JPS62282953A - Thermal recording apparatus - Google Patents

Abstract

PURPOSE:To always obtain a recording image having stable density, by mounting a means estimating the heat accumulation state of a thermal head and a means determining an injection energy level to a recording picture dot on reference to the heat accumulation state and the density data of a picture element. CONSTITUTION:A digital signal having the density information of each picture element is inputted to a heat accumulation state operation part 2 to calculate the heat accumulation state at the time of the recording of a pixel and the calculated data is outputted to a multivalue pattern forming part 3 along with the original density information. A fixed pattern and the injection energy level to the mark dot in the pattern are determined corresponding to the density information of the pixel to be recorded and heat accumulation state data outputted from the heat accumulation state operation part 2. Then, the multivalue data of each dot determined is outputted to a pulse width forming part 4 and a pulse width is determined to be converted to the signal fitted to a thermal head driving part 5 and predetermined energy is injected in each dot. Therefore, the quantity of injection is determined on reference to circumferential image data with due regard to an estimated heat accumulation state and a recording image having stable density can be always obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Field of Industrial Application) The present invention relates to a thermal recording device capable of stabilizing recording density.

(Prior art) In the melt-type thermal transfer recording method, in which heat-melting ink is heated by a thermal head to melt or soften the ink and transfer it to recording paper, the density expression by ink transfer is "1".
Because of this, gradation expression by area modulation, so-called binary area gradation, is generally performed.

However, in this method, when one pixel is represented by a matrix of n×n pixels, the number of gradations that can be expressed is only n2+1, including the background density level of the recording paper. Therefore, for example, only 17 gradations can be expressed with a matrix size of 4×4. However, in order to obtain an image that is natural to the eye, the resolution of color pixels is generally 4 dots/a.
It is said that I11 or more and the number of gradations is 64 or more.

In order to satisfy this requirement using the binary area modulation method, it is necessary to use a thermal head with an 8×8 matrix size and a heating element density of 32 dots/mm or more. However, the currently available thermal head is 16 dots/IQr.
A is the limit, and it is currently difficult to realize thermal transfer recording with a resolution of 32 dots/city.

For this reason, it has been extremely difficult to obtain smooth density gradation using binary area gradation.

Therefore, in order to solve this problem, we used two inks with different densities.
A method of obtaining multiple gradations by using a laminated ink ribbon and combining dots with different densities and dot arrangement (Japanese Unexamined Patent Publication No. 1983-1999)
193377) and density (7) A method of sequentially overlapping recording dots from low-density ink to high-density ink using an ink ribbon coated with different inks sequentially in the longitudinal direction (Japanese Patent Application Laid-open No. 59-55768). etc. have been proposed.

However, all of these methods have drawbacks such as unstable ink transfer and long printing time.

On the other hand, there are dots made up of multiple dots.

A fixed pattern is set in the risk, and the fixed pattern is configured.

The present applicant has proposed a multi-gradation recording method in which the formation energy of a pixel is multivalued and pseudo gradations are obtained by using a plurality of types of fixed patterns with different density levels (Japanese Patent Application No. No. 60-16768). This method is
Even with a small matrix size, it is possible to obtain more gradations than before, and the gradation characteristics are extremely smooth. Furthermore, since the same number of gradations as before can be expressed with a smaller matrix size than before, it has the advantage of increasing the number of gradations and simultaneously increasing the resolution.

However, conventional thermal recording devices have a unique problem in that the recorded images become non-uniform due to the heat storage effect of the heating resistor that constitutes the thermal head. In other words, if the period of the energization cycle is shortened in order to improve the recording speed, when the heating resistor that was energized in the previous energization cycle is energized again, the heat generated in the previous energization cycle will not be sufficiently dissipated. As a result, the temperature of the heating resistor continues to rise. Therefore, when the energization cycle is repeated at such short intervals, the current temperature of each heating resistor differs depending on the past history of energization. If the heating resistors are energized at the same time in such a state, for example in the case of a thermal transfer recording device, the area where the ink on the ink film is softened will be different, and the area of the ink transferred onto the recording paper will also be different. In the case of thermal recording, the color density of the thermal paper changes.

For this reason, there is a problem that the image density of the recorded image becomes non-uniform. In particular, in thermal transfer color printers, color images are obtained by polymerizing translucent ink, so that the hue changes as the density of the image points changes. Further, when characters or the like are recorded, ink is transferred to places where there is actually no image data and should not be transferred, such as small gaps, for example, resulting in problems such as the characters being crushed.

In particular, in the conventional binary method, pixels of a fixed size are recorded in a matrix from one dot to all dots, so the smaller the number of white dots, the more the white areas are affected by the heat accumulation effect, such as collapse. becomes large, causing density reversal and the like, resulting in significant deterioration of gradation characteristics.

In order to solve this problem, for example, a patent application filed in 1986-13
As shown in No. 5934, the energy injected into the thermal head is changed according to the pixel data (binary data of whether or not the pixel is recorded) in the reference area around the pixel of interest. Several methods have been proposed.

On the other hand, the method using the fixed pattern described above is intended to actively utilize such a heat storage effect. In other words, the ink transfer area is gradually changed by making use of multivalued energy for forming each pixel within a fixed pattern, and by utilizing temporal and spatial thermal interference between each pixel to create smooth and fine gradations. This is a method of obtaining characteristics. Therefore, the maximum density (solid density) can be expressed even with a fixed pattern composed of about half the total number of dots in the matrix. For these reasons, the influence of heat storage is smaller compared to the conventional binary method. Furthermore, for one fixed pattern, there is a monotonically increasing relationship between the amount of implanted energy and the recording density, and by combining an appropriate fixed pattern and distributing the implanted energy, smooth gradation characteristics can be achieved even at the switching point of the fixed pattern. can be retained. Therefore,
Locally, there is an advantage that the relative gradation characteristics are approximately maintained. However, the absolute concentration characteristics change.

This is because, in principle, this method utilizes thermal interference only between pixels within one matrix, but in reality, it is possible to avoid thermal interference from pixels surrounding the pixel to be recorded. This is because it is extremely difficult.

In order to reduce heat accumulation in such a recording method, it is necessary to simply write "1" and "0" independently for each pixel as described above.
It is clear that the heat storage control method used in binary recording is insufficient.

(Problem to be Solved by the Invention) Heat storage control in the conventional binary recording method is relatively simple because the recording situation around the pixel of interest can be expressed using binary data of "1 degree" and "θ". Met.

However, the recording method targeted by the present invention sets a fixed recording pattern consisting of a plurality of pixels, treats this as one pixel, multi-values the pixel formation energy within that pixel, and thermally Because interference is effectively used, control methods that use data for each individual pixel, such as conventional thermal control methods, cannot cope with this method. SUMMARY OF THE INVENTION Therefore, an object of the present invention is to provide a thermal recording apparatus that can obtain an image with stable recording density from the start of recording to the end of recording even in the above-mentioned multi-value fixed pattern method.

[Structure of the Invention] (Means for Solving the Problems) The present invention provides a means for predicting the heat storage state of a thermal head by referring to image data around a pixel to be recorded, and a method for predicting the heat storage state and the density of a thorn pixel. The invention is characterized in that it includes means for determining the level of energy to be injected into a recording pixel constituting the pixel by referring to the data.

(Function) Since it has a means of predicting the heat storage state using density data for each pixel composed of multiple pixel points, the number of combinations of reference data and the heat storage state are better than the method of predicting the heat storage state for each pixel. 7! ! The number of calculations is considerably reduced. Furthermore, by means of determining the amount of energy injected into the recording pixel according to the state of heat accumulation, it is possible to faithfully reproduce recording with reduced density changes due to the influence of heat accumulation. It is also possible to support a recording method that utilizes thermal interference between each pixel within a pixel.

(Example) Hereinafter, an example of the present invention will be described based on the drawings.

FIG. 2 is a diagram for explaining the method of determining the implantation energy level in this embodiment.

In this embodiment, a line thermal head in which a large number of heat-generating resistors are arranged in a row is used to create fixed patterns a1 to a1 as shown in FIG.
The case where a4 is used will be explained. For example, if the current pixel row to be recorded is the Nth pixel row, a multi-value pattern (mark dot pattern in the pixel matrix and the energy level injected into each mark dot) is used to form the first pixel P in the N row. ) is the density data of the (I±1)th, (N-2), and (N-1)th pixels in the Nth column. Determined by reference. N-1 row, N-2
The purpose of referring to the density data of the I±1th pixel in the column and N column is to predict/estimate (calculate) the heat storage state of the four heat generating resistors about which the current pixel P is to be recorded. For example, the further away from the pixel of interest P, the smaller the influence from that pixel, so small overlapping is performed. Here, the past data of the pixel of interest P in the sub-scanning direction is referred to over two columns because generally heat accumulation in the sub-scanning direction, that is, heat accumulation due to temporal thermal interference, is due to
This is because the influence is greater than heat accumulation from the main scanning direction, that is, heat accumulation due to spatial thermal interference.

In the conventional binary recording method, since the heating state of the heating resistor is basically the same when recording each mark dot, injection into that mark dot is performed by counting the number of mark dots in the reference area around the mark dot to be recorded. It determined the energy level. However, if the energy injected into each mark dot is multivalued as in this embodiment, the heat generation state of the heating resistor during recording of each mark dot will be different. In the method of determining the injection energy level, the number of combinations of heat generation states in the reference area becomes enormous, making it extremely difficult to calculate the heat storage state, and it is not possible to use thermal interference between image points. On the other hand, by using the density data of each pixel as in this embodiment, the number of combinations is significantly reduced and the number of times the heat storage state is calculated is also reduced. (In the binary method, it is necessary to calculate the number of heating resistors for each line. In this example, it is necessary to calculate the number of heating resistors for each line.
4 matrix units), it is only necessary to calculate 1/4 of the number of heating resistors for every 4 lines, and the total number of calculations is 1/10 of the conventional number. ) Then, it is possible to consider the use of thermal interference between image points. The reason for using such pixel-by-pixel density data as reference data is
This recording method basically has a one-to-one correspondence between the density data and the multi-value pattern, so if the density data is known, the heating state of the heating resistor in the corresponding pixel is determined, and therefore the pixel is This is because the influence of heat storage due to recording is also determined.

FIG. 1 is a block diagram showing an embodiment of the configuration of an apparatus that enables recording as described above.

That is, a scanner, an A/D converter, a color conversion circuit,
Yellow (Y), magenta (M), cyan (
A digital signal containing density information of each pixel of each color of the three primary colors and black (BL) (BL may not be used in some cases) of the printer shown in C) is input manually to the heat storage state calculating section 2. As described above, the heat storage state calculation unit 2 calculates the N-th pixels in the reference area in the figure for the N columns of pixels P to be recorded shown in FIG.
The heat storage state at the time of pixel P recording is calculated based on the density information of the pixels in columns 2, N-1, and N, and the data is output to the multi-value pattern generation section 3 together with the original density information. The multi-value pattern generation unit 3 generates a fixed pattern consisting of a 4×4 matrix and mark dots within the pattern according to the density information of the pixel P to be recorded and the heat storage state data for the pixel P outputted from the heat storage state calculation unit 2. Determine the injection energy level. The multi-value data (hereinafter referred to as the implanted energy level value) of each dot determined by the multi-value pattern generating section 3 is processed by a line counter 7 and a line counter 7, which determines the position of each dot in the sub-scanning direction during recording. The pulse width generator 4 outputs the pulse width in conjunction with the dot counter 8 which determines the position in the main scanning direction. The pulse width generation section 4 determines the energization pulse width according to the multi-value data for each dot output from the multi-value pattern generation section 3, and converts the result into a signal tailored to the specifications of the thermal head drive section 5. Convert and output. The thermal head drive section 5 is an IC circuit consisting of a shift register, a latch, and a gate driver, and is mounted on the thermal head substrate, and drives each heating resistor according to the output data from the pulse width generation section 4. The width of the energizing pulse to R is changed, and a predetermined energy is injected into each dot. The timing controller 9 not only controls the series of operations described above, but also controls the operations of the printer mechanical section 6, such as paper feeding of the printer and raising and lowering of the ribbon feeding thermal head. Between the thermal head and the recording paper, there are y, M,
The ink ribbon is sandwiched by applying heat-softening c and BL inks sequentially, for example, and a yellow image is transferred first, followed by magenta, cyan, and finally black. Images of each color are transferred in this manner, and color recording is completed.

Next, the specific configuration and operation of the heat storage state calculation section, multi-value pattern generation section, and pulse width generation section, which are the main parts of the present invention, will be explained in detail.

FIG. 4 is a diagram showing a specific configuration of the heat storage state calculating section. The density data output from the multi-gradation signal processing circuit 1 shown in FIG. 1 is sufficient with the number of bits corresponding to the number of gradations to be expressed. For example, if 64 gradations of each color are to be expressed, a 6-bit parallel signal is sufficient. be.

This 6-bit signal is input to a pixel column memory 12 that stores three pixel columns. For example, an A4 size 12 dot/dragon resolution line thermal head (2 heating resistors)
560), the matrix size is 4×4 and 3
Since the pixel row is 6-bit data, the pixel row memory 12
Recently, it has been constructed from a RAM with a capacity of 2560/4 x 6 x 3 bits. The pixel column memory 12 contains address data output by a pixel row address counter 10 that determines the position of each pixel in one pixel column in the main scanning direction, and address data that determines the positions of three pixel columns in the sub-scanning direction. Density data for each pixel is stored in memory in accordance with address data output by the pixel column address counter 11. Now, it is assumed that density data of pixels in pixel columns N-2, N-1, and N columns are stored in correspondence with address output data CI, 1.2 of the pixel column address counter 11, respectively. When recording N pixel columns, the output of the pixel row address counter 10 is first set to a value corresponding to the first pixel in each column. Next, the output of the pixel column address counter 11 is set to a value corresponding to O, that is, the pixel column of the N-2 column, and the density data of the first row of the N-2 column is output from the pixel column memory.

This density data is stored in the latch 13, and then the output of the pixel column address counter 11 is set to 1, that is, the value corresponding to the pixel column of the N-1 column, and the density data of the first row of the N-1 column is stored. Output. This density data is stored in the latch 14, and then the output of the pixel column address counter 11 is set to a value corresponding to 2, that is, the N pixel column, and the density data of the first row of the N column is output. Latch this concentration data to 1
5 to be memorized.

At this time, it is assumed that the latch 16 stores the density data of O, which is not printed, but this corresponds to the data of the pixel one position before the pixel P in the main scanning direction in FIG. 16 stores the pixel data of the I+1th row. (As for the pixel at the farthest position in this way, there is no pixel in the reference area shown in FIG. 2, so density data in which nothing is printed is stored in this part in advance. In the embodiment, by storing the data for two columns before the first pixel column and the data for one row at both ends of the pixel column, the second pixel column is stored for every pixel.
The reference area shown in the figure will exist. Alternatively, these pixels may be processed by a method such as not recording the first and second pixel columns and one row of pixels at both ends of the pixel column. ) In this way, the N-2nd column, the N-1st column, the 1st of the Nth column
The density data of the first row is stored in the latches 13, 14, and 15, respectively, and the density data of the previous row of the first row is stored in the latch 16.

Next, the pixel row address counter 10 is incremented by one to cause the pixel column memory 12 to output the density data of the pixel in the Nth column and second row. At this time, the heat storage state calculation circuit 17 calculates the lunch 13.1.
4. Retrieve each data of 16 and pixel column memory 12.

These data are density data of pixels within the reference area for predicting the heat storage state for the pixel P as shown in FIG. The heat storage state calculation circuit 17 predicts (calculates) the heat storage state of the pixel to be recorded, that is, the pixel in the first row of column N based on these density data, and multi-values the heat storage state data corresponding to the state. It is output to the pattern generation section 3. At this time, the density data of the latch 12, that is, the pixel in the Nth row and the first column, is also output to the multi-value pattern generation section 3. This heat storage state calculation circuit 17 is specifically constituted by a ROM, is addressed by inputted 6 bits x 4 to 24 bits of data, and outputs heat storage state data corresponding to this. However, since it is not possible to specify a 24-bit address in one ROM, the address is actually divided into multiple ROMs and input.

Next, the heat storage state data for the pixel in the N column, second row is predicted. First, the output of the latch 15, that is, the density data of the pixel in the N column, first row, is stored in the latch 15. Next, the value of the pixel column address counter 19 is set to 0, and
The density data of the pixel in the row is stored in the latch 13. Similarly, the density data of the pixel in column N-1 and row 2 is latched to 1.
4, the density data of the pixel in the Nth column and second row is stored in the latch 15, respectively. Next, the pixel row address counter 10 is further advanced by one to output the density data of the pixel in column N, row 3 from the pixel column memory 12, and these data are taken into the heat storage state calculation circuit 17, and the density data of the pixel in column N, row 2 is output. The heat storage state data of the eye pixels is output to the multi-value pattern generation section 3. The above operation is repeated for the number of pixels in the main scanning direction for N columns. In this case, it will be repeated 2560/4-640 times.

When the calculation of the heat storage state as described above is completed for the N column, the same operation is performed for the N+1 column. In this case, the reference areas are column N-2, column N-1, column N to column N-1,
The pixel column memory 12 is changed to three columns, N column and N+1 column.
In the area where N-2 pixel column data was stored, N+
Data for one pixel column is stored. That is, the pixel column address counter 11 is set to O, and the pixel row address counter 10 stores the data of the N+1 pixel column. Thereafter, the same operation as described above is performed except that the value of the pixel column address is different from that in the case of the N column described above, and heat storage state data of the pixels in the N+1 column is output. Such an operation is performed for all pixel columns.

Next, the operation of the multi-value pattern generation section 3 will be explained. The heat storage state data outputted from the heat storage state calculation circuit 17 and the density data of the pixel outputted from the latch 15 for the pixel to be recorded are processed by the multi-value pattern generation unit 3 into a fixed pattern of pixels in a 4×4 matrix and each Corresponding to a multi-value pattern consisting of the injection energy level of mark dots, the multi-value data of each dot of the multi-value pattern is specified by the line counter 7 and dot counter 8 and output from the multi-value pattern generating section 3.

In this embodiment, since the pixel matrix is 4×4, the dot counter 8 determines the dot position in the main scanning direction.
And the line counter 7 for determining the dot position in the sub-scanning direction counts 4 such as 0, 1, 2, 3. For example, when attempting to record N rows of pixels, first set the line counter 7 to the first line O, and each time the multi-value pattern data of that pixel consisting of heat storage data and density data is input, the dot counter 7 is set to the first line O. By changing 8 to 0.1.2.3, multi-value data of 2560 dots for the first line of each pixel will be output. When the output for the first line is completed, the line counter 7 is set to 1, and the multi-value data of the second line of each pixel is output.
The multi-value data of the 4th line and the 4th line are output, and the above operation is repeated when the next pixel column is reached.

This multi-value pattern generation section 3 is also specifically constituted by a ROM, and is addressed by input heat storage state data, concentration data, dot counter, and line counter data, and outputs multi-value data corresponding to this. . In order to speed up recording, the heat storage state data and density data are alternately stored in two RAMs for each pixel column, although they are not shown in the same figure. When recording, data from each RAM is taken out four times (one pixel column for four lines).

In this way, while four lines of one pixel column are being recorded, the heat storage state of the next pixel column can be calculated, thereby enabling high-speed recording.

The multi-value data output from the multi-value pattern generation section 3 is output to the pulse width generation section 4. The operation of this pulse width generator 4 will be explained. The energy injected into each pixel is
A plurality of short energizing pulses as shown in FIG. 5 are combined and varied. As shown in FIG. 6, the pulse width generation section 4 receives data on each of these energizing pulse widths and from the multi-value pattern generation section 3.

It is comprised of each pulse width generation section 18 and each pulse selection data generation section 19, each outputting data for selecting each energization pulse according to multi-value data. For each recording of one line, the multi-value data is manually generated as many times as the number of energizing pulses, and the pulse width data EN corresponding to the energizing pulse number and the energizing pulse selection data (1 bit) corresponding to the energizing pulse number and multi-value data are generated. SIN to thermal head drive circuit 5
It will be output to . For example, as shown in Figure 5, 0
．． 5m 5% 0.4m S % 0.3m s
, 0.2 m s.

Consisting of five pulses of 0.1 m s, when recording with a pulse width of 0.7 ms when multilevel data is 7, the pulse width data (0.5 m s,
0.4ms, 0.3ms,
0.2 m s, 0.1 m s (and energization pulse selection data (0, 1, 1, 0, O
) are output to the ENSSIN terminal. Of course, the energization pulse selection data is output for one line, ie, 2560 bits, in accordance with the multi-value data of each pixel for each pulse output.

Data for each pulse width is stored in the ROM, and the output of the counter 20 that counts pulse numbers 1 to 5 is input to the address, and 8-bit data corresponding to each value is stored in the pulse width counter (not shown in the figure). output). At the same time, the EN signal is turned on. The count loaded by this data is counted by CLK, and after counting a predetermined number of times, the EN signal is turned off and at the same time, the value of the pulse number counter 20 is incremented by one.

The value of this counter is input to the ROM to output the next pulse width data. In this example, this operation is performed five times for each line recorded, and is repeated for all lines. In addition to the counter value, this ROM is input with the detected temperature of the thermal head board (the output voltage from the thermistor converted into a digital signal by the A/D converter) to the address, and the pulse width is determined. It is possible to increase the effect even more by reducing the influence of local heat accumulation as well as overall heat accumulation.

Each pulse selection data is also stored in the ROM, and the output of the pulse number counter and multi-value data are manually entered at the address, and 1-bit data corresponding to the value is output. Note that it may be pulse data. For example, if the multivalued data is 6, it is 00110, so 0.
A pulse width of 5ms will be selected. However, in this case, a circuit is required to extract data corresponding to the pulse number from this data string.

FIG. 7 shows an example in which the energy injected into a recording pixel actually changes depending on the surrounding pixels due to the above-described operation.

The recording density increases corresponding to the density code O to 30, and the multi-value data O to 10 is 0.1 m from O to 1 ms, respectively.
It is assumed that the pulse width corresponds to s steps. When the pixel P of interest has a density of density code 30 and all surrounding pixels are in the state of 0 (1)'', the multi-value pattern is (1).

On the other hand, when the density of the surrounding pixels of the pixel of interest P is large to some extent, as in the state (n), the multi-value pattern is (
2), and the implantation energy is smaller than in (1). Furthermore, when the density of the surrounding pixels increases as shown in (III), the multilevel pattern becomes as shown in (3), and the implantation energy becomes even smaller. In this way, since the heat storage state changes depending on the concentration (injection energy m) of the surrounding pixels, the multivalue pattern is changed to correspond to that (taking into account each pixel within the pixel and mutual thermal interference). perform an action.

FIG. 8 is a diagram showing another embodiment of the present invention.

In this case, the heat storage state data predicted by the storage calculation section 2 is output to the pulse width generation section 4.

FIG. 9 shows the configuration of the pulse width generator 4. The basic configuration is the same as the previous embodiment, but the difference is the operation of each pulse selection data generation section 19. In addition to the multi-value data and the output of the pulse number counter, heat storage state data is input here, and each pulse selection data changes depending on the heat storage state. For example, if the multi-value data is 5 and there is no heat storage,
Each pulse selection data is (0, 1, 1, 010), that is, a 0.7ms pulse, but if the data indicating that heat storage is not to be inputted, a 06m5 pulse shorter than the pulse width is selected. In other words, each pulse selection data is (0,0
, 1, 1, 1). In this way, in this example, instead of changing the multi-value data depending on the heat storage state as in the previous embodiment, each pulse selection data is directly changed depending on the heat storage state.

In addition, in these embodiments, the heat storage state j13'ItI calculation unit 2
An example using a ROM table has been shown, but since it is not possible to cope with an increase in the number of bits of input data from the reference area, an arithmetic circuit may also be used.

As described above, the present invention can be implemented with various modifications within the scope of the invention.

[Effects of the Invention] According to the present invention, the amount of energy to be injected into a mark dot of a pixel to be recorded can be determined by also taking into consideration the heat storage state of the thermal head predicted by referring to image data of surrounding pixels. can. Therefore, it is possible to always obtain a recorded image with a stable density without causing problems such as gradual changes in image density as in the conventional method.

[Brief explanation of the drawing]

) Figure 1 is a block diagram showing the configuration of a thermal recording device according to an embodiment of the present invention, Figure 2 is a schematic diagram for explaining the principle of the present invention, and Figure 3 is a fixed recording diagram in the above embodiment. A diagram showing the relationship between the amount of energy injected into the pattern and the recording density, FIG. 4 is a block diagram showing the configuration of the heat storage state calculating section, FIG. 5 is a diagram showing the method of generating the energizing pulse, and FIG. 6 is the pulse width generating section. FIG. 7 is a diagram showing an example of a multi-value pattern of a pixel of interest that changes depending on the above embodiment.
FIG. 8 is a block diagram showing the configuration of a thermal recording apparatus according to another embodiment of the present invention, and FIG. 9 is a diagram showing the configuration of a pulse width generating section thereof. P...Pixel 2...Heat storage state calculation unit 4...Pulse width generation unit 21...Thermal head attorney Yudo Noriyuki Chika Kikuo Takehana Figure 2 Figure 3 Figure 5 Procedures Written amendment (method) 1. Display of the case Japanese Patent Application No. 1982-125692 2, Name of the invention Thermal recording device 3, Person making the amendment Relationship to the case Patent applicant (307) Toshiba Corporation 4, Agent 105 Description and drawings 7, Amendment contents of

Claims (1)

[Claims] (1) In a thermal recording device that expresses halftones by modulating the coloring area formed within a pixel consisting of multiple dots, the thermal head and means for determining the level of energy to be injected into the mark dot of the pixel by referring to the data of the heat storage state prediction means and the density data of the pixel. Thermal recording device.