composition.

[0168] Identifying 120 the colour recipe for the desired fabric colour can comprise calculating the colour recipe. Calculating a colour recipe can be a computer-implemented method. Alternatively, identifying 120 the colour recipe can comprise retrieving a pre-calculated colour recipe, for example, from computer memory. Methods for calculating a colour recipe for a given desired fabric colour are discussed in further detail later on in this specification, with reference to Figure 3.

[0169] After identification 120 of the colour recipe, a user, or machine, can create a woven fabric by weaving 130 yarns of the plurality of yarns, according to the identified colour recipe.

[0170] A manufacturer may be able to weave yarns according to the colour recipe directly. However, in other examples, the colour recipe is converted into a weaving pattern which is easier for a manufacturer to understand. A weaving pattern defines the repeating pattern of yarns in the warp and weft of the fabric. Example weaving patterns are shown and discussed in greater detail later in the description.For clarity, and as would be recognised by the skilled person, the order of steps 110 and 120 can be reversed, such that only the yarns required by the colour recipe are provided.

[0171] Figure 3 shows a flowchart 120 representing a method for calculating a colour recipe.

[0172] The method 120 comprises: identifying 121 a desired fabric colour; providing 122 a colour predictor; and identifying 123 the colour recipe which, when provided as an input to the colour predictor, produces a minimal colour difference between the output colour of the colour predictor and the desired fabric colour.

[0173] Identifying 121 a desired fabric colour can comprise identifying a spectral reflectance profile for the desired fabric colour. For example, once a user has decided upon a colour for the fabric (i.e., the desired fabric colour), the spectral reflectance profile for that colour can be retrieved from a memory or an external server (i.e., via the internet), can be determined experimentally, or can be calculated. The spectral reflectance profile of a colour can be defined via the spectral reflectance values of a colour for a plurality of different wavelengths. In other examples, identifying the desired fabric colour can comprise identifying the coordinates for the desired fabric colour in a given colour space.

[0174] A colour predictor can be used to calculate a colour recipe for a desired fabric colour. A colour predictor is a function / model which takes a colour recipe as an input and produces a predicted fabric colour as an output. Using one of the earlier examples, providing a colour recipe of red and yellow yarns with a ratio of 2: 1 as an input to a colour predictor would result in the colour predictor providing dark orange as an output.

[0175] Example colour predictors are discussed later in the description with reference to Figure 5.

[0176] The difference / similarity between a predicted fabric colour of the colour predictor and a desired fabric colour can be measured as a colour difference. The colour difference can be defined as a metric which measures the distance between two colours in a colour space. The colour difference can be defined using a simple RGB colour space, or can be defined using any other colour space, such as a CIELAB or CIELUV colour space.The colour difference can be determined using any suitable metric to measure the distance between two colours in colour space. For example, the colour difference can be calculated using the Euclidean distance between two colours in RGB space. In other examples, the CMC, CIE 1976, CIE94, or CIEDE2000 colour distance metrics can be used in a CIELAB colour space.

[0177] In order to measure the colour difference, the desired fabric colour needs to be defined in the chosen colour space (i.e . , the colour space for which the colour difference metric is defined). For example, once a desired fabric colour is identified and defined in terms of spectral reflectance values, it’s CIELAB coordinates can be determined such that the CIEDE2000 colour distance formula can be used. In a particular example, if the colour distance is to be calculated using a metric defined in CIELAB coordinates, spectral reflectance values can be converted into CIEXYZ coordinates and then into CIELAB coordinates, in a process which would be well known to the person skilled in the art.

[0178] In order to measure the colour difference, the predicted fabric colour from the colour predictor should be defined using the appropriate colour space coordinates. For example, the colour predictor can be configured to produce a predicted fabric colour as an output directly in the appropriate colour space coordinates, or the output from the colour predictor can be converted into the appropriate colour space coordinates. Again, the “appropriate colour space coordinates” refer to coordinates in the colour space for which the colour distance is defined.

[0179] The colour recipe for a desired fabric colour can be calculated using an optimisation algorithm to minimise the colour difference between a desired fabric colour and a predicted fabric colour. To calculate the colour recipe for the desired fabric colour, the colour recipe which is input to the colour predictor is adjusted to minimise the colour difference. The colour recipe which minimises the colour difference can be defined as the colour recipe for the desired fabric colour.

[0180] Any suitable optimisation algorithm can be used to minimise the colour difference between the desired fabric colour and the predicted fabric colour.

[0181] In a simple example, the optimisation algorithm can be a so called “brute force” algorithm in which every possible colour recipe is tested and the recipe providing thesmallest colour difference is taken to be the colour recipe for the desired fabric colour. In a similar example, a mixture of tested colours and predicted colours generated by the colour predictor are combined in a look-up table, and the recipe providing the smallest colour difference is taken to be the colour recipe for the desired fabric colour. Alternatively, colour recipes can be provided as inputs until a colour recipe is found which produces a predicted output colour with a colour difference from the desired fabric colour which is below a predetermined threshold.

[0182] In other examples, the optimisation algorithm can be a gradient descent algorithm. In other examples, the optimisation algorithm can be a genetic algorithm. For example, a genetic algorithm can be employed to determine the colour recipe by iteratively evolving candidate solutions through selection, crossover, and mutation to minimise the colour difference.

[0183] In other examples, the optimisation algorithm is, or comprises, one or more of: a generative adversarial network, a naive Bayesian optimisation, a Markov chain Monte Carlo algorithm, simulated annealing, reinforcement learning, a LSTM network, or a transformer.

[0184] After an optimisation algorithm has been used to determine the input to the colour predictor which provides a predicted output colour which is equal to, or close to, the desired fabric colour (i.e., by minimising the colour difference), weaving instructions for the desired fabric colour can be obtained based on the colour recipe.

[0185] In some examples, the colour recipe input to the colour predictor is simply provided to a manufacturer. However, in other examples, the colour recipe is converted into weaving instructions which can be more easily read / understood by a manufacturer.

[0186] Figure 4 shows an example method 125 for generating weaving instructions 128 for a desired fabric colour. The method 125 converts the colour recipe 126, which was input to the colour predictor to provide the predicted output colour with a minimum colour difference relative to the desired fabric colour, into a weaving instructions 128 that are easier for a manufacturer to comprehend.The colour recipe (i.e., the colour predictor input) is shown as vector 126 comprising twelve entries. The first six entries in the vector 126 represent the proportions of each of the primary yarn colours in the warp of the fabric. The first six entries in the vector 126 are the primary yarn colour warp composition. The final six entries in the vector 126 represent the proportions of each of the primary yarn colours in the weft of the fabric. The final six entries in the vector 126 are the primary yarn colour weft composition. In the present example, there are six primary yarn colours: White, Black, Red, Blue, Green, Yellow. In other examples, any alternative number of primary yarn colours can be used, meaning the vector representing the colour recipe would have a different number of entries.

[0187] In the present example, the colour recipe 126 is subject to various constraints which are discussed below. In other examples, alternative and / or different constraints can be placed on the colour recipe. The constraints are determined by the manufacturer.

[0188] A first constraint is that the values of the entries within the colour recipe vector 126 sum to 1. This is necessary, as the values of the entries represent the proportion of the associated primary yarn colours within the warp / weft of the fabric.

[0189] A second constraint is that four yarns are present in the warp of the weaving pattern and four yarns are present in the weft of the weaving pattern. Given the restriction that the weaving pattern comprises eight yarns (four in the warp, four in the weft), the proportion of each primary yarn colour must be a multiple of 0.125 (i.e., - ) in the colour 8

[0190] recipe.

[0191] A third constraint is that there are an equal number of yarns in the warp and weft of the fabric, the sum of the first six entries must equal 0.5, and the sum of the second six entries must equal 0.5.

[0192] In some examples, the colour recipe is restricted such that only a single yarn colour is used in the warp of the fabric. In such cases, one warp yarn colour proportion will be equal to 0.5, and the remaining warp yarn colour proportions will be equal to 0.

[0193] In Figure 4, the primary yarn colour warp composition comprises a value of 0.5 for black yarn, and values of 0 for the remaining five colours. This means that, for colourrecipe 126, only black yarns are present in the warp composition. The primary yarn colour weft composition comprises values of 0.25 for blue yarns and 0.25 for red yarns, and values of 0 for the remaining four colours. This means that, for colour recipe 126, only blue and red yarns are present in the weft composition, and they are present in equal parts.

[0194] In some examples, the colour recipe provided to the colour predictor is constrained as set out in the paragraph above. In other examples, the colour recipe provided to the colour predictor is not constrained, or is partially constrained (e.g., only one of the constraints apply), and the relative proportions of each primary yarn colour in the warp and weft are adjusted when generating the weaving instructions for the manufacturer to comply with all the constraints.

[0195] For example, in Figure 4, if the entry in the colour recipe vector 126 corresponding to the proportion of black within the warp was equal to 0.543, this would not satisfy that constraints because (i) the total proportion of warp yarns exceeds 0.5; and (ii) 0.543 is not a multiple of 0.125, so it cannot be represented using eight yarns. To generate the weaving instructions, 0.543 would be rounded to 0.5.

[0196] From the colour recipe 126 (after the relative proportions of each primary yarn colour are adjusted to meet constraints, if necessary), preliminary weaving instructions 127 are generated. The preliminary weaving instructions 127 simply shows the number of each yarn colour within the warp and weft of the fabric.

[0197] In the example of Figure 4, the proportion of black warp yarns in the fabric equals 0.5. Given there are eight yarns in total, this means that there are four (8 multiplied by 0.5) black warp yarns. Similarly, there are two red weft yarns (8 multiplied by 0.25) and two blue weft yarns (8 multiplied by 0.25). The preliminary weaving instructions 127 shows that of the eight yarns in the fabric, there should be four black warp yarns, two red weft yarns, and two blue weft yarns. In some examples, the preliminary weaving instructions can be sent to the manufacturer and may form the final weaving instructions.

[0198] In some examples, the preliminary weaving instructions 127 can be reorganised to form final weaving instructions 128 to be used by a manufacturer. The preliminary weaving instructions 127 which are generated by simply converting the relative proportions fromthe colour recipe into yarn numbers, contains repeated yarns in the weaving pattern 127. For example, there are two red yarns side-by-side, and two blue yarns side-by-side. This can create stripes, rather than producing a resultant colour as desired. Therefore, to generate the final weaving instructions 128, the order of the yarns in the weft is reorganised to ensure yarns of the same colour are not positioned side-by-side. In other examples wherein it is not possible to avoid repeated yarn colours in the weaving pattern, the number of repeated yarns can be minimised.

[0199] Figure 5 shows a schematic of a colour predictor 200 according to one example. The colour predictor 200 is a neural network and, more specifically, is a multilayer perceptron.

[0200] In other examples, the colour predictor can be a different model, rather than a neural network. For example, the colour predictor can be, or comprise, one or more of: a convolutional neural network, an autoencoders, a recurrent neural network, a transformer, a graph neural network regression model, or an ensemble learning algorithm (e.g., decision trees, gradient boosting machines, or support vector machines.)

[0201] The colour predictor 200 comprises an input layer 210, a hidden layer 220, and an output layer 230. The input layer 210 comprises a plurality of input nodes {ai, a2, ... an}. The hidden layer 220 comprises a plurality of hidden nodes {bi, bz, ... bn}. The output layer 230 comprises a plurality of output nodes {ci, C2, ... cn}. The output layer 230 of the colour predictor 200 represents the predicted colour of a woven fabric which is woven using the colour recipe defined by the input layer 210.

[0202] As is known for neural networks, the nodes in each layer are functions of the values of the nodes in the previous layer. In a specific example, the value of each node (other than the input nodes) in the neural network 200 is given by the tansig function of a weighted sum of the values of the nodes in the previous layer.

[0203] The nodes shown in Figure 5 are for illustrative purposes only and the number of nodes displayed in each layer is not to be construed as limiting for the invention. In other examples, the colour predictor 200 can comprise additional hidden layers, or no hidden layers.The input layer 210 of the colour predictor 200 comprises nodes representing yarns with different primary yarn colours. For example, if there are three primary yarn colours, the input layer 210 can comprise nodes for yarns with each of the three colours. The input nodes representing the yarns can have values between 0 and 1, representing the relative proportion of the primary yarn colour associated with that yarn within the fabric.

[0204] In some examples, the input layer 210 of the colour predictor 200 comprises input nodes representing the proportion of each primary yarn colour within the warp of the fabric and input nodes representing the proportion of each primary yarn colour within the weft of the fabric.

[0205] For example, if there are four primary yarn colours - Blue, Red, Yellow, and Green -the input layer 210 can comprise 8 input nodes representing each of the four primary yarn colours in each of the warp and weft.

[0206] In other examples, the input layer 210 can comprise additional inputs / nodes. For example, the input layer 210 can comprise nodes representing: yarn thickness, yarn density, or any other properties.

[0207] The colour recipe 126 shown in Figure 4 is an example of values which can be provided to the input layer 210 of the neural network 200.

[0208] As discussed above, the possible values of the input nodes can be restricted depending on the manufacturing requirements. For examples, a first restriction is that the sum of the node values in the input layer 210 is equal to 1. If the woven fabric is to be manufactured with an equal number of warp and weft yarns, the sum of the node values representing the weft yarns must equal 0.5, and the sum of the node values representing the warp yarns must equal 0.5. Depending on the number of yarns in each of the weft and warp patterns, the allowable values for the nodes in the input layer are restricted. For example, if there are five warp yarns and five weft yarns in the weaving pattern, the input nodes are restricted to having values which are multiples of 0.2. As another example, if there are four warp yarns and four weft yarns in the weaving pattern, the input nodes are restricted to having values which are multiples of 0.25. The skilled person will recognise that other systems, such as percentages, can be used instead of decimals to represent the proportions, and the values can be scaled as desired.The output layer 230 comprises nodes, wherein each node represents the reflectance value of the resultant woven fabric at a given wavelength. Defining the reflectance of a fabric at various wavelengths defines the colour of the fabric. Each node in the output layer 230 represents the reflectance via a value between 0 and 1, wherein 0 indicates no reflectance and 1 represents complete reflectance. As is known to the skilled person, reflectance is a measure of reflected light as a proportion of incident light.

[0209] The output layer 230 comprises nodes to represent reflectance across the visible light range of the electromagnetic spectrum. The visible light range can be taken to be between 380-700nm, 400-700nm, or any suitable range. The number of nodes in the output layer 230 can vary depending on how precise the colour predictor is intended to be. The more nodes used in the output layer 230, the more precise the colour predictor is, as the reflectance is defined for a greater variety of wavelengths. In some examples, each node in the output layer 230 represents a wavelength with increments of lOnm.

[0210] As discussed above, in order for a colour difference to be calculated, the output colour represented by the nodes of the output layer 230 can be converted into coordinates in an appropriate colour space.

[0211] In other examples, the output layer 230 can directly provide coordinates in the relevant colour space. For example, the nodes in the output layer 230 can represent red, green and blue values within an RGB colour space, or L*, a*, and b* values within a CIELAB colour space.

[0212] For examples wherein the colour predictor 200 is a neural network, the neural network can be trained using conventional methods, such as gradient descent and backpropagation.

[0213] Figure 6 shows a flowchart 300 of a method for training a colour predictor.

[0214] The method 300 comprises: preparing 310 training fabrics; measuring 320 the colour of the training fabrics; and training 330 the colour predictor using the training fabrics and the measured colour of the training fabrics.For example, to train the neural network, a plurality of fabrics are manufactured as training fabrics. A variety of different colour recipes are used to produce training fabrics, and the resultant colours of those training fabrics are measured. In some examples, the resultant colour of the training fabric can be measured by measuring the spectral reflectance of the fabric at a variety of wavelengths to obtain a spectral reflectance profile. If the output layer 230 of the neural network being trained defines the output colour in terms of colour space coordinates, the colour space coordinates for the resultant colour for each training fabric is determined for the training process.

[0215] To perform the training step 330, in one example, a loss function is defined (for example, RMS error) between the output of the neural network for a colour recipe and the actual resultant colour for a training fabric produced using said colour recipe. Gradient descent and backpropagation is used to minimize the loss function, thereby training the colour predictor. The skilled person will recognise that alternative supervised learning methods can be implemented to train the colour predictor.

[0216] Figure 7 shows a schematic of a system 400 for providing a colour recipe. The system 400 comprises an interface 410, a processor 420, and an output 430.

[0217] The interface 410 is configured to enable a user to input a desired fabric colour. For example, the interface can comprise one or more of a keyboard, a touchscreen, a dial, or any other suitable means for enabling user interaction. The interface 410 can be configured to enable a user to input a desired fabric colour defined in terms of coordinates in a colour space, in spectral reflectance values, or any other manner for defining a colour. In some examples, the interface 410 is configured to scan an object and take the colour of that object to be the desired fabric colour. In some examples, the interface 410 is configured to receive an image from a user and take the colour of that image to be the desired fabric colour.

[0218] In some examples, the interface 410 is configured to receive one or more manufacturing constraints. For example, a user can provide (i) a number of yarns to be used in total to produce the coloured fabric; (ii) a number of warp yarns to be used to produce the coloured fabric; and / or (iii) a number of weft yarns to be used to produce the fabric.The processor 420 is configured to receive the desired fabric colour (which is input via the interface 410) and identify a corresponding colour recipe.

[0219] In some examples, the processor 420 is configured to identify the colour recipe corresponding to the desired fabric colour by calculating the colour recipe using the methods described throughout this description (by minimising a colour difference between the desired fabric colour and a predicted output colour from a colour predictor).

[0220] Additionally, or alternatively, the processor 420 can be configured to retrieve the colour recipe from a memory, wherein the colour recipe was pre-calculated using the methods described herein, and / or wherein the colour recipe was determined experimentally.

[0221] The output 430 is configured to provide the colour recipe to the user. The output 430 can be, or comprise, any means suitable for conveying information to a user. For example, the output 430 can comprise a screen configured to display the colour recipe. In some examples, the output 430 can comprise a communications system configured to transmit the colour recipe directly to a manufacturing machine. In some examples, the output 430 can comprise a printer configured to print the colour recipe for the user.

[0222] In some examples, the interface 410 and the output 430 can be parts of a user device, such as a PC or smartphone. In such examples, the processor 420 could be the processor of the device (such that calculations are performed on the device itself) or the processor 420 can be an external server which the device connects to (e.g., via the internet).

[0223] Figure 8 shows a fabric (11) made of 3 colours in weft (4), 1 colour in warp (2). Figure 8b shows a fabric (I T) made of 3 colours in weft (4) but 2 colours in warp (2’).

[0224] Figure 9 shows a polychromatic yarn (530) made by twisting yarns (510) and (520) having different colours. Coloured fabric (540) is made of the polychromatic yarn (530) in weft and the polychromatic yarn (530) and other yarn (510) in warp.

[0225] From reading the present disclosure, other variations and modifications will be apparent to the skilled person. Such variations and modifications may involve equivalent and other features which are already known in the art of predictivemodelling and woven textile manufacturing, and which may be used instead of, or in addition to, features already described herein.

[0226] Although the appended claims are directed to particular combinations of features, it should be understood that the scope of the disclosure of the present invention also includes any novel feature or any novel combination of features disclosed herein either explicitly or implicitly or any generalisation thereof, whether or not it relates to the same invention as presently claimed in any claim and whether or not it mitigates any or all of the same technical problems as does the present invention.

[0227] Features which are described in the context of separate examples may also be provided in combination in a single example. Conversely, various features which are, for brevity, described in the context of a single example, may also be provided separately or in any suitable sub-combination. The applicant hereby gives notice that new claims may be formulated to such features and / or combinations of such features during the prosecution of the present application or of any further application derived therefrom.

[0228] For the sake of completeness, it is also stated that the term "comprising" does not exclude other elements or steps, the term “a” or “an” does not exclude a plurality, and a single processor or other unit may fulfil the functions of several means recited in the claims and any reference signs in the claims shall not be construed as limiting the scope of the claims.

[0229] Examples

[0230] Yarn having the properties described in Table 1 was woven to form coloured fabric.

[0231] The results are given in Table 1.

[0232] Table 1

[0233]

[0234]

[0235] *Optical colour mixing qualitative index: (1) Not perceived as a solid colour, (2) Perceived as a solid colour from a distance of 10 cm or more, (3) Perceived as a solid colour at any distance, even 1 cm away.

Claims

Claims1. A method of producing a coloured fabric, the method comprising:providing a plurality of yarns, each yarn having one of a set of primary yarn colours;optionally, identifying a colour recipe for a desired fabric colour, the colour recipe comprising a primary yarn colour warp composition and a primary yarn colour weft composition; andweaving yarns of the plurality of yarns to form the fabric optionally according to the identified colour recipe.

2. The method of claim 1, wherein providing the plurality of yarns comprises: adding colourant to a material comprising one or more polymers to dye the material; andextruding the material to form the plurality of yarns.

3. The method of claim 1 or claim 2, wherein the set of primary yarn colours comprises between 2-50 colours, such as 5-50 colours, 5-30 colours, 5-15 colours, or 5-10 colours.

4. The method of any one of the preceding claims, wherein weaving yarns of the plurality of yarns comprises weaving warp yarn and weft yarn together, wherein the warp yarns comprise yarns of at least one colour or at least two colours of the set of primary yarn colours, such as at least two colours (e.g. two colours).

5. The method claim 4, wherein the weft yarns comprise yarns of one, two, three or four colours of the set of primary yarn colours, such as two or three colours (e.g. two colours).

6. The method of any one of the preceding claims, wherein the plurality of yarn comprises yarn and / or polychromatic yarn having a thickness from ID to 100 D, such as from 5D to 60 D.

7. The method of any one of the preceding claims, wherein the weaving of the plurality of yarns comprises weaving warp yarns and weft yarns together to achieve adensity of the coloured fabric from 100 EPI to 600 EPI, preferably from 100 EPI to 300 EPI, such as from 150 EPI to 250 EPI.

8. The method of any one of the preceding claims, wherein the plurality of yarn comprises yarn formed of, or comprising, one or more filaments having colours which are different to the primary colour of the resultant yarn, optionally wherein the yarn is formed via twisting and / or grouping the one or more filaments.

9. The method of any preceding claim, wherein identifying a colour recipe for the desired fabric colour comprises retrieving a pre-calculated and / or experimentally determined colour recipe from a memory.

10. The method of any of claims 1-8, wherein identifying a colour recipe for the desired fabric colour comprises calculating the colour recipe.

11. The method of claim 9 or claim 10, wherein the colour recipe is calculated using an optimisation algorithm to minimise a colour difference between a desired fabric colour and a predicted fabric colour.

12. The method of claim 11, wherein the colour difference is defined using a function in CIELAB coordinates.

13. The method of claim 11 or claim 12, wherein the optimisation algorithm is a gradient descent algorithm.

14. The method of claim 11 or claim 12, wherein the optimisation algorithm comprises determining the colour difference between the predicted fabric colour for each possible colour recipe and the desired fabric colour.

15. The method of any of claims 11-14, wherein the predicted fabric colour is determined using a colour predictor, wherein the colour predictor is configured to receive a colour recipe as an input and provide a predicted fabric colour as an output.

16. The method of claim 15, wherein the colour predictor is a neural network, and wherein the neural network comprises primary yarn colour compositions as inputs.

17. The method of claim 16, wherein, for each primary yarn colour, the weft composition and warp composition are represented as separate nodes in the input layer of the neural network.

18. The method of claim 16 or claim 17, wherein the values of the input nodes are constrained based on one or more of:(i) the total number of yarns to be used to produce the fabric;(ii) the number of warp yarns to be used to produce the fabric; and / or(iii) the number of warp yarns to be used to produce the fabric.

19. The method of any of claims 16-18, wherein the nodes of the output layer of the neural network represent the spectral reflectance values of the predicted fabric colour for a plurality of wavelengths.

20. The method of any preceding claim, further comprising:generating weaving instructions based on the identified colour recipe.

21. A computer-implemented method of calculating a colour recipe for producing a coloured fabric, the method comprising:using an optimisation algorithm to minimise the colour difference between a desired fabric colour and a predicted fabric colour;wherein the predicted fabric colour is determined using a colour predictor, and wherein the colour predictor comprises primary yarn colour compositions as inputs; and wherein, for each primary yarn colour, the weft composition and warp composition are represented as separate inputs to the colour predictor.

22. The method of claim 20 or 21, wherein the colour predictor is a neural network, and wherein the primary yarn colour warp compositions and primary yarn colour weft compositions are represented via separate nodes in the input layer of the neural network.

23. The method of any one of claims 20-22 , wherein the optimisation algorithm is a gradient descent algorithm.

24. The method of any of claims 20 to 23, wherein the neural network is a multilayer perceptron.

25. The method of any of claims 20 to 24, wherein the nodes of output layer represent the predicted reflectance value of the coloured fabric for a plurality of wavelengths.

26. A system for providing a colour recipe, the system comprising:an interface configured to enable a user to input a desired fabric colour;a processor configured to receive the desired fabric colour and identify a corresponding colour recipe; andan output configured to provide the colour recipe to the user;wherein the processor is configured to identify the colour recipe corresponding to the desired fabric colour by:(i) calculating the colour recipe using the method of any of claims 21-25; or(ii) retrieving the colour recipe from a memory, wherein the colour recipe was pre-calculated using the method of any of claims 21- 25 or determined experimentally.

27. The system of claim 26, wherein the interface is configured to receive one or more constraints from the user.

28. A coloured fabric obtained from the method of any one of claims 1 to 20.

29. A coloured fabric comprising:warp yarn of at least one colour or at least two colours, each colour individually selected from a set of primary yarn colours; andweft yarns of at least one colour or at least two colours, each selected from the set of primary yarn colours.

30. The coloured fabric of claim 29 having any features of claims 1 to 8.

31. The method, system or coloured fabric of any preceding claim wherein the plurality of yarn comprises one or more polychromatic yarns.

32. The method, system or coloured fabric of any preceding claim wherein the warp yarn and weft yarn comprises one or more polychromatic yarns.

33. The method, system or coloured fabric of any of claims 1 to 31 wherein one of the warp yarn or weft yarn comprises one or more polychromatic yarns.

34. The method, system or coloured fabric of claim 31, 32 or 33 wherein the one or more polychromatic yarns have a thickness from ID to 100 D, such as from 5D to 60 D.

35. The method, system or coloured fabric of any preceding claim wherein the weaving comprises plain weave, twills and satins, e.g. plain weave.

36. The method, system or coloured fabric of any one of the preceding claims, wherein the plurality of yarns comprises (e.g. consists of) dope dyed yarns.

37. The method, system or coloured fabric of any one of the preceding claims, wherein:(i) the set of primary yarn colours does not comprise all three of red, yellow and blue; and / or(ii) the set of primary yarn colours does not comprise all three of cyan, magenta and yellow.

38. The method, system or coloured fabric of any one of the preceding claims, wherein at least one of the colours of the set of primary yarn colours is not selected from red, yellow, blue, cyan, magenta, white and black.

39. The method, system or coloured fabric of any one of the preceding claims, wherein the method, system or coloured fabric does not substantially comprise jacquard weaving.

Images