Method of and apparatus for recording halftone image.

BACKGROUND OF THE INVENTION

Field of the invention

0001The present invention relates to technique for substantially improving gradation reproducibility of a halftone image by divisionally exposing each elementary area which serves as a unit for a value of screen pattern data.

Description of the Prior Art

0002In an apparatus for recording a halftone image such as a graphic arts scanner of electronic control type, halftone dots are produced by exposing a photosensitive material while turning on and off an optical exposure beam in response to the result of comparison between an image signal with a prescribed screen pattern signal. The screen pattern signal is generated on the basis of screen pattern data prepared in advance.

0003As is well known in the art, the screen pattern data express threshold values for respective small areas Ap within a halftone dot HD illustrated in Fig. 1. The small areas Ap (hereinafter referred to as "elementary areas") are formed by dividing the halftone dot HD in the form of a matrix. Fig. 2 illustrates screen pattern data prepared for the halftone dot HD and stored in a memory. Numerals in this figure represent data values P supplied to respective elementary areas Ap. Referring to Fig. 1, shaded areas have respective screen pattern data P which are at most 12, whereby the shaded areas are exposed with an optical exposure beam when an image signal having a value of 12 is supplied for the halftone dot HD.

0004The size of the elementary area Ap is determined according to a spot diameter d of the optical exposure beam focused on the photosensitive material. In more concrete terms, the size of the elementary areas Ap is so determined that a value of the spot diameter d is from about a length a (hereinafter referred to as "elementary area size") of one side of each elementary ärea Ap to about twice fo that. It is to be noted that the spot diameter d is assumed to be equal to the diameter of a circle inscribed in the elementary area Ap in the drawing, for convenience of illustration.

0005Judgement is made as to whether the optical exposure beam is turned on or off for every scanning progress corresponding to the spot diameter d, whereby intensity of the optical beam is changed in response to the result of the judgement. In the conventional method of recording a halftone image, therefore, a spatial interval which is a unit of the intensity change of the optical beam is limited to integral multiples of the elementary area size a.

0006Thus, when the following relation (1) holds for a square halftone dot having a screen pitch K, gradation number M defined by the equation (2) is the upper limit of number of levels for gradation expression: <ul id="ul0001" list-style="none"><li>K = n a (1)</li><li>M = (K_la)² = n² (2)</li></ul> where n is an integer.

0007In order to increase the gradation number M, therefore, it is necessary to increase (K/a) in the equation (2). In a scanner of a flat bed type, however, the focal length of an image forming lens in a recording optical system is attaining rather long in order to ensure scanning length to some extent. Thus, the spot diameter d of the optical exposure beam cannot be so much reduced in the flat-bed type scanner. For example, implementation of a spot diameter of about ten to twenty micrometers causes significant technical difficulty and increase in cost. Consequently, it is unavoidable that the elementary area size a is relatively increased and the gradation number M is reduced along the equation (2) in the flat-bed type scanner.

0008Also in a drum type scanner which can employ relatively small value of the spot diameter d (and the elementary area size a as the result), the gradation number M becomes small along the equation (2) if the screen pitch K is small

0009Thus, in the conventional apparatus for recording a halftone image, the gradation number M remains small when the ratio of the screen pitch K to the elementary area size a cannot be much increased, whereby gradation reproducibility of the image remains rather poor.

0010Even if the spot diameter d of the optical exposure beam can be reduced, number of the screen pattern data must be increased in order to increase the gradation number M, and hence capacity of a screen pattern memory must be increased, thus causes considerable increase in cost.

SUMMARY OF THE INVENTOIN

0011The present invention is directed to a method of and an appratus therefore for producing a halftone image comprising a plurality of halftone dots on a photosensitive material, a full region of one halftone dot being divided into a plurality of unit areas, said method comprising the steps of: (a) preparing image data expressing density for each unit area on an image plane, (b) preparing threshold data for each unit area on the image plane, (c) comparing the image data with the threshold data for each unit area on the image plane, thereby designating each unit area as a full-exposure area to be entirely exposed, a semi-exposure area to be divisionally exposed or a non-exposed area not to be exposed, and (d) exposing unit areas on the photosensitive material corresponding to the full-exposure area and the semi-exposure area on the image plane, thereby producing the halftone image, where a ratio of a exposed portion in the semi-exposure area to a whole of one unit area is determined on the basis of a difference between the image date and the threshold data for the semi-exposure area.

0012Preferably, the photosensitive material is exposed with a light beam scanning the halftone image at a prescribed speed, and a size of the exposed portion of the semi-exposure area is controlled by adjusting exposure time in the semi-exposure area in proportion to the ratio.

0013The threshold data may be formed as a set of threshold values assigned for respective unit areas included in the full region of one halftone dot, where the threshold values are arranged into an arithmetic sequence of numbers having a prescribed number interval.

0014The image data and the threshold data may be a digital data consisting of a same number of bits.

0015According to an aspect of the present invention, the step (c) further comprises the steps of: (c-1) finding which of the image data and the threshold data is larger, to thereby disignate the unit area as the non-exposure area or as a exposure area to be exposed, (c-2) finding a value substantially expressing the difference between the image data and the threshold data for each unit area, and (c-3) designating the exposure area as the full-exposure area when the difference is out of a prescribed range, or as the semi-exposure area when the difference is within the prescribed range.

0016The exposed portion of the semi-exposure area may contact a boundary between the semi-exposure area and a neighbor unit area neighboring the semi-exposure area, where the neighbor unit area corresponds to the exposure area.

0017An object of the present invention is to provide a method of and an apparatus for recording a halftone image, where gradation reproducibility of a recorded image is substantially improved without reducing the ratio of a screen pitch to a spot diameter of an optical exposure beam and without increasing capacity of a screen pattern memory.

0018These and other objects, features, aspects and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

0019<ul id="ul0002" list-style="none"><li>Fig. 1 is a diagram showing a halftone dot;</li><li>Figs. 2 and 3 are diagrams showing screen pattern data for the halftone dot;</li><li>Figs. 4A to 4D illustrate processing corresponding to several exposure conditions for a preferred embodiment of the present invention;</li><li>Figs. 5A to 5D, 6A to 6D, 7A, 8 and 9 are diagrams schematically showing methods of divisional exposure according to the present invention;</li><li>Fig. 10 is a block diagram showing an apparatus according to the preferred embodiment of the present invention;</li><li>Figs. 11 A, 11 B, 12 and 13A to 13C are block diagrams showing internal structure of a halftone dot generater in the embodiment of the present invention; and</li><li>Figs. 14, 15A to 15C and 16 are timing charts showing operation of the embodiment of the present invention.</li></ul>

DESCRIPTION OF THE PREFERRED EMBODIMENT

0020The basic idea of the present invention will be first described with reference to some examples.

0021As to exposed portions (shown as shaded areas) of a halftone dot HD shown in Fig. 1, elementary area (or pixel) Ap thereof is exposed when a value of screen pattern data P for the area elementary Ap is not more than a value of image data N, as shown in the following equation (3): <ul id="ul0003" list-style="none"><li>Exposure condition: P S N (3)</li></ul>

0022For the purpose of simplification, it is assumed that both of the screen pattern data P and the image data N are of 8-bit digital signals and the screen pattern data P are set as shown in Fig. 3 for the halftone dot HD. In other words, the image data N, which express density for each elementary area, are within a range of zero to 255 in decimal notation, and can express 256 gradation levels. On the other hand, one halftone dot HD has 16 elementary areas (or pixels) Ap, and the screen pattern data P discriminates only 16 gradation levels although the same are 8-bit data.

0023Now, an i-th bit value of the image data N, where i is an integer, is expressed as n_i. The image data N, more significant bit data (MSBD) N_u and less significant bit data (LSBD) N_l, are defined as follows in binary notation:<maths id="math0001" num=""><img file="EP0337427A2_D0001.tif" /></maths><maths id="math0002" num=""><img file="EP0337427A2_D0002.tif" /></maths><maths id="math0003" num=""><img file="EP0337427A2_D0003.tif" /></maths>where ni through n₈ represent 1 or 0, respectively.

0024The MSBD N_u is obtained by neglecting four tew significant bits of the image data N and the LSBD N_l, is composed of only the four less significant bits.

0025The number of effective bits (= 4) of the MSBD N_u is determined so that the MSBD N_u can express the number of gradation levels (2⁴ = 16) which is identical to the number of levels of the screen pattern data P. The LSBD N_l, is obtained by subtracting the MSBD N_u from the image data N.

0026When either of the following conditions holds for the elementary area Ap in which the exposure condition expressed by the equation (3) holds, only a part of the elementary area Ap is exposed: <ul id="ul0004" list-style="none"><li>Divisional Exposure Condition C-1:<maths id="math0004" num=""><img file="EP0337427A2_D0004.tif" /></maths></li><li>Divisional Exposure Condition C-2:<maths id="math0005" num=""><img file="EP0337427A2_D0005.tif" /></maths></li></ul>

0027In the equations (7) and (8), the values of the screen pattern data P, the image data N and the MSBD N_u are supposed to be expressed in decimal numbers, respectively. The part of the elementary areas Ap to be exposed is decided in accordance with the LSBD N . One elementary area Ap is divided into unit portions to be exposed discriminatively. The number of the unit portions is referred to as "dividing number D".

0028The relation among the image data N, the screen pattern data P, the gradation number M and the dividing number D will be described below.

0029Assuming that G_N represents the gradation number (or number of gradation levels) of the image data N, the dividing number D of the elementary area Ap is so decided that the following relation holds between the halftone dot gradation number M and the dividing number D:<maths id="math0006" num=""><img file="EP0337427A2_D0006.tif" /></maths>G_N = 256, M = 16 and D = 16 in the above example.

0030The number of digit of MSBD N_u are so determined that it can express the same gradation number (= 2⁴) as the halftone dot gradation number M. Therefore, the gradation number G_N of the image data N can be expressed by a combination of the halftone dot gradation number M and the dividing number D if the dividing number D corresponds to the gradation number (= 2⁴) of the LSBD N_i. The screen pattern data P are throught to be threshold values for classifying all the image data N into a certain number of groups, which number is identical to the halftone dot gradation number M.

0031For example, Table 1 shows the screen pattern data P of Fig. 3, expressed in both decimal and binary numbers. <tables id="tabl0001" num="0001"><img file="EP0337427A2_D0007.tif" /></tables>The screen pattern data P are expressed in eight bits, as the same with the image data N, and obtained by adding "1" to respective fifth bit values (least significant bit values within the four more signaificant bit values) in binary notation. The least significant bits of the screen pattern data P are regularly "1" and second to fourth bit values are regularly "0".

0032As the result of such structure of the screen pattern data P, at least one screen pattern data P is present, in which the equality in the equation (7) or (8) holds for any image data N.

0033Figs. 4A to 4D are diagrams showing of processing in each case of the two divisional exposure conditions C-1 and C-2. In the following description, it is assumed that the screen pattern data P, the image data N, MSBD N_u and the LSBD N_l are expressed in decimal numbers.

0034In the case of N = 5, an elementary area A₁ (see Fig. 3) is subjected to exposure, becasue the exposure condition of the equation (3) holds for the elementary area A₁. Since N_u + 1 = 1 in this case, the divisional exposure condition C-1 (above equation (7)) also holds for the elementary area A₁, . Fig. 4A is a partially enlarged view of Fig. 3, for illustrating the state of divisional exposure of the elementary area A₁. As shown in Fig. 4A, the elementary area A₁ is divided into 16 (the dividing number.D = 16) along a main scanning direction y, to be divisionally exposed at the rate of N₁/16. When the image data N is 5, the LSBD N₁ is also 5. Therefore, only a portion A₁ (5) which is 5/16 of the elementary area A₁ is exposed as shown as shaded area with slanting lines in Fig. 4A.

0035Fig. 4B shows the case of N = 15. N_u + 1 = 1 holds also in this case, and hence divisional exposure is performed on the elementary area A₁ similarly to the above. As shown by shaded area with slanting lines in Fig. 4B, only a portion A, (15) which is 15/16 of the elementary area A₁ is exposed.

0036Also in the case of N = 16, the elementary area A₁ is subjected to exposure because the equation (3) holds. However the equation (7) is not satisfied because N_u + 1 = 17, and hence divisional exposure is not performed on the elementary area A₁ , but it is entirely exposed. Other elmentary areas other than the elementary area A₁ are not exposed since the equation (3) does not hold for them. In another point of view, an elementary area A,₇ is subejcted to divisional exposure because N_u + 1 = 17, and it is exposed at the rate of 0/16 because N_l, = 0.

0037Fig. 4C shows the case of N = 17. In this case the elementary areas A₁ and A,₇ are subjected to exposure, in which the equation (3) holds. The elementary area A₁₇is divisionally exposed because N_u + 1 = 17, while the elementary area A₁ is' entirely exposed. A portion A₁₇(1) which is 1/16 of the elementary area A₁₇ is exposed as shown in Fig. 4C, because N₁ = 1.

0038Fig. 4D shows the case of N = 32. In this case, the elementary areas A₁ and A₁₇ are subejcted to exposure according to the equation (3), and no divisional exposure is performed because N_u + 1 = 33. Thus, the elementary areas A_l and A₁₇ are entirely exposed as shown in Fig. 4D.

0039As described above, whether each elementary area Ap is subjected to exposure or not is determined on the basis of the exposure condition of the equation (3). Further, the elementary areas to be divisionally exposed (hereinafter referred to as "divisional exposure areas") are judged on the basis of the divisional exposure condition C-1 of the equation (7), while the rate of exposure is determined by the LSBD N_I. As the result, the size of exposed areas is increased in proportion to the image data N, whereby a reproduction gradation number becomes identical to the gradation number of the image data N.

0040Sequence of processing along the conditions of the equations (3) and (7) may be inverted to the above. That is, judgement may be first made as to whether or not divisional exposure is performed along the condition of the equation (7) with the image data N, whereby a portion of an elementary area to be divisionally exposed is determined by the LSBD N_l. Then, judgement may be made on the other elementary areas not to be divisionally exposed as to whether or not the same are entirely exposed according to the condition of the equation (3). For example, the elementary area A₁ is divisionally exposed when N = 0, since N_u + 1 = 1. However, the rate of exposure is 0/16 since N = 0, and hence the elementary area A₁ is not exposed as the result. Thus, the same result is obtained whichever condition of the equations (3) and (7) is judged first.

0041In the case of the divisional exposure condition C-1, the image data N and the screen pattern data P are compared with each other with repect to the respective elementary areas A₁ to A₂₂₅ in the halftone dot HD as hereinabove described, whereby some poritons of the respective elementary areas are exposed at need. Further, it is desirable that the exposed portions are linked with the exposed portions of adjacent elementary areas, in order to form a halftone dot having an excellent configuration. Therefore, position of an exposed portion in a divisional exposure areas is decided as follows, in consideration of exposure states of adjacent elementary areas: <ul id="ul0005" list-style="none"><li>Figs. 5A to 5D are is an explanatory diagrams illustrating a method of deciding position of an exposed portion of a divisional exposure area. Referring to Fig. 5, symbols B_m-1 and B_m+1 denote boundaries between a divisional exposure area A_m and two elementary areas A_m-1 and A_m+1which are adjacent to the same along the main scanning direction y respectively. There are four combinations of exposure states for respective elementary areas A_m-1 and A_m+1 The exposed portion of the divisional exposure area A_m is located at either of the boundaries B_m-1 and B_m+1 in response to the combination, as shown in Table 2.<img file="EP0337427A2_D0008.tif" />Referring to Table 2, the exposure state "0" indicates that the elementary area is not exposed and the exposure state "1" " indicates that it is exposed. The exposure state "1" also includes divisional exposure of the elementary area. These exposure states are judged through the above equation (3), so that the exposure state "1" is selected when the equation (3) is satisfied while the exposure state "0" is selected when the equation ( 3) is not satisfied.</li></ul>

0042When the adjacent elementary area A_m-1 is exposed, an exposed portion (hereinafter referred to as "divisional exposure portion") A_md of the divisional exposure area A_m is formed to be in contact with the boundary B_m-1 as shown in case 1 and case 2 of Table 2. Figs. 5A and 5B-show divisional exposure states corresponding to the case 1 and the case 2, respectively.

0043Figs. 6A to 6D shows the states of divisional exposure corresponding to the case 1 and the case 2 in detail. Figs. 6A to 6D correspond to the cases where the less significant bit data (LSBD) N₁ are 1, 8, 10 and 15, respectively. When the LSBD N₁ is increased, the divisional exposure portion A_md is increased in the direction (+y) from the boundary B_m-1' Thus, the divisional exposure portion A_md is exposed to regularly contact the adjacent elementary area A_m+1 in the case 1 and the case 2 of Table 2. This condition is hereafter referred to as "divisional exposure condition C-1 a".

0044When the adjacent elementary area A_m-1 is not exposed as shown in the case 3 and the case 4 in Table 2, on the other hand, the divisional exposure portion A_md is formed to be in contact with the boundary B_m+1. Figs. 5C and 5D show divisional exposure states in the case 3 and the case 4, respectively.

0045Figs. 7A to 7D shows the states of divisional exposure of the case 3 and the case 4 in detail. Figs. 7A to 7D correspond to the cases where the LSBD N₁ are 1, 8, 10 and 15, respectively. When the LSBD N₁ is increased, the divisional exposure portion A_md is increased in the direction (-y) from the boundary B_M+1' Thus, the divisional exposure portion A_md is exposed to regularly contact the adjacent elementary area A_m+1 in the case 3 and the case 4 of Table 2. This condition is hereafter referred to as "divisional exposure condition C-1 b".

0046As hereinabove described, the divisional exposure portion A_md is linked with exposed portion of the elementary area which is adjacent to the divisional exposure area A_m by performing divisional exposure in accordance with the divisional exposure condition C-1 a or C-1 b, thereby to form a halftone dot of an excellent configuration where exposed portions contact each other.

0047The divisional exposure condition C-2 expressed by the equation (8) denotes that the case where the image data N are in a range of 240 &lt; N(d) 5 255 is independently treated. The image data N of this range designates an elementary area A₂₄₁ to be exposed, while the number of gradation levels of the image data N (= 241 to 255) in this range is 15, dissimilarly to the gradation number 16 of the image data N corresponding to other elementary areas. Thus, the image data N in the range (N = 241 to 255) is treated dissimilarly to those in other ranges in performing divisional exposure.

0048Consider such case where the dividing number of the elementary area A₂₄₁ is 16, similarly to that for the other elementary areas. Fig. 8 is an explanatory diagram showing the elementary area A241, which is divided into 16, in an enlarged manner. The LSBD N₁ correspondinig to the maximum value Nmax (= 255) of the image data N is 15. Hence the elementary area A₂₄₁ is exposed only at the rate of 15/16 in response to the maximum value N_max when divisional exposure is performed similarly to the divisional exposure condition C-1. Namely, such a problem is caused in this case that the halftone area rate is not 100 % even for the maximum value N_max.

0049In order to solve this problem, (N_i + 1)/16 of the elementary area A₂₄₁ is exposed in response to the LSBD N₁ when the divisional exposure condition C-2 holds. When N = 241, i.e., N₁ = 1, for example, a portion A₂₄₁(2), which is 2/16 of the elementary area A₂₄₁, is divisionally exposed, as shown in Fig. 8. Thus, the entire elementary area A24, is exposed when N = 255.

0050Fig. 10 shows the structure of a graphic arts scanner of a flat bed type in which an embodiment of the present invention can be implemented.

0051Referring to Fig. 10, digital image data N including image information to be recorded are inputted in a halftone signal generator 1, the structure of which will be described in detail. The halftone signal generator 1 generates an exposure signal S on the basis of the digital image data N which is inputted into the generator for every elementary area sequentially along scanning lines. The exposure signal S controls formation of halftone dots. A microcomputer 2 for operation control is connected to the halftone signal generator 1. This microcomputer 2 has a CPU 3 and a memory 4, and is connected to a keyboard 5 for inputting control parameters.

0052A laser beam L is generated in a laser oscillator 7, which serves as a light source for exposure. The laser beam L is modulated in an acoustic optical modulator (AOM) 8 in response to the exposure signal S, then introduced into a beam expander 14. A laser beam L outputted from the beam expander 14 reaches a galvano mirror (or polygonal rotating mirror) 15 to be deflected therein for scanning, and is focused onto the surface of a photosensitive material 17 through an fe lens 16.

0053The laser beam L is periodically and repeatedly moved in a direction Y in response to vibration of the galvano mirror 15, or rotation of the polygonal rotating mirror provided in place of the galvano mirror 15, to thereby move along the main scanning direction. The photosensitive material 17 is moved in a direction perpendicular to the plane of the figure, to thereby move along a subscanning direction. Consequently, the photosensitive material 17 is scanned and exposed along the main scanning and subscanning directions, whereby a halftone image corresponding to the digital image data N is recorded on the photosensitive material 17.

0054In this apparatus, divisional exposure is controlled as follows, for example: Fig. 9 is an explanatory diagram showing exposure control of the divisional exposure portion A_md. Referring to Fig. 9, scanning of an optical exposure beam is made along the main scanning direction y, in order to expose the divisional exposure portion A_md in the elementary area A_m within the elementary areas A_m-1 to A_m+1. The main scanning direction y is horizontal in this figure, for convenience of illustration. An exposure signal S keeps an ON state during an interval between times t_b and t_c, at which a center 0 of the optical exposure beam reaches respective ends of the divisional exposure portion A_md. As the result, the divisional exposure portion A_md is photosensitized by receiving light in an amount exceeding critical exposure of the photosensitive material, while other portions are not photosensitized. The optical exposure beam is thus controlled to form a recording image in which the divisional exposure portion A_md as a part of the elementary area A_m is photosensitized. Although the shape of the photosensitized area is not completely rectangular as shown in Fig. 9 in practice, exposure of the divisional exposure poriton A_md is substantially achieved because the interval for keeping the exposure beam at an ON state is proportional to the photosensitized area.

0055In order to expose the entire elementary area A_m, an ON state is maintained during an interval between times t_a and t_c, at which the center O of the optical exposure beam reaches respective positions of the boundaries B_m-1 and B_m+1.

0056Now, the structure and operation of the halftone signal generator 1 will be described in more detail.

0057Figs. 11A and 11B illustrate internal structure of the halftone signal generator 1. Referring to Fig. 11A, the digital image data N inputted sequentially for respective elementary areas are supplied to comparators 41 and 42 and an adder 31, respectively.

0058The screen pattern data P for respective elementary areas are also sequentially inputted in the comparator 41 from a screen pattern memory 30. The screen pattern data P are outputted from the screen pattern memory 30 in synchronism with a first clock signal φ₁, which is inputted in the memory 30. The first clock signal φ₁ has a cycle (one clock cycle) which corresponds to the width a (see Fig. 1) of the image, and the image data N are also inputted into the halftone signal generator 1 in synchronism with the first clock signal φ₁. Thus, the image data N and the screen pattern data P are synchronously and sequentially inputted in the comparator 41 for respective elementary areas.

0059The comparator 41 compares the values of the image data N with those of the screen pattern data P, to generate a signal S₄₁ which becomes "1" level when P ≦ N holds while becoming "0" level when N &lt; P. In other words, the comparator 41 judges whether or not the exposure condition of the equation (3) is satisfied, and outputs a "1" level signal when it is satisfied.

0060This output signal S₄₁ is sequentially latched by D flip-flops 51 and 54 on the leading edge of the first clock signal φ₁. Namely, an output signal S₅₄ of the flip-flop 54 is the same with an output signal S₅₁ of the flip-flop 51 which is outputted one clock cycle before. Assuming that symbols S₅₁(m) and S₅₄(m) represent the signals S₅₁ and S₅₄ relating to an m-th elementary area A_m in general, the flip-flop 54 latches a signal S₅₄(m-1) relating to an (m - 1)-th elementary area at the time when the flip-flop 51 latches the signal S₅₁(m) relating to the m-th elementary area A_m.

0061As the result, the output signals S₅₁(m) and Ss₄(m) relating to the image data N of the m-th elementary area A_m are formed as follows:<maths id="math0007" num=""><img file="EP0337427A2_D0009.tif" /></maths>

0062The comparator 42 outputs a signal S₄₂, which becomes a "1" level when the image data N satisfies the relation of 240 &lt; N while becoming a "0" level when N ≦ 240. In other words, the comparator 42 substantially selectes one of the divisional exposure conditions C-1 and C-2 (see the above equations (7) and (8)).

0063This output signal S₄₂ is latched in the flip-flop 52 in synchronism with the first clock signal φ₁, to become a signal S₅₂ which is synchronous with the output signal Ss of the flip-flop 51. Therefore, an output signal S₅₂(m) relating to the image data N of the m-th elementary area A_m is as formed follows:<maths id="math0008" num=""><img file="EP0337427A2_D0010.tif" /></maths>

0064The adder 31 generates a signal S₃₁, which is obtained by adding "1" " to the MSBD N_u of the image data N, and this output signal S₃₁ is supplied to the comparator 43. The screen pattern data P from the screen pattern memory 30 are also inputted in the comparator 43, to be compared with the signal S₃₁. An output signal S₄₃ of the comparator becomes "1" level when P = N_u + 1 holds. In other words, the comparator 43 judges whether or not either of the divisional exposure condition C-1 or C-2 holds.

0065The output signal S₄₃ is latched in the flip-flop 53 in synchronism with the first clock signal φ₁, to become a signal Ss₃ which is synchronous with the signals S₅₁ and Ss₂. Therefore, an output signal Ss₃(m) with respect to the image data N of the m-th elementary area A_m is as formed follows:<maths id="math0009" num=""><img file="EP0337427A2_D0011.tif" /></maths>

0066The output signals S₅₂, S₅₃ and Ss₄ thus obtained are inputted in three AND gates 71, 72 and 73 in the following manner, whereby divisional exposure areas are detected and the exposure condition is judged: <ul id="ul0006" list-style="none"><li>First, the signals S₅₃ and S₅₄ and an inverted signal S₅₂ of the signal S₅₂ obtained in an inverter 65 are inputted in the AND gate 71. Assuming that the signals S₅₂ and S₅₃ are signals S₅₂(m) and S₅₃(m) with respect to the m-th elementary area A_m respectively, the signal S₅₄ which is synchronous with them is a signal S₅₄(m-1) for the elementary area A_m-1.</li></ul>

0067Therefore, from the equations (10) to (12), an output signal S₇₁ of the AND gate 71 becomes "1" level under the following condition:<maths id="math0010" num=""><img file="EP0337427A2_D0012.tif" /></maths>In other words, the AND gate 71 outputs the signal S₇₁ of a "1" level when rhe elementary area A_m satisfies the divisional exposure condition C-1 a (the case 1 or 2 in Table 2).

0068Similarly, the signals S₅₂ and S₅₃, and an inverted signal S₅₄ of the signal S₅₄ obtained in an inverter 64 are inputted in the AND gate 72, and hence an output signal S₇₂ thereof becomes "1" level under the following condition:<maths id="math0011" num=""><img file="EP0337427A2_D0013.tif" /></maths>In other words, the AND gate 72 outputs the signal 8₇₂ of a "1" " level when the elementary area A_m satisfies the divisional exposure condition C-1 b (the case 3 or 4 in Table 2).

0069The signals S₅₂ and S₅₃ are inputted in the AND gate 73, and hence an output signal S₇₃ thereof becomes a "1" level under the following condition:<maths id="math0012" num=""><img file="EP0337427A2_D0014.tif" /></maths>In other words, the AND gate 73 outputs the signal S₇₃ of a "1" level when the elementary area A_m satisfies the divisional exposure condition C-2.

0070As can be understood from the above description, the adder 31, the comparators 41 to 43, the flip-flops 51 to 54, the inverters 64 and 65, the AND gates 71 to 73 and the like form comparator means for comparing the image data N with the screen pattern data P with respect to the elementary area A_m while generating a signal for divisional exposure indicating that the elementary area A_m is subjected to divisional exposure when any one of the conditions of the equations (13) to (15) holds.

0071Now, judging operation of the divisional exposure conditions through the comparator means will be described in more detail.

0072Fig. 14 is a timing chart showing the judging operation of the divisional exposure conditions by the AND gate 71. The figure is related to processing executed on the column of the elementary areas including the elementary area A₁ shown in Fig. 3. Referring to Fig. 14, the screen pattern data P and the image data N relating to respective elementary areas are inputted at an internal of a cycle T₁ in synchronism with the leading edge of the first clock signal φ₁. That is, the image data N and the screen pattern data P are inputted with respect to an elementary area A₆₅ during an interval between times to and t₁, with respect to the elementary area A₁ during an interval between times t₁ and t₂, with respect to an elementary area A₁₇ during an interval between times t₂ and t₃ and with respect to an elementary area A₁₄₅ during an interval between times t₃ and t₄, respectively.

0073The output signal S₄₁ from the comparator 41 becomes "1" level when P ≦ N, while becoming a "0" level when N &lt; P. In the example shown in Fig. 14, therefore, the signal S₄₁ is at a "1" " level with respect to the elementary areas A₆₅, A₁ and A₁₇.

0074The signal S₅₁ is delayed by the cycle T₁ from the signal S₄₁, and the signal S₅₄ is further delayed by the cycle T₁ from the signal S₅₁.

0075The output signal S₄₃ from the comparator 43 becomes a "1" level when P = N_u + 1. Relation between the screen pattern data P and (N_u + 1) is shown in the lower part of Fig. 14, where P = N_u + 1 holds only when P = 1 and N = 14. Therefore, the signal S₄₃ is at a "1" level with respect to the elementary area A₁ during the interval between the times t₁ and t₂. The output signal S₅₃ from the flipflop 53, which is delayed by one cycle T₁ from the signal S₄₃, becomes a "1" level during the interval between the times t₂ and t₃.

0076The output signal S₄₂ from the comparator 42, which signal is not shown in Fig. 14, is maintained at a "0" level during the interval between the times to and t₄ since the image data N are not more than 240. Thus, the inverted signal S₅₂ of the output from the flip-flop 52 is maintained at a "1" level.

0077These signals S₅₂, S₅₃ and S₅₄ are inputted in the AND gate 71, whose output signal S₇₁ becomes "1" level only during the interval between the times t₂ and t₃. In other words, the signal S₇₁ is at a "1" " level since the divisional exposure condition C-1 a expressed by the above equation (13) holds for the elementary area A₁. It is to be noted that A_m = A₁ and A_m-1 = A₆₅ in the equation (13).

0078Operation of the AND gates 72 and 73 for detecting divisional exposure areas is similar to the timing chart shown in Fig. 14, and hence detailed description is omitted.

0079The signals S₇₁ to S₇₃ thus obtained are inverted in the inverters 61 to 63 shown in Fig. 11B, respectively, and supplied to a four-input AND gate 91. The output signal S₅₁ from the flip-flop 51 is also supplied to the four-input AND gate 91. Therefore, an output signal S₉₁ from the four-input AND gate 91 becomes "1" level only when the elementary area A_m satisfies the exposure condition of the equation (10) while satisfying none of the divisional exposure conditions of the equations (13) to (15). In other words, the signal S₉₁ becomes "1" level when the elementary area A_m is entirely exposed. This signal S₉₁ is supplied to the AOM 8 as an exposure output signal S through a four-input OR gate 92.

0080On the other hand, a first signal generator 81 for generating divisional exposure signal shown in Fig. 11 B receives the output signal S₇₁, the LSBD N_i, a second clock signal φ₂ and a clear signal S₃₂ from a -clear output circuit 32. The second clock signal φ₂ is a clock signal which is synchronous with the first clock signal φ₁ and has a frequency 16 times that of the first clock signal φ₁. The clear signal generator 32 is adapted to generate the clear signal S₃₂ which is necessary for operation of the first signal generator 81, as described later in detail.

0081Similarly, a second signal generator 82 receives the output signal S₇₂, the LSBD N_i, the second clock signal φ₂ and the clear signal S₃₂. A third signal generator 83 receives the output signal S₇₃, the LSBD N_i, the second clock signal φ₂ and the clear signal S₃₂.

0082The first to third signal generators 81 to 83 generate divisional exposure signals S₈₁ to S₈₃ in accordance with the divisional exposure conditions C-1a, C-1b and C-2 expressed in the above equations (13) to (15), respectively. The divisional exposure signals S₈₁ to S₈₃ and the signal S₉₁ are inputted in the four-input OR gate 92, so that any one of the signals of "1 " level is supplied to the AOM 8 as the exposure signal S.

0083Internal structure of the signal generators 81 to 83 and their operation for each divisional exposure condition C-1a, C-1b and C-2 will be described in more detail.

0084Figs. 13A to 13C are block diagrams showing internal structure of the signal generators 81 to 83. Operation of the first signal generator 81 under the divided exposure condition C-1a is now described with reference to a timing chart shown in Fig. 15A.

0085In the case of the divisional exposure condition C- 1a, the AND gate 71 shown in Fig. 11A outputs the signal S₇₁ of a "1" level. This signal S₇₁ is supplied to an input terminal A of a shift register 81 L provided in the first signal generator 81 shown in Fig. 13A. The shift register 81 L, which receives serial input and sends parallel output, sequentially receives the signal S₇₁ at the input terinal A and output the same at output terminals P1 to P16 as a delay signal which is synchronous with the second clock signal φ₂ received at a input terminal C.

0086As shown in Fig. 15A, a cycle T₂ of the second clock signal φ₂ is 1/16 of the cycle T₁ of the first clock signal φ₁. Fig. 15A also shows relation between the input signal S₇₁ and an output signal S_81L3 at an output terminal P3. Referring to Fig. 15A, the first clock signai φ₁ rises at a time t₁₀ and the signal S₇₁ simultaneously becomes "1" level, whereby the signal S_81L3 at the output terminal P3 becomes "1" " level at a time t₁₁ delayed by a period of three times T₂ (3T₂). In other words, output signals S_81L1 to S_81L15, which appear at respective output terminals Pn(n = 1 to 15) rise in delays by periods of n times T₂ from the time t₁₀, respectively. The output signals S_81L1 to S_81L15 are inverted in inverters 81M1 to 81M15, respectively, to become signals S_81M1 to S_81M15. Fig. 15A shows only the signal S_81M3, as an example thereof.

0087On the other hand, the LSBD N_i of the image data N are supplied to an input terminal B of a decoder 81 N provided in the first signal generator 81. The LSBD N_i are in the form of a four-bit binary signal, and only one of signals S_81N0 to S_81N15 outputted from respective output terminals YO to Y15 beocmes "1" level in accordance with the decimal value (0 to 15) of the LSBD N_i. When N_i = 3, for example, only the output signal S_81N3 becomes "1" level while the other signals remain at a "0" level.

0088Three input AND gates 81 H1 to 81 H15 provided in the first signal generator 81 receive signals S_81M1 to S_81M15, respectively, and also signals S_81N1 to S_81N15, respectively, while the signal S₇₁ is commonly inputted in the three-input AND gates 81 H1 to 81H15. Referring to Fig. 15A showing the case of N_i = 3, an output signal S_81H3 of the three-input AND gate 81 H3 is at a "1" level during the interval of 3T₂ between the times t₁₀ and t₁₁. Output signals (not shown) of the remaining three-input AND gates are maintained at "0" levels. All of the output signals S_81H1 to S_81H15 are inputted in a 15-input OR gate 81 G, which in turn generates a first divisional exposure signal S₈₁.

0089As hereinabove described, in the case of the divisional exposure condition C-1 a, the divisional exposure signal S₈₁ holds "1" " level during a period of N₁ times T₂ in response to the output signal S₇₁ of the AND gate 71 and the LSBD N_i inputted in the first signal generator 81.

0090Since the cycle T₂ is 1/16 of the cycle Ti, the period of N₁ times T₂ in which the divisional exposure signal S₈₁ is at a "1" " level is N,/16 of the cycle T₁. Further, the cycle T₁ corresponds to the width a of one elementary area. Comparing Fig 15A with Fig. 9, therefore, it can be understood that the divisional exposure signal S₈₁ is an exposure signal S for controlling exposure of N_i/16 of the elementary area A_m from the boundary B_m-1 (or time ta).

0091The clear signal S₃₂ is inputted in a clear input terminal CLR of the shift register 81 L, to thereby clear the shift register 81 L every time the first clock signal φ₁ rises.

0092The internal structure of the clear signal generator 32 for generating the clear signal S₃₂ is shown in Fig. 12, while its operation is shown in 'Fig. 16. A flip-flop 32A provided in the clear signal generator 32 latches the value of the first clock signal φ₁ in response to the second clock signal φ₂, to produce a signal S_32A which is delayed by the cycle T₂ from the first clock signal φ₁. This signal S_32A and the first clock signal φ₁ are inputted in an exclusive OR gate 32B. An output signal S_32B from the exclusive OR gate 32B as well as the first and second clock signals φ₁ and φ₂ are inputted in a three-input AND gate 32C generating inversion output. As the result, the three-input AND gate 32C outputs the clear signal S₃₂, which holds "0" level during half the cycle T₂ every time the first clock signal φ₁ rises, as shown in Fig. 16. When the clear signal S₃₂ is inputted in the shift register 81 provided in the first signal generator 81, the shift register 81 L is cleared every time the first clock signal φ₁ rises. Thus, the shift register 81 L operates to process a new input signal S₇₁ at an interval of the cycle T₁ of the first clock signal φ₁.

0093In the case of the divisional exposure condition C-1 b, the AND gate 72 shown in Fig. 11 A outputs the signal S₇₂ of a "1" level. This signal S₇₂ is supplied to an input terminal A of a shift register 82L provided in the second signal generator 82 shown in Fig. 13B. The shift register 82L receives serial input and generates parallel output, similarly to the shift register 81 L. Therefore, it outputs signals S_82L1 to S_82L15, which rise with delays of periods T₂ to 15T₂ from the rise time of the signal S₇₂, respectively, in synchronism with the second clock signal φ₂ supplied to its input terminal C.

0094Fig. 15B is a timing chart showing the operation of the second signal generator 82. Referring to Fig. 15B, the signal S_82L3 rises with a delay of a period 3T₂ from a rise time t₂₀ of the first and second clock signals φ₁ and φ_2, and the output signal S₇₂.

0095A decoder 82N has function similar to that of the decoder 81 N, such that only one of signals S_82N1 to S_82N15, which are outputted from its output terminals Y1 to Y15, respectively, becomes "1" level in response to a decimal number of the LSBD N of the image data N supplied to its input terminal B.

0096Each of two-input AND gates 82J1 to 82J15 receives a combination of one of the signals S_82L1 to S_82L15 and one of signals S_82N1 to S_82N15. It is to be noted here that the two-input AND gate 82J1 receives the signals S_82L1 and S_82N15 while the two-input AND gate 82J2 receives the signals S_82L2 and S_82L14, for example. In other words, output signals S_82J1, S_82J2, ..., S_82J15 of the respective two-input AND gates 82J1 to 82J15 correspond to the output signals S_82L15, S_82L14, ..., S_82L1 of the shift register 82L. Consequently, only one of the output signals S_82L1 to S_82L15 of the shift register 82L, which one is delayed by a period of (16 - N ) T₂ from the signal S₇₂, is maintained as it is while the remaining output signals are entirely brought into "0" levels, when they are outputted from the two-input AND gates 82J1 to 82J15.

0097All of the output signals S_82J1 to S_82J15 from the respective two-input AND gates 82J1 to 82J15 thus obtained are inputted in a 15-input OR gate 82G, which in turn outputs a second divisional exposure signal S₈₂.

0098Referring to Fig. 15B showing the case of N₁ = 13, the divisional exposure signal Ss₂ rises at a time t₂₁ which is delayed by a period 3T₂ from the rise time t₂₀ of the signal 8₇₂, since (16 - N_i) = 3.

0099The shift register 82L is cleared by the clear signal S₃₂ received in its clear input terminal CLR every time the first clock signal φ₁ rises, and hence the divisional exposure signal Ss₂ is also simultaneously cleared at a time t₂₂. As shown in Fig. 15B, therefore, a period during which the divisional exposure signal S₈₂ holds "1" level is N_i times T₂(13T₂) in response to the decimal number of the LSBD N₁ (= 13). The rate of the duration is N/16 to the cycle T₁ of the first clock signal φ₁.

0100It can be understood by comparing Fig. 15B with Fig. 9, therefore, the divisional exposure signal S₈₂ is an exposure signal S which controls exposure from an intermediate portion of the elementary area A_m to the boundary B_m+1 between the elementary areas A_m and A_m+1. It is also understood that the exposed portion is N_i/16 of the elementary area A_m.

0101In the case of the divisional exposure condition C-2, the AND gate 73 shown in Fig. 11A outputs the signal S₇₃ of a "1" level. This sigal S₇₃ is supplied to an input terminal A of a shift register 83L provided in the third signal generator 83 shown in Fig. 13C. The shift register 83L is similar to the shift registers 81 L and 82L. However, its output terminal P1 is open and signals S_82L2 to S_82L16, which are delayed by periods 2T₂ to 16T₂, respectively, from the signal S₇₃, are outputted in synchronism with the second clock signal φ₂ received in its input terminal C. The signals S_82L2 to S_82L16 are inverted in inverters 83M1 to 83M15, respectively, to become signals S_83M1 to S_83M15.

0102Fig. 15C is a timing chart showing the operation of the third signal generator 83. Referring to Fig. 15C, trailing time of the signal S_83M1, which is inverted from the signal S_83L2, is delayed by a period 2T₂ from the rise time t₃₀ of the first and second clock signals φ₁ and φ₂, and the output signal S₇₃. That is, the signal S_83M1 is maintained at a "1" level during an interval 3T₂ between times t₃₀ and t₃₁, and falls to a "0" level at the time t₃₁.

0103Fig. 15C also shows the signal S_83L16 and its inverted signal S_83M15. Although the signal S_83L16 should rise with a delay of a period 16T₂ from the time t₃o, it is maintained at a "0" level until a time t₃₂ for a lapse of one cycle T₁, since 16T₂ = T₁. Therefore, its inverted signal S_83M15 is maintained at a "1" " level during the interval between the times t_3o and t₃₂.

0104On the other hand, a decoder 83N has function substantially equal to that of the decoders 81 N and 82N, such that only one of signals S_83N1 to S_83N15 outputted from respective terminals Y1 to Y15 becomes "1" level in response to the LSBD N_i supplied to its input terminal B.

0105Three-input AND gates 83H1 to 83H15 receive the respective signals S_83M1 to S_83M15 and the respective signals S_82N1 to S_83N15, while the signal S₇₃ is commonly inputted in the three-input AND gates 83H1 to 83H15. All of output signals S_83H1 to S_83H15 from the respective three-input AND gates 83H1 to 83H15 are inputted in a 15-input OR gate 83G, which in turn outputs a third divisional exposure signal S₈₃.

0106Fig. 15C shows the divisional exposure signal S₈₃ in the case of N₁ = 15. In this case, only S_83N15 is at a "1" level within the output signals S_83N1 to S_83N15 of the decoder 83N, whereby the signal S_83M15 corresponding thereto is outputted as the divisional exposure signal S₈₃ through the three-input AND gate 83H15 and the 15-input OR gate 83G. Since the signal S_83M15 is maintained at a "1" level as shown in the figure, the divisional exposure signal S₈₃ is also maintained at a "1" level throughout an interval between times t_3o and t₃₂.

0107Thus, the third signal generator 83 outputs the divisional exposure signal S₈₃ for controlling exposure of (N_i + 1)/16 of the elementary area A_m in response to the LSBD N_i of the image data N in the third divided exposure condition C-2.

0108As described above, one of the divisional exposure signals S₈₁ to S₈₃ is brought into a "1" level in response to each of the divided exposure conditions C-1a, C-1b and C-2. Incidentally, respective output terminals YO of the decoders 81 N to 83N are open in order not to expose the elementary area A_m in the case of N_i = 0. When N₁ = 0, all of the divisional exposure signals S₈₁ to S₈₃ become "0" levels becasue the signals S_81N1 to S_81N15, the signals S_82N1 to S_82N15 and the signals S_83N1 to S_83N15 of the decoders 81 N to 83N are all at "0" level.

0109When none of the divisional exposure conditions is satisfied, on the other hand, the divisional exposure signals S₈₁ to S₈₃ hold "0" level since the signals S₇₁ to S₇₃ are at "0" level.

0110As described above, the four-input AND gate 91 shown in Fig. 11 B receives the signals S₇₁ to S₇₃, which are inverted by the inverters 61 to 63 in advance, as well as the output signal S₅₁ of the flip-flop 51. Therefore, when none of the divisional exposure conditions is satisfied and the exposure condition of the equation (3) is satisfied, an exposure signal S which controls exposure of the entire elementary area is generated.

0111As described above, the image recording apparatus of this embodiment judges the exposure condition and the divisional exposure conditions for each elementary area and performs exposure in accordance wih the conditions, to thereby form a halftone dot having the same gradation number as the image data N. Further, it is not required to increase capacity of the screen pattern memory 30 etc. since no data are required other than the image data N and the screen pattern data P in order to generate the divisional exposure signal for controlling the divisional exposure.

0112The present invention is not restricted to the above embodiment, but the following modifications can be attainable, for example: <ul id="ul0007" list-style="none"><li>In the above embodiment, a portion of (N, + 1)/16 of the unit area A₂₄₁ is exposed in response to the LSBD N₁ of the image data N in the case of the divisional exposure condition C-2, whereby the gradation number is increased. However, a similar effect cna be attained, for example, by exposing a portion of N,/16 of the elementary area A₂₄₁ when N₁ is not more than 14 and exposing the entire elementary area A₂₄₁ when N_i = 15.</li></ul>

0113Alternatively, divisional exposure for the divisional exposure condition C-2 may be performed in the same way as the divisional exposure condition C-1 in the above embodiment, while the entire elementary area A₂₄₁ is exposed when the image data N is equal to 255.

0114The rate of exposed portion in a divisional exposure area is determined in response to the LSBD N, of the image data N in the above embodiment. The LSBD N₁ can be also expressed in the following equation, by employing the equality in the equation (7) or (8):<maths id="math0013" num=""><img file="EP0337427A2_D0015.tif" /></maths>

0115The third term " + 1" in the right hand side of the equation (16) appears because the screen pattern data P is formed so asto be suitable for judgement along the equations (7) and (8). In general, therefore, the equality in the equation (7) or (8) and the equation (16) are respectively provided as follows:<maths id="math0014" num=""><img file="EP0337427A2_D0016.tif" /></maths><maths id="math0015" num=""><img file="EP0337427A2_D0017.tif" /></maths>

0116In other words, the data N₁ for deciding the rate of divisional exposure are obtained by adding a constant value Co to a difference between the image data N and the screen pattern data P. When the equations (17) and (18) are employed, the equation (3) is replaced by the following equation (19):<maths id="math0016" num=""><img file="EP0337427A2_D0018.tif" /></maths>

0117Therefore, it can be said that, generally in divisional exposure, only a portion which is responsive to a difference between the image data N and the screen pattern data P of an elementary area is exposed.

0118The range of the image data N for which the equation (17) holds for a certain value of the screen pattern data P is provided as follows:<maths id="math0017" num=""><img file="EP0337427A2_D0019.tif" /></maths>where AP represents the number of gradation levels (AP = 16 in the above embodimet) between two adjacent screen pattern data P. In other words, as to judgement of the divisional exposure condition for each elementary area, the elementary area can be assumed to be divisionally exposed when the image data N are in a certain range, which is expressed by the equation (20) for example, predetermined on the basis of the screen pattern data P.

0119In the above embodiment, the halftone dot HD is assumed to be formed of 4 x 4 elementary areas. However, the present invention is also applicable to a halftone dot which is divided into 5 x 5, 8 x 8, 16 x 16 or the like. When one halftone dot HD is formed of 5 x 5 elementary areas, for example, each elementary area can be further divided by a dividing number of 11 for performing the divisional exposure. In this case, an expressible gradation number is 275 (=5x5x11). Accordingly, the image data N are converted to be expressible of 275 gradation levels. As can be seen from the above example, the number of elementary areas forming the halftone dot HD and dividing number of each elementary area can be arbitrary determined. In this case, the rate of exposed portion in divisional exposure is not necessarily decided on the basis of the LSBD N, of the image data N, but is determined in response to the difference between the image data N and the screen pattern data P, similarly to the above embodiment.

0120However, when the number of elementary areas forming the halftone dot HD and a dividing number of each elementary area are expressed as 2^m and 2" (m, n are integers), respectively, as is the case with the above embodiment, processing with a binary signal is facilitated.

0121When a preceding elementary area along a scanning line is exposed, the portion of the divisional exposure is formed to contact the preceding elementary area in the above embodiment. However the portion may be formed to be closer to the center of the halftone dot. The "center of the halftone dot" means, in Fig.1 or 3, a line along the subscanning (horizontal) direction located at the center of the main scanning (vertical) direction of the halftone dot HD, for example. Through the divisinal exposure, exposed portions of all elementary areas A₁ to A₂₄₁ are concentrated to the center of the halftone dot HD, which is accordingly formed with an excellent configuration.

0122According to the present invention, as described above, the elementary area to be divisionally exposed is designated on the basis of the image data and the screen pattern data while the elementary area is divisionally exposed in response to the difference between those data. As a result, reproducibility of gradation levels of a recording image can be substantially increased to the same extent as the image signal without reducing the ratio of the diameter of the optical exposure beam to the screen pitch and without increasing the capacity of the screen pattern memory. ,

0123Although the present invention has been described and illustrated in detail, it is clearly understood that the same is by way of illustration and example only and is not to be taken by way of limitation, the spirit and scope of the present invention being limited only by the terms of the appended claims.

0124The features disclosed in the foregoing description, in the claims and/or in the accompanying drawings may, both separately and in any combination thereof, be material for realising the invention in diverse forms thereof.