JPH05307307A - Color image forming device - Google Patents

Abstract

PURPOSE:To provide a compact and inexpensive color image forming device which is provided with an image producing device obtained by closely arranging plural developing devices and plural exposure devices around a single photosensitive body, which requires only a small-scale data delay circuit as a data delay circuit, which can form a clear color image at a high speed and which does not cause color slippage in a formed image by vibration. CONSTITUTION:On the inside of the single photosensitive drum 218 provided with a light transmissive supporting body, a cylindrical heat conductive body obtained by fitting four LED arrays which correspond to the colors of C, M, Y and Bk and expose the photosensitive body of the drum 218 on the outside circumferential surface is fitted to the rotary shaft of the drum 218 concentrically to the drum 218. On the outside of the drum 218, four developing devices 220 (C, M, Y and Bk) corresponding to the colors of C, M, Y and Bk are arranged.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a color image forming apparatus, and more particularly, to an electrophotographic system in which a plurality of specific color exposure devices and developing devices are provided around one cylindrical or endless web-shaped photosensitive member. The present invention relates to a color image forming apparatus.

[0002]

2. Description of the Related Art Several different systems are known for electrophotographic color image forming apparatuses for forming color images of two or more colors, such as color copying machines. One method is a method including an image forming station including the same number of photoreceptors, chargers, and developing devices as the number of colors required for image formation. This method is suitable for high-speed color image formation, but has a drawback that the apparatus is inevitably large in size because a plurality of photoconductors are arranged in parallel, a large-capacity delay circuit memory is required, and it is expensive. In the system disclosed in Japanese Patent Laid-Open No. 2-254868, only a single photoconductor is provided, and a color image is formed on the photoconductor by repeating the same number of image forming cycles as the color of the developer. Although this system can realize a small-sized and low-cost apparatus, it has drawbacks that the image size is limited to the surface size of the photoconductor and the image forming speed is slow. Furthermore, JP-A-58-2
As disclosed in JP-A No. 3058 and JP-A No. 61-51534, there is a system in which a single photoconductor is provided and latent images of a plurality of colors are formed on the photoconductor within one rotation. Although this method is capable of high-speed multicolor image formation, it is necessary to secure a space for the same number of exposures as the developing colors between the developing devices, and therefore data for synchronizing the exposure position of the subsequent exposure device. The amount of delay was large, and a large-diameter photoconductor drum was required in a full-color device. Further, when the exposure device is arranged close to the developing device in order not to increase the capacity of the delay circuit memory so much, there is a drawback that the exposure device arranged close to the developing device is easily soiled with toner. In any case, in a color copying machine having only a single photoconductor, the second and subsequent exposures are performed from the top of the toner image developed on the photoconductor by the first exposure. Since the irradiation light of the exposure is scattered or absorbed by the toner adhering to the photoconductor, it is not possible to form a clear latent image of the relevant color, and therefore the color image formed on the transfer paper is also the original color. There is an essential problem that it is different from.

[0003]

When the rotational vibration of the drive motor for rotating the photosensitive drum propagates to the photosensitive drum through the drive mechanism, the rotational speed of the photosensitive drum changes delicately. In a copying machine using an end face light emitting diode (LED), which is useful as a light emitting body expressing multiple gradations, as an electro-optical conversion element, the shape of an optical image formed by the LED is flat in the sub-scanning direction. Therefore, the unevenness of rotation of the photosensitive drum is more significantly affected, the exposure light images overlap, and it becomes difficult to form the latent image of the original recorded image according to the recording signal.

Further, in a copying machine in which a plurality of exposure devices corresponding to different colors are arranged around a single photoconductor, precise alignment of the exposure device is performed so that color deviation does not occur in a recorded image. However, there is a large difference in the amount of heat generated by the exposure equipment, the heat generated by the exposure equipment is not evenly dissipated, and air is generated by other exposure equipment due to the effects of air convection. When the heat is transferred, a difference occurs in the thermal expansion of the support member that supports the exposure apparatus, and the image formation positions of the exposure light emitted from the plurality of exposure apparatuses are displaced,
Color misregistration occurs in a recorded image.

In view of the above-mentioned drawbacks of the prior art, the first aspect of the present invention
The purpose of is to install a plurality of developing devices and a plurality of exposing devices in the vicinity of a single photoconductor, for example, even if it is a four-color full-color system, only a small-scale data delay circuit is required. To provide a small-sized and inexpensive color image forming apparatus capable of forming a clear color image faithful to the color of an original at a high speed and causing no color shift in a recorded image due to vibration of a motor or the like. It is in. Also,
A second object of the present invention is to provide a color image forming apparatus in which scanning exposure by a plurality of exposure devices is accurately performed even in various temperature environments, and a resist quality of a formed multicolor image is high.

[0006]

In order to achieve the above-mentioned first object, an interface means (212a) for receiving image data and a rotary shaft of a photoconductor are used to support the first means.
A supporting member in which the scanning exposure means and the second scanning exposure means are concentrically attached around the rotation axis, and a data signal of another color (M) in response to a data signal of a specific color (C) of image data. A delay means (212b) for delaying, and a photoreceptor (218) having a cylindrical or endless web shape and having a photosensitive layer on the outer surface of a light-transmissive support, and at least a specific color (C A) and a second exposure means (212M) that exerts an exposure action based on the signal and the delayed other specific color (M) signal, respectively, are inside the photoconductor (218), and Body (218)
The first scanning exposure means (212C) and the second scanning exposure means (212M) are arranged on a support member (212s) mounted concentrically around the rotation axis and are supported by the rotation shaft of the support body of Color developing means (220C,
M) was placed outside the photoreceptor (218). Also, the second
In order to achieve the purpose (1), a heat conductor (212s) is provided inside the photoconductor (218), and a plurality of scanning exposure means (212C, M, Y, K) are fixed to this.

[0007]

In the first means, the first exposure means exposes the image of the first color from the inside of the photoconductor, and the first phenomenon means arranged outside the photoconductor forms a visible image of the first color. , The second being accepted by the interface means and being delayed by the delay means
Based on the data signal of the color, the second exposure means performs image exposure of the second color from the inside of the photoconductor, and the second phenomenon means arranged outside the photoconductor forms a visible image of the second color. Therefore, the second exposure means is used for the second support of the photoreceptor support.
Since the exposure is performed from the side opposite to the image of the first color,
It is possible to form an image after the second color without being affected by the color (toner) image. Further, in the above-mentioned second means, the heat conductor conducts the heat generated by each of the plurality of scanning exposure means to another scanning exposure means different from the scanning exposure means, and the temperature of all the scanning exposure means. The image resist for each color is stably maintained because the image is made uniform.

[0008]

EXAMPLES The present invention will be described in detail with reference to the following examples. First, the drawings referred to in the description of the embodiments will be briefly described. FIG. 1 is a configuration diagram mainly showing a mechanical portion of a digital color copying machine suitable for carrying out the present invention, and FIG. 2 shows details of an image forming portion. FIG. 3 is a block diagram of image signal processing for showing the flow and processing of image signals of the copying machine, FIG. 4 is a block diagram of an intelligent image processing circuit, and FIG. 5 is image processing related to image formation by image area. FIG. 6 is a block diagram, FIG. 6 is a diagram for explaining a method of configuring recording pixels,
7 is a pixel formation timing diagram, FIG. 8 is an image processing block diagram of the image exposure unit, FIG. 9 is a diagram showing a dot growth direction of character pixels, FIG. 10 is a diagram showing a data format of character pixels,
11 is a diagram showing a specific example of a character pixel pattern, FIG. 12 is a diagram showing an example of dividing a character into pixels, FIG. 13 is a partially enlarged view of a reproduced character, and FIG. 14 is a diagram showing a different part of the reproduced character. FIG. 15 is an enlarged view, FIG. 15 is a diagram showing a gradation pixel data format, FIG. 16 is a diagram showing a specific example of a pixel pattern for each color of a gradation image, FIG. 17 is a diagram showing a specific example of a color gradation image for each color,
FIG. 18 is an enlarged view of a general color printing original, FIG.
20 is a block diagram of the image area separation circuit, FIG. 20 is a detailed block diagram of the original image area separation circuit, FIG. 21 is a partially enlarged view of FIG. 18, and FIG.
Is a diagram for explaining a green signal output from the scanner, FIG.
24 is a detailed block diagram of a bit map expansion circuit, FIG. 24 is a detailed block diagram of an image area separation interpolation circuit, FIG. 25 is an enlarged view showing a specific example of a binary bit map image, and FIG. 26 is a view showing recording pixels of different sizes. FIG. 27 is an explanatory diagram of image formation of recording pixels of different sizes, and FIG. 28 is a diagram illustrating low density character reproduction. The symbols used in the following description and their meanings are listed below. That is, R: red (red), G: green (green), B: blue (blue), C:
Cyan, M: Magenta, Y: Yellow, K: Black, LE
D: light emitting diode, LEDA: light emitting diode array, CCD: charge-coupled device or color imaging device,
SC: scanner or scanner module, PR: printer or printer module, SCON: system controller.

[Module Configuration] Referring to FIG.
The copying machine of this embodiment is roughly divided into two parts (called modules) in terms of its mechanism. One is a scanner module SC200 for reading an image, and the other is a printer module PR280 for forming an image. The rear end of the scanner module SC200 is rotatably connected to the upper rear end of the printer module PR280 by a hinge 200h. A torsion spring is incorporated in the hinge 200h, and the scanner module S
The C200 employs a free-stop structure that stays in any open position, and is prepared for smooth paper jam handling and maintenance work. [Image Processing Functional Block Configuration] With reference to FIG. 3, an outline of the image processing functional block that carries out the image processing function will be described. In FIG. 3, arrows between the image processing function blocks represent the flow of image signals or control signals. The image signal processing circuit is generally a basic image processing circuit 241 that processes an image signal read by the scanner SC200, an image memory 24.
Intelligent image processing circuit 242 for performing high-level image processing on the image data in 3, an external device connection connector 286N, an external interface (IF) circuit 286, an interface (IF) memory 287, and a bitmap (BM) expansion circuit 288. , A system controller SCON 285 to which the console device 250 is connected. Image recording data read by the scanner SC200 and processed by the basic image processing circuit 241 or input from an external device, processed to be recording data in a form suitable for the printer PR280, and output from the bitmap expansion circuit 288. The recorded image recording data is input to the printer PR280. Mechanically, the two image processing circuits 241, 241,
42 is inside the SC200, and the console device 250 is S
The external device connection system, that is, the external device connection connector 286N, the external IF circuit 286, I
F memory 287, BM expansion circuit 288 and SCON2
80 is built into the PR 280.

[Schematic Function of Each Image Processing Function Block and Flow of Image Signal] The system controller SCON285 has a function of integrally controlling the entire system of the copying machine, and each module constituting the copying machine, for example, SC2
00, PR 280, or intelligent image processing circuit 242, etc., a communication means SCONin, indicated by a thick arrow,
The system is controlled while sending and receiving control signals of commands and responses via SCONout. In addition,
When an optional module such as an automatic document feeder or a sorter is added, it has a function of controlling the optional modules as well. The console device 250 has a function as an input / output device for outputting a message of the copying machine to the operator or inputting various commands to the copying machine by the operator. Scanner module SC2
Reference numeral 00 has a color original image reading function, and outputs R (red), G (green), and B (blue) image signals through the original image reading circuit 207a to the basic image processing circuit 241 and the image memory 24.
Output to 3. The basic image processing circuit 241 uses the original image R,
Image processing is performed on the G and B image signals to obtain C (cyan), M
It has a function of converting (magenta), Y (yellow), and K (black) image recording signals, and outputs the converted C, M, Y, and K image signals to the printer module PR280 or the image memory 243. The printer module PR280 receives C, M, and
A permanent visible image is formed on the fed transfer paper based on the Y and K image signals. The external device connection system image processing function block (286N, 286, 287, 288) receives an image signal or a character code signal input from an external device outside the copying machine, and receives these signals of C, M, Y, and K. It has a function of converting into a recording signal and sending it to the PR 280. A mode in which an image is reproduced according to an image signal input from an external device in this manner is called a printer mode,
The mode in which the image of the original image read by the SC200 is reproduced is called the copy mode. The image memory 243 is the original image 1
R, G, B data for one sheet or C, M, for one copy
It has a function to store Y and K data. The intelligent image processing circuit 242 has a function of performing advanced image processing on the image data in the image memory 243. Magnetic disk unit (M
D) 240 has a function of storing image data and an intelligent image processing program. Note that, in FIG. 3, the intelligent image processing circuit 242 and the image memory 243 are illustrated in a slightly schematic form in order to facilitate understanding of a logical signal transfer relationship.

[Detailed Configuration of SC200] In FIG. 1 showing the detailed configuration of the scanner module SC200, 2
00c is a scanner control circuit, 202 is a platen glass,
208 is the first carriage, 209 is the second carriage, 203
Reference numerals a and b are original illumination lamps, 204a, b and c are first, second and third mirrors respectively, 205 is an imaging lens, 207
Is a color image pickup device (CCD), 207a is an original image reading circuit, 240 is a magnetic disk device (MD), 241
Is a basic image processing circuit, 242 is an intelligent image processing circuit, 24
3 is an image memory, 210 is a carriage scanning motor, 251
Is an operation panel including a transparent touch switch and a liquid crystal display device. [Original image reading operation] Scanner module SC20
In 0, the original image is sampled at a sampling density of 1/16 mm in both main scanning and sub-scanning, and R, G, and B three-color image data is obtained.
Each has a function of quantizing and reading in 256 tones. The illumination lamps 203a and 203b and the first mirror 204a are mounted on the first carriage 208, and the second mirror 204b and the third mirror 204c are fixed to the second carriage 209. When reading the original image, the first carriage is scan-driven by the sub-scanning speed Vsub and the second carriage is Vsub / 2 by the carriage scanning motor 210 while maintaining the optical conjugate relationship. The color image pickup device 207 has a structure in which an R image pickup section, a G image pickup section, and a B image pickup section, each of which has one-dimensionally arranged 4752 pixels covered with a red filter, a green filter, and a blue filter, are arranged in parallel in three columns. There is. Note that 202ar and 202a in FIG.
Reference characters g and 202ab show the positions of the R, G, and B image reading scanning lines on the platen surface 202a in a slightly exaggerated manner. Actually, the three scanning lines are almost close to each other. Lined up in parallel at 3/16 mm intervals. First,
The original is placed on the platen 202 with the copy surface facing down. The image forming lens 205 forms an original image on the light receiving surface of the CCD 207 in a reduced size. CCD 207 is an imaging lens 2
One line of main scanning of the original image projected by 05 is 16 for each color.
It is resolved at a density of pixels / mm, sampled and read. The CCD 207 that has read the original image reads R in pixel units of the original image,
An analog voltage corresponding to the reflected light of each of G and B is output, and is quantized by the original image reading circuit 207a into an 8-bit digital signal, that is, 256-tone digital data. The image data quantized for each color of R, G, and B in this way is the basic image processing circuit 241 and the image memory 2.
It is sent to 43. The original image reading circuit 207a
In order to prevent the problem of color misregistration caused by the R, G, B sampling positions of the color image pickup device 207 being different from each other by 3/16 mm, the delay circuit using the memory outputs the R and G signals to the B signal respectively. 203ab-
The sampling position shift is compensated by delaying by a time corresponding to the distance between 203ag and 203ab-203ar.

[Basic Image Processing Operation] The RGB image information of the original image read as described above is the basic image processing circuit 24.
Input to 1. The functions of the basic image processing circuit 241 are divided into two categories. The first category is a function for supporting the operation of the image signal, not the direct operation of the image signal. For example, the document size detection process, the color document / monochrome document identification process, the image area separation process for identifying and separating the character area and the gradation image area of the original image, and the like. In this category of processing, for example, all original image information on the platen 202 needs to be checked, such as document size detection. In this case, a so-called prescan is required before reading an original image for copy image formation. And Most of the processing results belonging to the first category are transmitted to the system controller SCON285. Upon receiving the processing result, the SCON 285 issues a control command to each module constituting the copying machine based on the processing result and advances the image forming process.
For example, if the basic image processing circuit 241 is “original is monochrome monochrome”
If it is determined that the result of this determination is sent to the SCON 285, the SCON 285 sends a command signal for energizing K development and stopping C, M, Y development to the printer control circuit 280c of the printer PR280. As a result, the printer control circuit 280c stops the multicolor development, and the K developing device 220K
Only the image is formed by energizing it. If the original is identified as a black-and-white original, data having a value of 0 is output except for the K signal. The second category is a process for directly manipulating an image signal, and includes, for example, image processing such as scaling, image movement, color correction, and gradation conversion. Further, these processings are classified into processing contents common to each image area such as scaling, and processing different in the image area such as gradation processing. The image processing contents of the second category include those automatically activated by the processing result of the first category, those designated and input by the operator from the console device 250, and those by a combination thereof. In any case, the R, G, B image signals input to the basic image processing circuit 241 are finally the recording signals C (cyan), M (magenta),
Converted to Y (yellow) and K (black), printer PR2
It is input to the recording interface circuit 212a which is the input unit of 80.

[Structure Related to Intelligent Image Processing] FIG. 4 is a diagram showing details of the intelligent image processing circuit 242 including the image memory 243. It should be noted that the reference numerals starting with 242 indicate that they are elements that constitute the intelligent image processing circuit 242. Also,
In the present specification, intelligent image processing refers to image processing for exhibiting advanced functions other than those conventionally provided in copying machines. In the figure, 242 CPU is a 32-bit microprocessor unit (hereinafter abbreviated as MPU),
242bus is a 32-bit bus, 242DMA is a direct memory access controller (hereinafter abbreviated as DMAC), 242SCONIF is an S for connecting to the communication means SCONout and SCONin of SCON285.
The CON image signal interface circuit and 242SCIF correspond to the interface circuit 243a in FIG.
The SC image signal interface circuit 242PRIF for connecting to the image signal output unit of C corresponds to the interface circuit 243b of FIG. 3, and is connected to the image signal input unit of the printer PR or a basic image processing circuit 241 equivalent thereto. PR image signal interface circuit for 242d
Reference numeral c is a disk controller, and 242m is a program memory for storing programs and the like. The program memory 242m is further divided into 242 ml, 2, 3, 4, 5 according to the purpose. The image memory 243 is actually an MPU
A bus 242bus of a 242 CPU is assigned to a part of a memory space capable of linear addressing,
Other program memory 2 for MPU242 CPU
It is on an equal footing with 42m. As described above, the IFs 243a and 243b of the image memory 243 are not directly connected to the image memory 243, but the image signal interface circuits 242SCIF and 242PRIF are connected to the bus 242bus, and the image memory 243 is also connected to the bus 242bus. Connected to. Between these, MPU24
Image data can be transferred by the data transfer command of the 2CPU, and further, the bus 242bus can be occupied by the DMAC242DMA to perform high-speed transfer which can follow the image reading speed and the recording speed. Figure 3
Then, in this situation, we considered that the two are connected, and made a simple display.

[Intelligent Image Processing Operation] Intelligent image processing circuit 2
When 42 performs intelligent image processing, it is not easy or rational to perform the intelligent image processing function only by the proximity processing, and therefore the original image information is read by SC200 and output from the original image reading circuit 207a. The stored image data is temporarily stored in the image memory 243, and the stored image data is subjected to image processing by the intelligent image processing circuit 242 to be converted into print data. This recording data is sent to the recording IF 212a of the PR 280, which is an input portion for image formation, and a visible image is formed on the recording medium by the PR 280. Therefore, when performing intelligent image processing on image data, there is a slight time delay between the reading of the original image and the output of the image recording signal. As described above, when performing the intelligent image processing operation, it is possible to perform the processing operation combined with the basic image processing operation. At this time, the image memory 2
The input data to 43 is obtained not from the output of the original image reading circuit 107a but from the output of the basic image processing circuit 141.

[Configuration of PR 280] The components of the printer module PR 280 will be described with reference to FIGS. 1 and 2. In these figures, 222a and 222b are a paper feed cassette and a paper feed tray 223 for storing transfer paper, respectively.
a and b are paper feed rolls for taking out the transfer paper, 224 is a pair of register rolls for timing of feeding the transfer paper, 218 is a photoconductor drum, 230 is a separation / conveyance belt for carrying the transfer paper separated from the photoconductor drum 218. 230c is a belt cleaner for cleaning the separation / conveyance belt 230, 236 is a fixing roll for heat-sealing the toner on the transfer paper, 237 is a fixing backup roll for pressing the fixing roll 236,
38b is a discharge roll for discharging the transfer paper to the outside of the machine, 238 is a discharge switching roll for switching the transfer paper to the discharge port and the double-sided feeding path side, 272 is a double-sided tray for storing the inverted transfer paper, and 273 is a double-sided tray 272. Double-sided feeding rolls 277a, 277b, and 275c for taking out the transfer paper are a pair of conveying rolls and 273a is a stacking roll. In addition, 219 (C, M,
Y, K) is a charging corotron that uniformly charges the surface of the photosensitive drum 218 prior to exposure for each C, M, Y, K, and 220 (C, M, Y, K) is each C, M, Y and K developing devices, 229 is a transfer corotron that charges the transfer paper to transfer the toner image, 221 is a cleaning device that cleans the photosensitive drum 218, and 221c is the photosensitive drum 21.
A static elimination corotron 221t for removing charges on the surface of 8 is a waste toner tank for collecting waste toner from the cleaning device. Further, 201 is a power switch for connecting and disconnecting the power source, 280c is a printer control circuit, 212b is a delay memory, 212dC, dM, dY and dK are recording control circuits for respective colors, and 212C, which is attached inside the photoconductor drum 218. M, Y, and K are light-emitting diode arrays (abbreviated as LEDA) of the respective colors, and 214C, M, Y, and K are converging optical transmitter arrays.

[Detailed Configuration of Image Forming Section] Referring to FIG.
The photoconductor drum 218 and the internal structure thereof will be described in detail. The inside of the photoconductor drum 218 is LEDA212C,
It is formed of a glass tube 218g having good transmissivity with respect to light having an emission wavelength of M, Y, K, for example, a wavelength of 720 nm. On the outer surface of the glass tube 218g, a transparent conductive layer and an organic photosensitive layer (OPC) are sequentially provided. The transparent conductive layer is grounded to 0 potential of the copying machine. An exposure module fixed to the main body of the copying machine is arranged inside the rotatable photoconductor drum 218. This module includes a heat conductor 212s, a heater 218h, a heat pipe 218p, a recording IF 212a, a delay memory 212b,
Recording control circuit 212 dC, dM, dY, dK, LEDA
212C, M, Y, K and converging optical transmitter array 21
4C, M, Y, K. LEDA212C, M,
Each of Y and K has a structure in which 14256 light emitting diodes are one-dimensionally arranged in a direction perpendicular to the plane of the drawing, and the light emitting point density is 48 dots / mm. The light emitting shape of each light emitting diode is a flat ellipse having a long array direction and a short orthogonal direction. Further, each LEDA 212 has a structure in which a plurality of divided semiconductor chips are arranged along the longitudinal direction on a ceramic substrate. When assembling the converging optical transmitter array 214, the LEDA2 is used.
12 emission points P1 and exposure points P2 on the photosensitive drum 218
The respective positions are adjusted so that and maintain the optical conjugate relationship. When the diameter of the photoconductor drum 218 is small, the difference in the optical refraction action between the axial direction and the circumferential direction becomes large. Good results can be obtained by using an anamorphic lens array or a roof mirror lens array in which the imaging action is different depending on the direction and the circumferential direction. further,
The anamorphic lens can also convert the light emission shape at the light emission point P1 of the LEDA 212 into a desired shape and form an image on the photosensitive drum 218. LEDA21
The forward voltage of the LED of No. 2 is about 1. When the current is 5 mA.
It is 7V, and the heat loss other than the light output that occurs when all LEDs are driven is considerably large. The heat generated from the LED may cause various problems such as destruction of the LED, change in light emission characteristics due to temperature rise of the diode junction portion, or deterioration in optical accuracy due to thermal distortion. For this reason, in this embodiment, the heat conductor 212s for closely supporting the ceramic substrate on which the LEDA 212 is mounted is made of a material having good heat conduction such as aluminum and has a large heat conduction cross section. . Furthermore, heat pipe 218p
1 is fixed to the central portion of the heat conductor 212s, and the other end is drawn out of the photosensitive drum 218 as shown in FIG.
A radiating fin is provided around the other end of the heat conductor 212s.
The heat conducted to the outside of the photosensitive drum 218 is released. In order to obtain high print speed, LE
As the power consumption of D increases, the power loss at the PN junction increases, so it is better to provide a fan near the heat radiation fins of the heat pipe 218p and perform forced air cooling.

On the other hand, when the temperature of the LEDA 212 is remarkably lowered, the light emitting performance of the LED is also lowered. The heater 218h embedded in the heat conductor 212s is provided to prevent this problem. The energization amount of the heater 218h is controlled in consideration of the detection temperature of a temperature sensor (not shown) attached to the heat conductor 212s and the driving load condition of the LEDA 212. Thus, the four LEDA 212C fixed to the heat conductor 212s,
M, Y and K are heat pipes 218p and heaters 218
The heat conductor 21 maintained at a desired temperature by the action of h
Since the temperature is controlled to reach a predetermined temperature range via 2s, the LEDA 212 is controlled regardless of the driving load condition.
The temperature difference of A212 is small, and the thermal expansion effect appears uniformly. Therefore, four LEDs A212C, M, Y, K
The accuracy of the irradiation light for each color can be kept good. Further, the inside of the photosensitive drum 218 is dust-proof sealed,
This sealed structure not only makes the temperature inside the photoconductor drum 218 uniform, but also helps prevent toner contamination of the exposure optical system. 29 and 30 are perspective views showing the photosensitive drum 218 and a part of its rotary drive system, respectively. As shown in FIG. 29, the heat conductor 212s has a cylindrical shape and is concentric with the photoconductor drum 218 and is integrated with the photoconductor drum 218.
Are supported and fixed to a rotary shaft (support shaft) supported via bearings. The drum ring gear 215 meshes with the drum drive gear 225, the drum drive pulley 226 fixed to the rotating shaft of the drum drive gear 225, and the drum drive belt 227 spanned by the drum drive pulley 226.
A driving force from a photoconductor drum drive motor (not shown) is transmitted via. The delay memory 212b electrically includes a recording IF 212a and a recording control circuit 212d (C, M,
Y, K), which is a circuit disposed between M, Y, and K of the four color image signals input to the recording IF 212a.
It is a circuit for delaying the Y and K signals with respect to the C signal by a predetermined time. The delay time is the time required to move the circumferential distance from the exposure point P2 of C on the photosensitive drum 218 to the exposure point P2 of M, Y, and K, respectively. Although the LEDs A212C, M, Y, and K are arranged at equal intervals in this embodiment, the LEDA 212C is replaced with the C developing roll 212Cm.
Just before, the LEDA212K was charged with the K charging corotron 219.
If they are moved and arranged immediately after K, the capacity of the delay memory 212b can be further reduced.

[Image Forming Operation of PR280] The printer module PR280 receives C, which is input to the recording IF 212a,
Pixel density 1 for both main scan and sub-scan for each color of M, Y, K
A full-color visible image consisting of a dot pattern with a recording dot density of 1/48 mm is formed on the transfer paper for each of C, M, Y, and K colors based on the recording data of 256 gradations of / 16 mm, and the copy output is made. To do. When the image forming operation is started, first, the photosensitive drum 218 is rotated counterclockwise by the drive motor 211. With the rotation, C latent image formation, C toner image formation, M latent image formation, M toner image formation, Y latent image formation, Y toner image formation, K latent image formation,
K toner image formation is performed, and finally, the toner images formed on the photosensitive drum 218 in the order of C, M, Y, and K are transferred onto the transfer paper. Image formation is performed first C
Imaging is performed as follows. The charging corotron 219C uniformly charges the surface of the photoconductor drum 218 to −700 V at a negative potential by corona discharge. After that, the LEDA 212C raster-exposes the surface of the photosensitive drum 218 based on the C recording signal. The C recording signal for image formation is supplied from the basic image processing circuit 241 in the general copy mode, and from the image memory 243 in the special copy mode including the intelligent image processing. The recording signal is input to the recording IF 112a, and the recording control circuit 212d
C controls the light emission of the LEDA 212C for each pixel unit based on the recording signal. For example, when the C recording signal is the highest density pixel, the LEDs corresponding to 3 in the main scanning and 3 in the sub scanning fully emit light, when the 0 density pixel does not emit light at all, and in the case of an intermediate density, light emission proportional to the density The number of diodes or the time is controlled to emit light. When raster exposure is performed in this manner, the charge on the exposed portion of the photoconductor drum 218, which was initially uniformly charged, disappears in proportion to the exposure light amount, and the electrostatic latent image corresponding to the distribution of the exposure light amount is lost. Is formed. The toner in the developing device 220C is negatively charged by the action of the doctor blade 212Cd and rotates in a non-contact manner with the photosensitive drum 218.
The developing roll 212Cm is biased by a voltage supply source (not shown) to a potential obtained by superimposing a negative DC potential and an AC potential on the metal conductive layer of the photosensitive drum 218. By applying the bias potential in this manner, the toner does not adhere to the portion of the photosensitive layer of the photosensitive drum 218 where the electric charge remains, and the C toner is adsorbed to the portion having no electric charge, that is, the exposed portion. , A C visible image corresponding to the C latent image is formed. Note that such a developing method may be called a reversal developing method.

The formation of the M image is also performed in the following manner, similarly to the formation of the C image. The charging corotron 219M uniformly charges the photosensitive drum 218 on which the C toner image is placed to −700 V by negative charge by corona discharge. L
The EDA 212M performs raster exposure based on the M recording signal. However, the M recording signal for image formation is inputted at the same time as the C recording signal at the time of being inputted to the recording interface circuit 212a, but when being inputted to the recording control circuit 212dC, the C exposure position is set at the delay memory 212b. M
It is input with a time delay corresponding to the distance of the exposure position. LEDA
Since the light emission of 212M is controlled based on the delayed M recording signal, the C image exposure position and the M image exposure position which are exposed by the recording signal at a single sample point of the original image exactly overlap. In this way, in the photosensitive layer portion of the photosensitive drum 218 that was initially uniformly charged, when the M raster exposure is performed, the charge proportional to the exposure light amount disappears and an electrostatic latent image of M color is formed. It The M toner in the developing device 220M is negatively charged, and the developing roller 2 of the developing device 220M is charged.
12 Mm is biased to the same potential as the C developing device 220C. Therefore, the toner does not adhere to the portion of the photosensitive layer of the photosensitive drum 218 where the charge remains, and the M toner flies and is attracted to the portion of the photosensitive drum 2 where the M exposure has been performed.
An M visible image corresponding to the electrostatic latent image on 18 is formed. Similarly, the Y image is on the C, M toner images on the photosensitive drum 218, the K image is on the C, M, Y images,
They are formed so as to overlap each other. Since the basic image processing circuit 241 performs UCR (undercolor removal) processing,
Since the recording signal components of the three C, M, and Y colors are replaced with the K recording signal, there is little chance that one pixel is developed with all the C, M, and Y color toners. In the conventional multi-developing color copying machine, since the second and subsequent exposures are performed on the toner images of other colors already formed,
Since the exposure light beam is scattered by the toner adhering on the photosensitive drum 218, the formed electrostatic latent image may be blurred, or the exposure light may be absorbed by the toner and the desired exposure amount may not be given. Although a problem was likely to occur anyway, in the present embodiment, the exposure light beam after the second exposure can reach the photosensitive layer portion of the photoconductor drum 218 without being blocked by anything, so that it is clear for all colors. An electrostatic latent image is formed. Strictly speaking, the surface potential of the electrostatic latent image of M has the dielectric effect of the C toner already developed and attached on the photoconductor drum 218. Therefore, the potential contrast of M exposure is the case of C exposure. Although a little different from the above, a generally good formed image can be obtained. That is, actually, the surface potential is about -80V when the M toner is exposed to the C toner, and the surface potential is about -150V when the M toner is not exposed to the C toner. The surface potential of the portion not exposed to M remains at −700 V because the surface potential is only dark decay regardless of the presence or absence of C toner. Further, FIG. 31 shows the photoconductor at the time of M exposure and Y exposure under the same conditions after the C toner image is formed, according to this embodiment and a conventional example in which a plurality of exposure devices are provided outside the photoconductor drum. It is a graph which showed the result of having measured the amount of exposure. (a) shows the measurement result at the time of M exposure, and (b) shows the measurement result at the time of Y exposure. As shown in the drawing, in this embodiment, the exposure light at the time of M exposure and Y exposure is not blocked by the C toner, so that C formed on the photoconductor is
It is possible to form a latent image that is hardly affected by the toner image.

On the other hand, when the image formation is started, the transfer paper is fed from the three feeding parts, that is, the paper feed cassette 222a, the paper feed tray 222b or the double-sided tray 272 to the feed rolls 223a, b and c. It is fed by the feeding action and is waiting at the nip of the register roll pair 224. The full-color toner image formed on the photoconductor drum 218 as described above is rotatably transferred to the transfer portion. When the front end of the toner image on the photoconductor drum 218 reaches the transfer separation corotron 229, the register roll pair 224 is driven so that the front end of the transfer paper coincides with the front end of the image, and the registration of the transfer paper and the image is performed. Is performed. When the transfer sheet is superposed on the toner image on the photosensitive drum 218 by the register roll pair 224 and passes under the transfer separation corotron 229 connected to the positive potential power source, the transfer sheet is charged to the positive potential by corona discharge, Most of the toner image on the photosensitive drum 218 is transferred onto the transfer paper. Subsequently, the separation / conveyance belt 2 connected to the separation power source
When passing under the sheet 30, a suction force stronger than the attraction force between the photoconductor drum 218 and the transfer paper acts between the separation conveyance belt 230 and the transfer paper, and the transfer paper is separated from the photoconductor drum 218 and separated. It is sucked and conveyed by the conveyor belt 230. The transfer paper on which the toner image is placed is sent to the fixing roll 236 by the separating / conveying belt 230. Heat and pressure are applied to the transfer paper held in the nip portion of the heated fixing roll 236 and backup roll 237, the toner is melted and bites into the fibers of the transfer paper to fix the toner image, and the copy image is obtained. The formation is complete. The completed copy is
After that, it is sent out of the main body by the discharge roll 238b and the switching roll 238. The ejected copy papers are piled down in the order of pages on a paper ejection tray (not shown). Also,
When the double-sided copy mode is selected, the switching roll 238 is moved to the position shown by the broken line to guide the transfer paper to the double-sided tray 272. The copied transfer paper is transported by the transport roll pair group 277 a, b, c, and the double-sided tray 272.
It is deposited with the copy side up. Photoconductor drum 218
A small amount of untransferred residual toner left on the photosensitive drum 2
The cleaning device 221 cleans 18 in preparation for reuse. At this time, the collected toner is collected by the collecting pipe 221p.
It is stored in the waste toner tank 221t via the. There is an opening (not shown) in the upper part of the double-sided tray 272, and the stacked transfer sheets can be easily taken out without opening the front panel of the copying machine to set the non-double-sided copy mode on the operation panel 251. It can be used as a paper discharge tray when it does.

[Image Forming Processing Operation by Image Area] According to FIG.
The processing of the image signal when forming an image for each image area will be described. In the present embodiment, the pattern forming processing operation of the recording pixels for the read original image is provided with a difference between the character image and the gradation image, and optimum image formation is realized for each. In addition,
A pixel pattern for a character image is called a character recording pixel pattern, and a pixel pattern for a gradation image is called a gradation recording pixel pattern. First, the outline of the processing will be described below. (1) In the copy mode, the image area separation circuit 241 converts the R, G, B image data obtained by reading the original image by the SC 200.
d is automatically separated into a character area and a gradation image area. PR28
At 0, based on the separation information from the image area separation circuit 241d, the recording data of the original image is formed by either the character recording pixel pattern or the gradation recording pixel pattern. (2) In the printer mode, when the image data sent from the host computer is in vector format such as page description language, the bitmap expansion circuit 288 directly
A pixel pattern having a character attribute and a pixel pattern having a gradation attribute are developed, and the developed recording data is received by the PR 280 to form an image. (3) In the printer mode, when the image data sent from the host computer is in a rough bit map format, a character recording pixel pattern obtained by interpolating the received bit map data by an image area separation interpolation circuit of a bit map expansion circuit 288 described later. And recording data composed of gradation recording pixel patterns are calculated, and the PR 280 forms an image based on the calculated recording data. Next, the processing flow of the image signal will be described. From the basic image processing circuit 241 in the copy mode, from the interface memory 287 of the external device connection system in the printer mode,
Recording data of 8 bits for each of C, M, Y, and K and a 1-bit selection signal SW for character / gradation discrimination are input to the recording IF 212a of the R280. The selection signal SW is a signal common to each color. The C recording data and the selection signal SW are directly input to the C recording control circuit 212dC. The C recording control circuit 212dC converts the C recording data into a character recording pixel pattern when the selection signal SW is 0, and converts it into a gradation recording pixel pattern when the selection signal SW is 1, and the photoconductor drum 21.
Image on 8. The M, Y, K recording data and the selection signal SW are respectively delayed by a delay memory 212b M, Y, K for a predetermined time, and then the M, Y, K recording control circuit 21 respectively.
Input to 21dM, Y, K, same as C recording data,
It is converted into a character recording pixel pattern or a gradation recording pixel pattern, and an image is formed on the photosensitive drum 218.

[Recording Pixel and Its Configuration] FIG.
FIG. 6 is a diagram illustrating standard type recording pixels when an image is formed by the image forming apparatus. In the figure, pixel represents a recording pixel, Xp is a pixel size in the main scanning direction, and Yp is a pixel size in the sub scanning direction. Further, the pixel pixel has three sub-pixels sub-p.
sub-pixel sub-pixel
The width of l is the width of the image formed when one LED of the LED A 212 emits light. light-spot is LEDA
This is the shape of a portion that exceeds a predetermined light energy when the converging light transmitting element array 214 forms an image of the bright spots 212 on the photoconductor drum 218. The shape is 1/48 mm in the main scanning direction and It is a flat ellipse with a width smaller than this. Note that the light energy distribution of the bright spots of the LEDA 212 itself is not flat, and the shape thereof is smaller than the image size in the main scanning direction and is smaller in the sub scanning direction due to separation between LEDs and other structural restrictions. Greater than size. When the LEDA 212 is lit for a time t while the photosensitive drum 218 is rotating at the peripheral speed Vpc, the photosensitive layer of the photosensitive drum 218 becomes 1 /
The exposure action is exerted in a rectangular shape of 48 mm × Vpc · t. Hereinafter, this shape is referred to as a dot dot. Sub-pixel sub-p
It is possible to select whether the pixels are separated by 1/256 mm in the sub-scanning direction and an image is formed in each of the divided parts, or a blank is formed. In addition, this divided "portion" can be continuously or intermittently imaged. Therefore, it can be said that the dot dot is an image formed in one isolated 1/256 mm section or two or more continuous sections. In other words, the minimum dot size in the sub-scanning direction is Yp / 256, and the maximum dot size is equal to Yp. There is a possibility that 0 to 128 dots dot can exist in one subpixel. However,
In the present embodiment, the number of recording control circuits 212dC, M, Y, and K is maximum 2 for simplification.

[Recording Pixel Size in Copy Mode] The size of the recording pixel in the copy mode is the pixel pixel of FIG.
Is the same as. The size is the main scanning direction Xp × the sub scanning direction Yp. In this embodiment, one recording pixel is 1 / side
It is a 16 mm square and is the same as the reading sampling density of the scanner SC200. [Printer Mode Recording Pixel Size] In the printer mode in which an image signal is received from the external device connection connector 286N and a visible image is recorded, there are three types of pixel sizes, and any one of them can be selected and designated. One of these is the recording pixel size selected when the pixel density of the image signal from the host computer is 16 pixels / mm and the pixel density is exactly the same as the image reading sampling density of the SC200. And is the same as the pixel size (standard type) in the copy mode. Pixels of the other two sizes are shown in FIG. (A) is a recording pixel selected when the pixel density of the image data sent from the host computer is as high as 24 pixels / mm. In the figure, the outer square (square) is one recording pixel pi.
In xel, both the pixel size Xp in the main scanning direction and the pixel size Yp in the sub scanning direction are 1/24 mm. Recording pixel pi
xel is composed of two sub-pixels sub-pixels 1 and 2, and the actual pixel formation is formed by exposing two LEDs of the LED A 212. In (b), the pixel density of the image data sent from the host computer is 12 pixels /
It is a recording pixel applied when the pixel density is relatively low as mm. The outermost quadrangle (square) is one recording pixel pixel, and the pixel size Xp in the main scanning direction and the pixel size Yp in the sub scanning direction are both 1/12 mm. The recording pixel pixel is composed of four sub-pixels sub-pixel1,
It consists of 2, 3 and 4 and the actual pixel formation is LEDA212
Are formed by exposing four LEDs of 27 is P
When the R280 receives the recording data and the binary number 10101111B, the image is formed when the image is formed by the recording pixels of different sizes. 9A shows the case where the recording pixel size is 1/24 mm, and FIG. 9B shows the case where the recording pixel size is 1/12 mm. On the other hand, the size is 1
An image in the case of forming an image with the same recording data 10101111B in a recording pixel of / 164 mm (standard type) is shown in FIG. 11 (c2). Comparing the three, it can be seen that the amount of jaggedness in the shaded area is the same for the three different size recording pixels. Therefore, high-quality character reproduction that is not dependent on the pixel density of the input record data can always be performed. This advantage is due to the low pixel density of the external IF286,
For example, connecting a high-definition signal output device to PR28
When 0 is set, it can be utilized when a recorded / reproduced image is formed.

[Image Forming Operation of Recording Pixel] FIG. 7 is a diagram showing on / off timings of dots forming a certain recording pixel and the recording pixel in the sub-scanning direction. As shown in the figure, for example, the third sub-pixel sub-pixel
3 is composed of two dots, and the first dot forms an image by turning on the LED of the LEDA 212 at time t1, off at t2, and turning on the second dot at time t3 and off at t4. FIG. 8 shows the internal circuit of the recording control circuit 212d and the LEDA.
It is the figure which showed a part of 212K. To form an image with the PR280, a raster recording method for recording a pixel row is used. Hereinafter, the driving operation of the LEDA 212 will be described with reference to FIG. LSYNC is a line scanning synchronization pulse generated at every 1 line scanning, and WCLK is one LSYNC.
The recording timing clock pulse is generated 256 times during the period, and 212p indicates the shape data (dot distribution configuration) of the recording pixels from the input 8-bit image data and the 1-bit character area / gradation area identification selection signal SW. A pixel generation circuit for outputting, 212ctr is a counter for counting WCLK,
212 dmux outputs the shape data of the recording pixel input to the input terminal in to the output destination of o0 or o for each LSYNC input.
Demultiplexer for switching to 1, 212sp is a toggle type serial / parallel conversion circuit of 8 bits × 57024 words × 2 columns, 212spmux n is the input terminal 0,
Out one of the shape data input to 1, 2, 3
The n-th selector that outputs from the terminal
spcmp n is the output value of the counter 212 ctr and the selector 212 spmux A comparator 212d that compares the output values of n and outputs a match pulse signal to the out terminal each time they match. n is the nth LED driver, 212
K n is the n-th LED of 14256. Record I
The C recording data sent from the F212a or the M, Y, K recording data sent from the delay memory 212b (M, Y, K) and the character area / gradation area identification selection signal SW are input to the pixel generating circuit 212p. .. The pixel generation circuit 212p generates dot pattern data of pixels, that is, blinking data of three LEDs from the recording data and the selection signal SW. That is, the light emission of the LED for one sub-pixel,
Four pieces of timing data for controlling extinction are generated. For example, in the case of the third sub-pixel sub-pixel 3 of FIG. 7, timing data of t1, t2, t3, t4 is generated. These timing data are for one scan line,
4752 pixels are generated. That is, 4752 (pixels)
X3 (LED) x2 (light emission, extinction pair) x2 (pair set) = 5
7024 pieces of timing data are calculated and sequentially sent out.

Toggle type serial / parallel conversion circuit 21
2sp has a capacity to store these data in two-stage registers side0 and side1 for two scanning lines, and while one stage register serially inputs the timing data, the other stage register records. It has a so-called toggle type configuration that outputs in parallel for driving.
Therefore, all the LEDs can be driven completely in parallel all the time, and high-speed image formation that makes full use of the light emission capability is possible. The timing data regarding the emission and extinction of the n-th LED is the register section 212sp. n, n
It is stored as 0, n1, n2, n3. Register division 212sp The timing data stored in each n is stored as 8-bit data, and the gradation value thereof ranges from 0 to 255. Selector 212 spmux n
Selectively outputs one of the light emission and extinction timing data n0, n1, n2, and n3 input to the input terminals 0, 1, 2, and 3 to the output terminal out. CLR
When the LSYNC signal is input to the terminal, it is initialized to No. 0 selection, and when a comparator match pulse described later is input to the CLK terminal, the selection number is incremented by one. Comparator 212 spcmp n is the output value of the counter 212 ctr and timing data n0, n1, n2, n for light emission and extinction.
The values of 3 are compared, and a pulse is output each time they match. LSY
Immediately after the NC signal is input, the comparison target timing data is n
It is 0. This out terminal is the selector 212 spmux
n CLK terminal. Therefore, each time the output value of the counter 212ctr and the values of the emission and extinction timing data n0, n1, n2, n3 match, the comparison target timing data sequentially changes from n0 to n1, n2, n3. Driver 212d When the pulse signal of LSYNC is input to the CLR terminal, n enters the extinction state. Input terminal in comparator 212 spcmp n
The LED212K emits light when a match signal is input from the LED and extinguishes when a match signal is input. Toggle n. Counter 212 ctr is LSYN
It is cleared by the pulse input of C and is incremented each time the pulse of WCLK is input. Therefore, the counting range is 0 to 255. However, 255 is counted at the same time as the pulse input of the next LSYNC, and the counter 212
Since the ctr is cleared when counting 255, the count is up to 254 substantially. Third sub-pixel sub of FIG.
Taking the example of -pixel3, n0 = t1, n1
= T2, n2 = t3, n3 = t4 timing data is generated by the pixel generation circuit 212p and set in the toggle type serial / parallel conversion circuit 212sp, and the pixel shape as shown in the figure is formed according to the above procedure. When there is no dot formation (blank), the value n0 = 255 is set in the serial / parallel conversion circuit 212sp. In this way, the counter value of the counter 212 ctr never reaches this value, and the comparator 212 spcm
p There is no generation of a match signal of n, and the LED does not emit light. When the sub-pixel is a single dot, n3 = 255
When all the sub-pixels are dots, n0 = 0, n1 = 25
Setting 5 is enough. The driving operation of one LED has been described above. The selector 212 spmux,
Each of the circuit elements of the comparator 212 spcmp and the driver 212 d is provided in the same number as the number of LEDs, and
The ED is driven similarly. Further, the circuit shown in FIG.
Common hardware is used for all colors of M, Y, and K. However, since the pixel generation circuit 212p has a different gradation pixel generation pattern, the signal output logic of this circuit is configured for each color. The signal output logic for character generation is common to all colors.

[Pixel Pattern] The dots in one recording pixel can be selected in various sizes, numbers and arrangements. 1
If a pattern formed by weaving dots and blanks in one recording pixel is referred to as a pixel pattern, the number of pixel patterns is extremely large. Note that it is also possible to classify pixel patterns having visually the same characteristics as one group. In the present embodiment, a pixel pattern having a property suitable for character reproduction is applied to the character area, and a pixel pattern suitable for gradation reproduction is applied to the gradation image area. Further, in the gradation image area, different pixel patterns are assigned to the C, M, Y, and K images to prevent the occurrence of color moire. [Character Pixel Pattern] The character pixel pattern is designed to reproduce clear lines so that isolated dots are not generated within and between subpixels. Further, in order to precisely reproduce a fine portion and reduce the jaggedness (jaggies) of the shaded portion, eight kinds of image directionality and 32 kinds of line thicknesses are selectably prepared. The eight arrows in FIG. 9 represent the directionality of the eight types of character pixel patterns. The origin of the arrow represents the first dot generation position, and the arrow represents the image growth direction. FIG. 10 shows a data format of character pixels. As shown in the figure, the character pixel pattern expression consists of 8-bit data, and the upper 3 bits (b7 to b5) are di.
The direction data represents the direction of the character pixel pattern, and the lower 5 bits (b4 to b0) represent the size of the character pixel pattern in the size data. Numerical values from 0 to 7 attached to the origin of the arrow in FIG. 9 represent the upper 3 bits, that is, the direction information in decimal notation. In addition, the thickness information is expressed such that the character pixel pattern becomes thicker as the numerical value increases, and 0 represents a blank, and 31 represents an image of all pixels regardless of directionality. Figure 11 is based on this rule.
It is the figure which drew one pixel data and the pixel pattern corresponding to the pixel data. 11 (a1), (a2), (a
In 2), since the value of the upper 3 bits is 0, it indicates that the dot grows from the upper side of the pixel. Since the lower 5 bits are represented by decimal numbers 5, 15 and 21, respectively, the thickness of the character pixel pattern increases at these ratios. Similarly, in FIGS. 11 (b1), (b2), and (b3), since the value of the upper 3 bits is 2, it can be seen that the type grows from the right side of the pixel. Since the values of the lower 5 bits are 5, 15 and 21, respectively, the dot width should be increased in parallel with the right side at these ratios, but the dot width is LE.
Since the length of the light emission pattern of D in the main scanning direction is fixed, there is a restriction that it cannot be changed continuously.
Within one sub-pixel, the method of increasing the dot length is adopted. In the actual image, the vertical line is 1/16 m
Experiments have shown that irregularities occur at m intervals, but they are not so noticeable visually. 11 (c1), (c
In 2) and (c2), since the value of the upper 3 bits is 5, it can be seen that the type grows from the lower left corner of the pixel. Since the values of the lower 5 bits are 5, 15 and 21, respectively, in decimal notation, the image size increases at these ratios. All of these nine types of character pixel patterns have the following properties.

(1) The ratio of the image area in a pixel to the total area of a pixel is determined by the value of the lower 5 bits of data, regardless of the shape of the character pixel pattern.

(2) When the shape of the image growth tip is macroscopically captured in the image growth process of each character pixel pattern, as shown in FIG.
As shown by the broken line in the middle, the tip shape of the character pixel pattern can be approximated by a line perpendicular to the growth direction.

[Character Reproduction by Character Pixel Pattern] FIG.
Is an enlarged view of the contour of the letter "K" having a height of 1 mm together with a 1/16 mm square. Characters of this size are the smallest characters used in small English-Japanese dictionaries and are suitable for evaluating copy reproduction in copy mode and print quality in printer mode. That is, it has been experimentally confirmed that a beautifully formed character image of this size is a necessary and sufficient condition for forming a character image. The mesh in FIG. 12 is a standard for pixel division when an image is formed with a recording pixel density of 1/16 mm. The upper numerical group represents the pixel number in the main scanning direction, and the left numerical group represents the pixel number in the sub scanning direction. By the way, the most difficult part in forming a character image using the LEDA 212 is an oblique demarcation line in the area indicated by area1 and a fine demarcation part at the end of the area indicated by area2. FIG. 13 is a diagram in which area 1 of FIG. 12, that is, a part of the hatched portion of the character “K” (main scanning pixel numbers 7 to 10, sub-scanning pixel numbers 3 and 4) is reproduced according to the character pixel pattern of this embodiment. is there. After all, although jaggies are generated, the degree of unevenness is greatly improved, and one step in a staircase shape is about 20 μm, and it is understood that it is improved to the extent that human eyes hardly recognize it. In the prior art, since an image is formed in pixel units, a jagged line that is three times as large as that in this figure occurs in a portion that should originally be a smooth diagonal line, which is visually noticeable. FIG. 14 is the are of FIG.
a2, that is, the upper left end of the character "K" (main scanning pixel number 1
4 to 4, sub-scanning pixel numbers 1 and 2) are reproduced in this embodiment, and it can be understood that a fine portion is approximately accurately reproduced.

[Reproduction of Low-Density Character] As described above,
The character pixel data of this embodiment is only the data of the direction and thickness of the character pixel pattern, and does not have density information. However, since there are many low-density character images in various manuscripts, there is concern that it may not be possible to reproduce low-density characters.
However, a survey of many manuscripts revealed the following.

(1) There are few original characters having a small size and low density as shown in FIG. Further, even in rare cases, it is easier to read by increasing the recording density and printing.

(2) Even the thinnest line segment forming a small size character having a height of about 1.5 mm has a width of about three isolated pixels. Therefore, in the present embodiment, when reproducing low density characters, an image is formed as follows.

(1) Low-density minute size characters are reproduced at a constant density regardless of the original image density.

(2) For low-density characters of a small size or more, the directionality of the character image contour matches or approximates the directionality of the original image, and the pixels of the non-contour portion (inside the line segment) are naturally in the original image. 1
Although the entire pixel is the density portion, the thickness of the density portion is reduced to form an image. FIG. 28 is a diagram for explaining the image reproduction method described in (2) above, where (a) is an example of a part of characters having an original image density of 0.4, and (b) is a reproduced image thereof according to the present embodiment. Is shown. If the toner having the same density as that of the original image is reproduced, toners having various densities must be prepared, which is not realistic at all. The reproduced image in FIG. 28B is reproduced with a toner having a toner density of 1.2, which is the same as a general solid development density. In the figure, the 9th main scanning pixel, the 4th subscanning pixel, and the 10th main scanning and 3rd subscanning pixel are non-contour pixels. For the above reason, these pixels have a partial blank area. By such an operation, it is possible to reproduce a character image having a low density and a small size or more while maintaining the sharpness of the outline with the naked eye.

[Gradation Pixel Pattern] In this copying machine, C, M,
It is controlled so as to generate different gradation pixel patterns for each of the Y and K color plates. A pixel pattern suitable for reproducing a copy image having a density gradation is called a gradation pixel pattern. Furthermore, each gradation pixel pattern is given a visually recognizable specific pattern, and when these pixel patterns are aggregated to form a color gradation image, the pattern of a single pixel is continuously formed into a linear stripe. Create a pattern (texture). Such a striped pattern is called a screen, and an angle formed by the screen with the main scanning direction is called a screen angle.
In this way, the screen angle differs for each color plate, and the screen pitch of each color plate can be made small.
Therefore, the spatial frequency of the moire fringes due to the interference of the screens of the respective color plates becomes extremely high, and the presence of the moire can be substantially ignored with the naked eye. When reproducing a color image, it is necessary to pay attention to the occurrence of color moire caused by a slight deviation in the registration of the C, M, Y, and K plates. Figure 15 shows the data format of grayscale pixels,
FIG. 16 is a diagram exemplifying the stage of image growth of grayscale pixels for each of C, M, Y, and K colors. In the character pixel pattern, the ratio of the image area occupied by pixels was set to 5 bits and 32 types.
For grayscale pixels, the density of 8-bit data densit
It is represented by y and is set to 256 types. This is set based on the fact that gradation pixels require finer density gradation. In FIG. 16, pixel patterns for each color are shown for an example of 6 kinds of common densities out of 256 kinds. (Ya), (Yb), (Yc),
(Yd), (Ye), (Yf) are Y pixel patterns, and (M
a), (Mb), (Mc), (Md), (Me), (M
f) is an M pixel pattern, (Ca), (Cb), (Cc),
(Cd), (Ce), and (Cf) are C pixel patterns, and (K
a), (Kb), (Kc), (Kd), (Ke), (K
In f), K pixel patterns are arranged in order of increasing density. The above-mentioned visually specific pattern attached to the gradation pixel pattern is an image that can be perceived as a "line pattern" having a specific directionality in any density pattern pixel pattern. The directionality of the “line pattern” of the Y pixel is always shown in FIG.
The screen angle indicated by the thick line angY in a) is in the direction of 90 degrees, and similarly, the directionality of the “line pattern” of the M pixels is always 0 degrees at the screen angle indicated by the thick line angM in FIG. The direction of the "line pattern" in Fig. 16 (C
The screen angle of 135 degrees indicated by the thick line angC in a),
The directionality of the "line pattern" of K pixels is always the direction of the screen angle of 45 degrees indicated by the thick line angK in FIG. 16 (Ka). The increase in concentration is always indicated by the thick lines angY and ang in FIG.
With M, angC, and angK as the center, the image dots around it are increased. In this way, the "line pattern" having the directionality of the thick lines angY, angM, angC, and angK can be perceived in any of the 256 kinds of gray scale pixel patterns illustrated in FIG. Concentration increasing method is thick line angY, angM, ang
The basic idea is to increase the image dots around C and angK while maintaining the symmetry, but in the process of increasing the density, the density increase position reaches the upper and lower limits of the sub-pixel to which the dot belongs and The maintained dot density growth may be hindered. For example, since the thick line angM is in contact with the upper side from the beginning, the dots of M gradation pixels cannot grow upward. In such a case, the density of dots is grown only downward so that the certain directionality of the "line pattern" is not impaired. Incidentally, the growth pattern of this M gradation pixel is the same as the growth pattern of the character pixel in the direction 0. However, the growth step of the main gradation pixel pattern is in increments of 1/256, and the growth step is finer than the growth step in increments of 1/32 of the character pixel pattern. Further, in the case of the C pixel or the K pixel, when the dot to be density-grown reaches one of the upper and lower ends of the sub-pixel, it is regrown from the side opposite to the reached side. In any case, it is necessary to secure the directionality of the "line pattern" and the image dot area proportional to the 8-bit (256 gradation) input data.
In this embodiment, in order to reduce the recording data size, the gradation pixel data of each color depends only on the C, M, Y and K color data, and the recording control circuit 212d (C, M, Y,
K) is fixedly generated in each hardware.

[Screen and Screen Angle] FIG. 1
7 is a diagram exemplifying a state where a single pixel pattern is continuously formed into a linear striped pattern when a color gradation image is formed by collecting gradation pixels, and (Ys) is mainly Y gradation pixels. And an image in which 3 × 3 pieces are arranged in the sub-scanning direction, similarly, (M
s), (Cs), and (Ks) represent images in which 3 × 3 M, C, and K pixels are arranged in the main and sub-scanning directions, respectively. These images show the case where the density gradation of all pixels is 149/256. As is clear from the figure, when the gradation image is actually formed by collecting the gradation pixels, the screen can be more clearly perceived than the "line pattern" of the single pixel. Although this figure shows an image portion where the density is uniform, it is clear that the same effect is produced even in a general gradation image in which the density continuously changes. As described above, the screen angle of each color is Y =
90 degrees, M = 0 degrees, C = 135 degrees, and K = 45 degrees are set. By the way, the reason why the gradation image is provided with a screen is to prevent the problem of color moire as described above, but in reality, a great effect is exhibited even when toner is developed by the developing device 220 (C, M, Y, K). To do. In dry toner development, the toner particle size is several microns, so if a large number of isolated and adjacent dot latent images are developed with toner of such a particle size, the toner is not developed as the latent image, and the photoconductor drum 218 is not developed. Phenomena that appear in series above often occur.
Furthermore, when the toner is pressed and melted by the fixing roll 236, it is crushed and this phenomenon becomes more remarkable. As a result, the linearity of the recorded image density with respect to the input image data may be lost, and the actual density gradation may be extremely poor. In this embodiment, as is clear from FIG. 17, there are no isolated dots even at an extremely low density, and the dots of the latent image formed on the photosensitive drum 218 are continuous. Therefore, there is less possibility of the occurrence of defects such as the above-mentioned non-linearity of the recorded image density, and the effective gradation number of the recorded image of the transfer paper on which the image is formed can be maintained high.

[Summary of Original Image Area Separation] When performing copy reproduction of an original image, it is first necessary to separate image information contained in the original image into a character / line image area and a gradation image area.
The original image area separation application processing is to separate the original image area into a character or line image area where a visible image is to be formed and a gradation image area such as a photograph or a halftone dot image. Image recording is performed with suitable color processing and gradation reproduction processing, and it is possible to reproduce high-quality copy images. By the way, as described above, referring to FIG. 5, the basic image processing circuit 241 controls the image area separation circuit 241d as one of the image processing functions of the first category for supporting the image operation. A character area processing circuit 241c and a gradation area processing circuit 241p are provided as a second category that is directly operated. The character area processing circuit 241c performs processing suitable for reproducing a sharp image with sharp edges and fine line segments of characters and line images, and outputs 5-bit C, M, Y, K density data. The gradation range processing circuit 241p performs processing suitable for gradation image reproduction, such as smooth gradation reproduction and faithful color reproduction, and outputs 8-bit C, M, Y, K density data. These two types of processing are performed in parallel on the input R, G, and B images, and only one of the processing results is selectively sent to the PR 280 by the selector sel 241. When the selection signal SW is a signal representing "character", 3-bit pixel direction data is added to the upper portion of the 5-bit C, M, Y, K density data, and the original density data is added. Different modifications are applied to the contour part and the non-contour part, and the result is output. The selection signal SW for switching the output to the selector sel 241 and the character pixel direction data are sent from the image area separation circuit 241d.

[Arrangement of Original Image Area Separation Circuit] FIG. 19 shows the arrangement of the image area separation circuit 241d. In the figure, 241d2rl to 241d2r6 are 36 bits ×
A 4752-word shift register performs first-in first-out (FIFO) for 6 main scanning lines of the original image data of 8 bits each for R, G, and B and the image area separation intermediate data. 241d2 is S
Separated intermediate data, character / gradation selection signal SW, and direction component of character pixel from the latest R, G, B data for one scanning line from C200 and the R, G, B data up to the preceding 6 scanning lines. It is an original image area separation circuit for extracting data.
FIG. 20 is a diagram for explaining the internal configuration of the original image area separation circuit 241d2 and its relation. In the figure, FP is a frequency analysis circuit for calculating the spatial frequency characteristics in the main, sub-scanning direction and diagonal directions included in the original image, and PM is a pattern for extracting the directions of the ink halftone dots and blank points and line segment contours of the halftone dot original portion. Matching circuit, CA is a color analysis circuit for checking the distribution of color components, T
A is a discriminating circuit for discriminating whether the original image is a character image region or a gradation image region, SM is a direction determining and correcting circuit for a character portion, EM is a correcting circuit for a character end portion, and WG is a character pixel thickness information generating circuit.

[Operation of Original Image Area Separation Circuit] FIG. 18 is an enlarged view showing an example of a general print document. The mesh is divided at 1/16 mm intervals, that is, in SC200 sampling units. Indicates that img1 is the outer shape of black character “K” with a height of 1 mm, and img2 is C, M,
The C image in the Y, K halftone image is shown. Further, p is a target pixel for separation determination, and ml is a range of neighboring pixels to be referred to when performing pattern matching with PM. In FIG. 20, a range ml representing a part of the shift register group 241d2r indicates that the reference pixel data is extracted from here. Similarly, m
Reference numeral 2 indicates the range referred to by FP and CA. F
The frequency data, pattern matching data, and color distribution data of each circuit of P, PM, and CA are input to TA, and are comprehensively discriminated by TA to be a character image area or a gradation image area and output as a selection signal SW. For the pixels determined to be the character image area, the directionality of the character pixel pattern is further determined by SM, and the density data is modified so that the line segments are smoothly connected from the continuity with the neighboring character pixels.

In EM, so that the end of the line segment is proper,
Data on the direction and thickness of the line segment is created. The WG corrects the thickness information of the line segment and adjusts the thickness information of the line segment outline and the middle portion according to the image density. Specifically, the thickness information of the contour pixels of low-density characters other than the highest density is increased so as to match the contour of the original image, and the middle portion is decreased so as to match the density of the original image. As described above, the original image area separation circuit 2
The data of the character / gradation selection signal SW, the direction of the character pixel pattern, and the thickness information output from the reference numeral 41d2 is the selector s.
The image data is input to the e241, is discriminated, and is sent to the PR 280 as modified image data. FIG. 21 is an enlarged view of a part of FIG. FIG. 22 is a diagram schematically showing the G data output by the original image reading circuit 207a of the SC200, and the area of pure black corresponds to the output value occupying the maximum density gradation 256. Referring to FIG. 22, for example, paying attention to the diagonal line portion of the character “K”, since the original image reading circuit 207a simply quantizes and outputs the reflection density, the output data includes only the pixel density size information. And the direction information is completely lost. In the conventional copying machine, when the image data read by the original image reading circuit is reproduced by the printer, the image is reproduced by the same dot pixel having no directionality. In the present embodiment, the direction determination correction circuit SM estimates and restores the direction information for each character pixel, and finally sends this to the PR 280 in the data format shown in FIG. 10, and the PR 280 forms an image based on this pixel data. Therefore, an image with a sharp contour can be reproduced.

[Summary of Image Interpolation Processing] The printer mode adaptation processing is bit map data in which print information data sent from the host computer is coarse, typically binary image data in which pixels are only black pixels / white pixels. In this case, the image area of the image data is separated into a character / line image area and a gradation image area such as a photograph or halftone dot image, and further, the slanted area of the character or line drawing is smoothed and the halftone dot image of the gradation image area is finely divided. By performing data interpolation processing such as digitization, it is possible to reproduce a high-quality original image. FIG. 25 is a partially enlarged view of an example of the binary bitmap image. The mesh has a spacing of 1/24 mm, that is, the intervals corresponding to "medium" of the three pixel sizes in the printer mode. The pixel density of this received data is 12 dots / mm that matches the coarse pixels shown in FIG.
Is. The right side of FIG. 25 is a part of the diagonal line of the character “K”, but if this received data is sent to the PR 280 as it is and printed, jaggies with 1/12 mm intervals appear in the outline of the character image, and the quality of the character image is improved. There is a risk of damage. Therefore, interpolation data obtained from the original information is interpolated in the received data to create a group of multi-valued character pixel data, and a clear character image as shown in FIG. 13 is restored and reproduced based on the created pixel data. . At that time, it is necessary to separate the received bitmap image into a character / line image area and a gradation image area.

[Structure of Bitmap Developing Circuit] FIG. 23 shows an internal circuit of the bit map developing circuit 288. In the figure, 287d2r1 to 287d2r6
Is a 1-bit x 7128-word shift register,
The original image data of a single color is first-in first-out (FI
FO). 7128 words are provided so that the original image data can be used even when the pixel density is 1/24 mm. The image area separating / interpolating circuit 287d2 separates from the latest bitmap raster data for one scanning line input from the external interface circuit 286, the raster data and the input pixel density dens up to six scanning lines before, and the color information color. The selection signal SW and the multi-valued character pixel data and the multi-valued gradation pixel data generated by interpolation are output. FIG. 24 is a diagram illustrating an internal circuit of the image area separation interpolation circuit 287d2 and a related circuit. In the figure, DC is a halftone dot interpolating circuit, SEL is a selector, and the circuits indicated by other symbols are the same as those described for the internal circuit of the image area separation circuit 241d, and therefore description thereof is omitted.

[Operation of Image Interpolation Circuit] In FIG. 24, m3 is PM
The range of neighboring pixels to be referred to at the time of pattern matching is shown, and the pixel data surrounding the reference pixel p of the shift register group 287d2r is extracted from here. Similarly, m4 indicates the range referred to by FP and CA. The frequency data, pattern matching data, and color distribution data output from the FP, PM, and CA circuits are input to the TA, and the TA comprehensively discriminates between the character image area and the gradation image area, and the selector SW is used as a selection signal SW. It is output to the SEL and recording interface circuit 212a. The directionality of the character pixel pattern is further determined by the SM for the pixels determined to be the character image area, and the direction component is 3 bits and the thickness component is 5 bits so that the line segments are smoothly connected from the continuity with neighboring character pixels. Multi-valued character pixel data consisting of In addition, the EM creates data on the direction and thickness so that the ends of the line segments are appropriate. The WG corrects the thickness information of the line segment, adjusts the thickness information of the middle portion of the line segment according to the image density, and outputs it to the first input terminal of the selector SEL. The halftone dot interpolation circuit DC creates multi-value gradation pixel data having 8-bit smooth density information from the isolated original pixel binary data and color information color, and outputs it to the second input terminal of the selector SEL. When the output of the discrimination circuit TA is 0, the selector SEL selects the first input data, that is, the character pixel information, and when the output of the determination circuit TA is 1, selects the second input data, that is, the gradation pixel information, and sends it to the PR 280. At the same time, the selection signal SW is also PR
Sent to 280.

[0044]

As described above, according to the invention described in claim 1, the photosensitive layer is provided on the outer surface of the light transmissive support, and at least the first exposure means and the second exposure means are circulated. Arranged inside a single photoconductor and concentrically supported and fixed to the rotary shaft of the photoconductor, the specific color data of the image data received by the interface means and the delay means delayed this image data. Based on the other specific color data, each of the exposing means scans and exposes the photosensitive layer, and the developing means of the first color and the developing means of the second color arranged outside the photoconductor develop the color image. Since the exposure means is housed inside the photoconductor, the apparatus can be downsized even with the four-color full-color system, and the recorded image will not be misregistered due to vibration. , Was made possible only with high speed sharp color image forming comprising a small data delay means.
According to the invention described in claim 2, since the plurality of scanning exposure means are fixed to the heat conductor so that the temperatures of the plurality of scanning exposure means become substantially uniform, the thermal expansion effect of the plurality of scanning exposure means. Appear almost the same, so that the scanning exposures of a plurality of colors are accurately matched and the resist quality of the multicolor image can be improved. Further, according to the third aspect of the present invention, since the first exposure means and the second exposure means are composed of the edge emitting diodes, it is possible to form a multi-tone recorded image with high accuracy.

[Brief description of drawings]

FIG. 1 is a configuration diagram mainly showing a mechanical portion of a digital copying machine according to an embodiment of the present invention.

FIG. 2 is a configuration diagram showing in detail an image forming unit of a digital copying machine.

FIG. 3 is an image signal processing block diagram of the copying machine.

FIG. 4 is a configuration diagram of an intelligent image processing circuit of a copying machine.

FIG. 5 is an image processing block diagram relating to image formation by image area of a copying machine.

FIG. 6 is an explanatory diagram illustrating a method of configuring recording pixels.

FIG. 7 is a pixel formation timing chart of the copying machine.

FIG. 8 is an image processing block diagram of an image exposure unit of a copying machine.

FIG. 9 is an explanatory diagram showing dot growth directions of character pixels.

FIG. 10 is an explanatory diagram showing a data format of character pixels.

FIG. 11 is an explanatory diagram showing a specific example of a character pixel pattern.

FIG. 12 is an explanatory diagram showing an example of dividing a character into pixels.

FIG. 13 is a partially enlarged view of a reproduced character.

FIG. 14 is a partially enlarged view of characters in a portion different from FIG.

FIG. 15 is an explanatory diagram showing a gradation pixel data format.

FIG. 16 is an explanatory diagram showing an example of a pixel pattern for each color of a gradation image.

FIG. 17 is an explanatory diagram showing an example of a color gradation image for each color.

FIG. 18 is an enlarged view of a general color print original.

FIG. 19 is a configuration diagram of an image area separation circuit of a copying machine.

FIG. 20 is a detailed block diagram of an original image area separation circuit of a copying machine.

FIG. 21 is a partially enlarged view of FIG. 18.

FIG. 22 is an explanatory diagram of a green signal output from the scanner.

FIG. 23 is a detailed block diagram of a bitmap expansion circuit of the copying machine.

FIG. 24 is a detailed block diagram of an image area separation interpolation circuit of a copying machine.

FIG. 25 is an enlarged explanatory diagram showing a specific example of a binary bitmap image of a character.

FIG. 26 is an explanatory diagram showing recording pixels of different sizes.

FIG. 27 is an explanatory diagram of image formation of recording pixels of different sizes.

FIG. 28 is an explanatory diagram illustrating low density character reproduction.

FIG. 29 is a perspective view of a part of the rotation drive system of the photosensitive drum.

FIG. 30 is a perspective view of a photosensitive drum.

FIG. 31 is a graph showing the amount of exposure of the photoconductor during M exposure and Y exposure according to the example of the present invention and the conventional example.

[Explanation of symbols]

200 Scanner Module 207 Color Imaging Device 207a Original Image Reading Circuit 212 (C, M, Y, K) Light Emitting Diode Array (LE
DA) 212a Recording interface circuit 212b Delay memory 212d (C, M, Y, K) Recording control circuit 212s Heat conductor 214 Convergence optical transmitter array 218 Photoconductor drum 218p Heat pipe 218h Heater 219 (C, M, Y, K) K) Charging scorotron 220 (C, M, Y, K) Developing device 229 Transfer corotron 241 Basic image processing circuit 241d Image area separation circuit 242 Intelligent image processing circuit 243 Image memory 280 Printer module 285 System controller 286 External interface circuit 287 interface Memory 287d2 Image area separation interpolation circuit 288 Bit map expansion circuit

─────────────────────────────────────────────────── ─── Continuation of the front page (51) Int.Cl. ⁵ Identification number Internal reference number FI Technical display location G03G 15/04 116 9122-2H 21/00 118 H04N 1/29 G 9186-5C

Claims (3)

[Claims] 1. A cylindrical or endless web having a photosensitive layer on the outer surface thereof, and one photosensitive member that moves in a circular manner, and at least the photosensitive layer of the photosensitive member has a specific color signal and other First scanning exposure means and second scanning exposure means that exert an exposure action based on the specific color signal, and a plurality of the specifics that respectively develop the latent image formed on the photoconductor by the exposure of each scanning exposure means. In a color image forming apparatus including a color developing unit, an interface unit that receives image data and a rotation shaft of the photoconductor are supported, and the first scanning exposure unit and the second scanning exposure unit rotate around the rotation shaft. A support member concentrically attached to the and a delay means for delaying the data signal of the specific color of the received image data with respect to the data signal of the specific color. The photosensitive member is provided with the photosensitive layer on the outer side of a light-transmissive support, the first scanning exposure unit and the second scanning exposure unit are arranged on the inner side of the photosensitive member, and the plurality of developing units are A color image forming apparatus, wherein the color image forming apparatus is arranged outside the photoconductor. 2. A heat conductor for fixing and supporting a plurality of scanning exposure means on the inside of the photoconductor and for maintaining the temperatures of the plurality of scanning exposure means at substantially the same temperature. 2. The color image forming apparatus according to item 1 above. 3. The color image forming apparatus according to claim 1, wherein the first scanning exposure means and the second scanning exposure means are composed of end face light emitting diodes.