JPH06225143A - Picture reader - Google Patents

Abstract

PURPOSE:To automatically correct a nonlinearity of an image pickup element and an amplifier or the like at a low cost by inputting regular read picture data of n-gradation subject to shading correction outputted from an A/D converter to a linearity correction circuit. CONSTITUTION:A linearity correction circuit 30 is connected to a post-stage of an A/D converter 13 and the reader is provided with time division read data collection control means 21, 22, 24, linearity correction data generating means 21, 22, and linearity correction data setting control means 21, 22. Then linearity correction data C1-Cn are generated by using time division read data obtained by reading a reference plane for each division unit time being equally n-divisions of a storage time of an image pickup element 11 for n-gradation reading and using a linearity constant revised for each division unit time and set automatically to the circuit 30. Thus, the picture reading of ngradation in which the nonlinearity of a read unit is corrected is executed at a low cost with high accuracy.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image reading apparatus, and more particularly to an apparatus capable of automatically correcting the non-linearity of a reading unit.

[0002]

2. Description of the Related Art In FIG. 7, a reading surface 2 on which an original (image) is placed is formed on the upper surface side of an image reading apparatus 1 of fixed original type, and a reference surface (for example, white) is placed at a position off the reading surface 2. A shading correction plate 5 is provided. The reading unit 10 is capable of reciprocating movement in the sub-scanning direction Y, and includes an image sensor 11, an amplifier 12, a multiplication A / D converter 13, a fluorescent lamp, and the like shown in FIG. However, amplifier 12, A / D
The converter 13 is often arranged stationary. A control unit 20 includes an MPU, ROM, RAM and the like.

By the way, the light emission amount of the fluorescent lamp and the image pickup device 11
Due to its nature, the sensitivity of 1 varies in the main scanning direction X. Therefore, it is well known that the shading correction data Dc is generated in advance using the reference surface 5 and is stored in the control unit 20, and the shading correction is performed when the image is read.

That is, the reading unit 10 is moved in the sub-scanning direction Y and the image is read by the image pickup device (CCD) 11.
Original analog read image data Da from the image sensor 11
After being amplified by the amplifier 12, the A / D converter 1
Input to 3. Then, the shading correction data Dc input from the control unit 20 is subjected to shading correction to obtain the regular read image data Dd. Therefore, the regular read image data Dd is one in which the influence of the variation in the light emission amount of the fluorescent lamp in the main scanning direction X is removed.

[0005]

By the way, since the regular read image data Dd is not only the binary data of "white data" and "black data" but also the intermediate color data thereof, it is necessary to identify a lot of ( n) Gradation is being read. For example, black data “0” to white data “255”. For this reason, it is necessary to select the image pickup device 11, the amplifier 12, and the A / D converter 13 which have good linearity characteristics, resulting in high cost. In particular, since the amplifier 12 often obtains linearity at the expense of the dynamic range, it is easy to make it an excessive device.

However, even if such inconvenience is tolerated, the non-linearity still remains, which hinders further improvement in accuracy. That is, the output of the A / D converter 13, that is, the regular read image data Dd is distorted in the middle of the ideal curve shown by the two-dot chain line in FIG. 9, for example, like the curve shown by the solid line. is there.

As one of the solutions to this problem, a large number (n) of reference planes (5) corresponding to each gradation are used to collect linearity correction data for each gradation, and the A / D converter 1 is used.
3 and the control unit 20 between the regular read image data D
It is conceivable to correct d. However, even if it is necessary to prepare a large number (n) of reference planes (5) each reflectance of which is accurate, it is difficult to realize technically, economically and handling.

It is an object of the present invention to provide an image reading apparatus which is capable of automatically correcting the non-linearity of an image pickup device, an amplifier and the like to read with high accuracy and which is easy to handle at low cost.

[0009]

According to the present invention, linearity correction data for each read gradation is automatically generated by utilizing the accumulation time of the image pickup device instead of using an empirical rule, that is, a large number (n) of reference planes. It was created based on a new idea, and it is configured so that linearity correction data can be collected accurately and quickly, and that linearity correction can be automatically performed during actual image reading.

That is, the image reading apparatus according to the present invention is
The original read image data read by moving the reading unit including the image sensor and the image relatively in the sub-scanning direction and the shading correction data generated by reading the reference plane in advance are input to the A / D converter. In an image reading apparatus that obtains n-gradation normal read image data while performing shading correction in the main scanning direction, the normal read image data is input to the subsequent stage of the A / D converter and corrected using linearity correction data. A linearity correction circuit for generating and outputting normal read image data is connected, and the reference plane is read using the image sensor for each division unit time obtained by equally dividing the accumulation time of the image sensor by n, and each time division read data is read. Linearity correction using time-division read data collection control means for obtaining the time-division read data, each time-division read data, and each linearity constant updated for each division unit time. And linearity correction data generating means for generating over data, characterized by providing a linearity correction data setting control means for setting the generated linearity correction data to the straight line correction circuit.

[0011]

According to the present invention having the above-mentioned structure, before the image reading operation, the time-division read data collection control means is operated to obtain each time-division read data while reading the reference plane with the image pickup device. At the same time, the linearity correction data generation means generates each linearity correction data by using the linearity constant updated for each division unit time and each collected time-division read data. Then, the linearity correction data setting control means causes the A / D
Each linearity correction data is automatically set in the linearity correction circuit connected to the latter stage of the converter.

Thus, when the image reading operation is started, the A / D
Shading-corrected n output from the converter
The normal read image data of gradation is input to the linearity correction circuit, corrected by the linearity correction data, and output as corrected normal read image data. Therefore, the non-linearity of the image sensor, the amplifier, and the A / D converter can be automatically corrected.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. As shown in FIG. 1, the image reading apparatus (1) has the same basic structure as the conventional example (FIGS. 7 and 8), and the linearity correction circuit 30 is connected to the subsequent stage of the A / D converter 13. At the same time, the time division read data collection control means (21, 2
2, 24) and linearity correction data generation means (21, 22)
And linearity correction data setting control means (21, 22) are provided, and the linearity correction data capable of canceling the nonlinearity of the reading unit 10 (11, 12, 13) by utilizing the accumulation time t of the image pickup device 11. Linearity correction circuit 3 for collecting C1 to Cn and
It is configured so that it can be automatically set to 0.

In FIG. 1, the control unit 20 has an MP
U21, ROM for storing various programs and fixed data
22, a RAM 23 for temporarily storing various data, a timer 24, an interface 25 for performing data communication with an external device, an image RAM for storing the read regular read image data Dd to the corrected regular read image data Ddc
26, DMAC 27, etc., and drive-controls the entire apparatus. The stepping motor 29 connected via the driver 28 reciprocates the reading unit 10 (11) in the sub-scanning direction Y. The shading correction data Dc generated in advance is stored in the RAM 23 and is input to the multiplication A / D converter 13 at appropriate times.

First, the linearity correction circuit 30 performs linearity correction by causing the linearity correction data C1 to Cn to participate in the shading-corrected normally read image data Dd input from the A / D converter 13. The normal read image data Ddc is generated and output, and as shown in FIG.
It is connected between the converter 30 and the control unit 20. Now, if all of the linearity correction data (C1 to Cn) are set as default values (1 to n), the relationship between the input IN and the output OUT is as shown by the curve (1) in FIG. The default value (1 to n) prior to the time-division read data collection is automatically set by the MPU 21 (ST11 in FIG. 2).

Now, the time division read data collection control means is
The accumulation time t of the image pickup device 11 is equally divided by "n" of the read n gradations, that is, t / shown in FIG.
, nt / n, the image pickup device 11 reads the reference plane (white shading correction plate) 5 while obtaining the time-division read data C1 to Cn.
It consists of an OM 22 and a timer 24.

That is, the MPU 21 is the reading unit 10
(Imaging element 11) is moved to the reference plane 5 (ST of FIG.
At the same time, the default values C1 to Cn (= 1 to n) for each division unit time t / n, 2t / n, ..., nt / n are set in the linearity correction circuit 30 (ST11). Then, in this state, the image sensor 11 is used to read the reference plane 5 at a normal accumulation time t, and the amplifier 12 and the A / D converter 13 are read.
Shading correction is performed through (ST12).

A linearity constant A updated every division unit time is set here (ST13). In the case of the first division unit time t / n, A = 1. Then, the accumulation time (At /
n) is obtained and set (ST14). Therefore, in this case, the time-division read data Ba output from the linearity correction circuit 30 can be obtained (ST15). Hereinafter, the same procedure is performed while updating the linearity constant A (A = A + 1) while A ≧ n
It is repeated until it becomes (ST17, 18, 14, 15). Both the linearity constant A and the time-division read data Ba are temporarily stored in the RAM 23.

That is, for example, in the case of A = i, that is, the division unit time it / n, as shown in FIG. 4, the time division output from the linearity correction circuit 30 in which the default value (Ci = i) is set. Read data Bi is obtained.
In other words, the non-linearity of the reading unit 10 (11, 12, 13) is divided equally into the storage time t of the image sensor 11 by the gradation n, and the time-division read data Ba for each division unit time At / n.
(A = 1 to n) are collected. Each time-division read data Ba (a: 1 to n) has the curve set in FIG. 4 because the value set in the linearity correction circuit 30 is the default value Ca (= 1 to n). The curve of FIG. 4 is the same as the curve of FIG.

The linearity correction data generating means that operates at the same time as this is composed of the MPU 21 and the ROM 22 and ST of FIG.
16 is executed. That is, as shown in FIG. 4, each time-division read data Ba, Bb, ..., Bi, ..., Bn and each linearity constant A (= 1), A (= 2), ..., A (= i),
, A (= n) and linearity correction data Ca, C
b, ..., Ci, ..., Cn are generated. In this example,
It is calculated as Ci = [A (= i) -Bi]. That is, as is clear from FIG. 4, each linearity correction data (Ci) has a difference (Bi−) between the time-division read data (Bi) read by the actual curve and the data (i) based on the ideal curve. It is understood that i) is given a minus sign.

Here, the linearity correction data setting control means replaces the generated linearity correction data C1 to Cn with the default values (1 to n) described above, and the linearity correction circuit 30 is used.
The MPU 21 and the ROM 22 are included and are executed in ST19 of FIG. Therefore, if the ideal input IN (1 to n) is added to the linearity correction circuit 30 after that, the linearity correction is performed by the linearity correction data C1 to Cn, so that the output OUT has an ideal curve. It does not become (1) but becomes a downwardly convex curve (2). In conclusion, if the input IN is the normal read image data Dd due to the non-linearity of the reading unit 10 (11, 12, 13) shown by the curve in FIG. 6, the output OUT ( Corrected regular read image data Ddc)
Can be obtained.

Next, the operation of this embodiment will be described. It is assumed that the shading correction data Dc has already been generated and stored. Here, the linearity correction data C1
~ Cn automatic setting operation command is issued.

First, the MPU 21 moves the reading unit 10 to the reference plane 5 (ST10 in FIG. 2). Further, the linearity correction circuit 30 has the default values (1 to 1) of the linearity correction data C1 to Cn which can be the curve (1) shown in FIG.
n) is set. Subsequently, shading correction is performed at the normal accumulation time t. That is, the A / D converter 13
From the normal read image data Dd after shading correction
Is output.

Next, the MPU 21 sets a linearity constant A for updating the storage time t of the image pickup device 11 every division unit time which is equally divided by "n" of n gradations. The first time, A =
It becomes 1 (ST13).

Then, the MPU 21 as the time division read data collection control means (21, 22, 24) sets the accumulation time (At / n) for each division unit time, and reads the time division read data Ba at that time. (ST14, 15). Immediately after this, the linearity correction data generation means (21, 22) operates to obtain linearity correction data Ca (= A-Ba) in ST16. Hereinafter, the linearity constant A is updated in ST17 (A = A
+1), and this is repeated until A ≧ n.

Therefore, when the linearity constant A is i (= A), the time-division read data at the division unit time it / n becomes Bi shown in FIG. 4 and its accumulation time (it
The linearity correction data in / n) is Ci. Therefore, all the linearity correction data C1 to Cn can be expressed as the curves shown in FIG.

Thus, the linearity correction data setting control means (21, 22) automatically sets the linearity correction data C1 to Cn thus generated in the linearity correction circuit 30 (ST19).

Thereafter, when the image is read while moving the reading unit 10 in the sub-scanning direction Y, the analog original read image data Da from the image pickup device 11 is amplified by the amplifier 12 and input to the multiplication type A / D converter 13. It Then,
The shading correction data Dc for each pixel in the main scanning direction X is subjected to shading correction, and is output as normal read image data Dd. That is, the main scanning direction X of the fluorescent lamp
The data Dd is the data Dd from which the influence of the variation in the light emission amount is removed.

Subsequently, the regular read image data Dd is linearly corrected in the linearity correction circuit 30 by the linearity correction data C1 to Cn on the curve (2) in FIG. That is, the regular read image data D on the curve shown in FIG.
d is an image pickup element 11, an amplifier 12, an A / D converter 1
The non-linearity of No. 3 is corrected, and the corrected regular read image data Ddc according to the ideal curve of FIG. 6 is obtained. The corrected regular read image data Ddc is stored in the image RAM 26 by the MPU 21, and is transmitted to an external device as necessary and through the interface 25.

According to this embodiment, however, the linearity correction circuit 30 is connected to the subsequent stage of the A / D converter 13 and the time-division read data collection control means (21, 22, 24).
And linearity correction data generation means (21, 22) and linearity correction data setting control means (21, 22) are provided, and the unit time of dividing the accumulation time t of the image sensor 11 by n for n gradation reading is divided. The time-division read data (Ba) obtained by reading the reference plane 5 every (At / n) and updated every division unit time (A
Since the linearity correction data C1 to Cn are generated using the linearity constant (A) that is equal to (= A + 1) and these are automatically set in the linearity correction circuit 30,
It is possible to highly accurately read the image of n gradations in which the non-linearity of the reading unit 10 (11, 12, 13) is corrected.

Further, the time-division read data collection control means (2
1, 22, 24) reads the time-division read data (Ba) accumulated every division unit time (At / n) obtained by equally dividing the accumulation time t of the image sensor 11 by the gradation n, and corrects the linearity. Since the data generation means (21, 22) generates the linear correction data (Ca) for canceling the non-linearity of the reading unit 10 (11, 12, 13) at that time, it is for each gradation. Since it is not necessary to prepare the n reference surfaces 5 having the light reflectance of, it is very easy to handle.

[0032]

According to the present invention, a linearity correction circuit is connected to the subsequent stage of the A / D converter, and time-division read data collection control means, linearity correction data generation means, and linearity correction data setting control means. By using the time-division read data obtained by reading the reference plane for each division unit time obtained by equally dividing the storage time of the image sensor by n in the n-gradation reading and the linearity constant updated for each division unit time. Since it is configured to generate linearity correction data and automatically set these in the linearity correction circuit, n-gradation image reading with non-linearity correction of the reading unit can be performed with high accuracy and handled at low cost. Is also easy.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention.

FIG. 2 is likewise a flow chart for explaining the operation.

FIG. 3 is a diagram for explaining the characteristics of the linearity correction circuit when not corrected.

FIG. 4 is a diagram for explaining the relationship among accumulation time, division unit time, time division read data, and linearity correction data.

FIG. 5 is a diagram for explaining the input / output relationship when the ideal linearly read image data is input to the linearity correction circuit and correction is performed using the linearity correction data.

FIG. 6 is a diagram for explaining the relationship between actual regular read image data and corrected regular read image data after being corrected using linearity correction data.

FIG. 7 is a schematic view showing a conventional example.

FIG. 8 is a block diagram showing a conventional example.

FIG. 9 is a diagram for explaining the problems of the conventional example.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Image reading device 2 Reading surface 5 Reference plane 10 Reading unit 11 Imaging element 12 Amplifier 13 A / D converter 20 Control unit 21 MPU (time-division reading data collection control means, linearity correction data generation means, linearity correction data setting control 22 ROM (time-division read data collection control means, linearity correction data generation means, linearity correction data setting control means) 23 RAM 24 Timer (time-division read data collection control means) 26 Image RAM 29 Stepping motor 30 Linearity Correction circuit X Main scanning direction Y Sub-scanning direction Da Original read image data Dc Shading correction data Dd Normal read image data Ddc Corrected normal read image data C1 to Cn Linearity correction data t Storage time

Claims (1)

[Claims] 1. An A / D converter which reads original read image data read by moving a reading unit including an image sensor and an image relative to each other in the sub-scanning direction, and shading correction data generated by reading a reference plane in advance. In an image reading apparatus for inputting to a converter and performing shading correction in the main scanning direction to obtain n-gradation normal read image data, the normal read image data is input to a subsequent stage of the A / D converter and linearity correction data is input. Connect a linearity correction circuit that generates and outputs corrected regular read image data using
Time-division read data collection control means for obtaining each time-division read data by reading the reference surface using the image-pickup element at every division unit time obtained by equally dividing the accumulation time of the image-pickup element by n, and each time-division read data And a linearity correction data generating means for generating linearity correction data by using each linearity constant updated every division unit time, and a straight line for setting the generated linearity correction data in the linearity correction circuit. An image reading apparatus comprising: a property correction data setting control means.