JPH04367985A - Automatic read position correcting method for image scanner device using plural line sensors - Google Patents

Abstract

PURPOSE:To automatize the adjusting work executed by a visual observation, to eliminate the variance and to improve the accuracy of a joint by calculating in advance a shift amount of a read position of each sensor, and executing automatically a correction of data as an offset amount at the time of reading general data. CONSTITUTION:In such a position as steps over a joint of line sensors No.0 and No.1, a test chart 3 on which a wedge shape 2 is printed clearly is read exactly without bending in the advance direction. Data which is read by the line sensors No.0 and No.1 is subjected to analog-to-digital conversion by a pre-processing part, stored in a line buffer as last data, and a multiplexer reads out the data from the respective line buffers of the line sensors No.0 and No.1. Subsequently, by a position correcting register, a data read-out start position is offset by an instruction amount of the register, and stored in a frame memory as composite image data of the line sensors No.0 and No.1, and outputted to a printer depending on the situation.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for automatically correcting the reading position of an image scanner using a plurality of line sensors for adjusting the accuracy of the reading position of an image scanner and maintaining the accuracy.

0002

2. Description of the Related Art In general, an image scanner device captures a two-dimensional image as time-series data by repeating one-dimensional scanning. A one-dimensional line sensor is often used as this sensor section. Devices that read large images often use a plurality of small line sensors for technical and economic reasons. In this case, data output from a plurality of sensors is combined to form data in the main scanning direction.

0003

When a plurality of line sensors are used, they are arranged so that the areas covered by the respective sensors overlap with each other in order to avoid missing data at the joints. However, due to reasons such as the arrangement method and assembly accuracy, variations often occur in the relative positional relationship between the sensors in each device. In order to remedy this variation, a method is used in which the relative position of each sensor is stored as a correction value unique to each device and corrected during data synthesis. Conventionally, data read by an image scanner device is output to a plotter or displayed on a display, and an operator repeatedly performs visual trial and error to determine this correction value. In addition, a method has been devised in which a test chart is printed with marks whose intervals have been precisely measured, and a position correction value in the main scanning direction is calculated based on the interval between the read marks. This is the original correction value, and the test chart must be set strictly parallel to the direction of the sensor or at a certain angle, requiring a lot of effort or special equipment, and is not very practical. . The present invention has been devised in view of the above-mentioned points, and its purpose is to provide an image scanner device using a plurality of line sensors that can automate visual adjustment work and improve the accuracy of seamless joints. An object of the present invention is to provide a reading position automatic correction method.

0004

[Means for Solving the Problem] In other words, the means for achieving the object is to use a triangular or inverted triangular shape having continuous differential coefficients in an image scanner device that reads images using a plurality of line sensors arranged in parallel. Using a specially printed test chart in the shape of , and a frame memory that stores the digital data obtained by reading this chart, the amount of deviation in the reading position of each sensor is calculated in advance, and the amount of deviation in the reading position of each sensor is calculated in advance. In this method, a two-dimensional position correction value is automatically calculated for data correction as an offset amount. Note that the position in the height direction may be determined by each sensor alone, and there is no problem with multiple sensors, and this is excluded from the present invention.

[0005]

[Operation] The image scanner device reads the figure printed on the test chart and temporarily stores it in the frame memory. This data is scanned in the sub-scanning direction, and for each column, the length of the continuous black part (hereinafter referred to as black run),
Calculate the starting and ending positions. By checking the length of this black run, it is possible to know the continuity in the main scanning direction, and by checking the start or end position, it is possible to know the continuity in the sub-scanning direction. A correction value is calculated so that data spanning multiple sensors is continuous in these two directions, and the correction value is written in the position correction register. The multiplexer offsets the start position of reading data from the line buffer of each sensor by the amount indicated by the register, thereby making it possible to automatically correct the position.

[0006]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below in detail with reference to the drawings. FIG. 1 is an explanatory diagram of an image scanner device showing one embodiment of the present invention, FIG. 2 is an explanatory diagram of a test chart of one embodiment of the present invention, and FIG. 3 is a sub-scanning diagram of data output from each sensor by a CPU. FIG. 4 is an explanatory diagram of the data table created in FIG. 3, and FIG. 5 is an explanatory diagram of the action of FIG. 3. Note that here, as shown in FIG. 1, line sensor No. 0 and No. An example using two sensors will be described, but the same applies to the case where three or more sensors are used. In FIG. 1, line sensor No. 0 and no
．． The data read by the two sensors No. 1 is converted from analog to digital in the preprocessing section, and stored in the line buffer as the last data. 0 and No. A multiplexer (MUX) reads data from each line buffer of 1. Then, the position correction register of the present invention, which will be described later, offsets the data reading start position by the amount specified by the register, and then the line sensor No. 0 and No. The combined image data is stored in a frame memory and, depending on the case, is immediately output to a printer. FIG. 2 shows the line sensor No. 1 arranged on the paper 1 for creating the position correction register of FIG. 0 and No. 1 and the corresponding placement position of a printed test chart 3 having a wedge shape 2 having continuous differential coefficients. Next, the process shown in FIG. 2 until data is obtained as a position correction register will be described with reference to FIGS. 3 to 5.

Line sensor No. shown in FIG. 0 and No.
Wedge-shaped 2 is placed in a position that straddles the joint between the two sensors in 1.
To accurately read a test chart 3 on which is clearly printed without bending in the direction of travel. The wedge shape 1a is assumed to have an angle of 45° or more so that the lengths of black runs in adjacent columns differ by at least one pixel, as shown in FIG. This data is then scanned in the sub-scanning direction by a CPU (not shown) to create the table shown in FIG. That is, for each column, the length BLL of consecutive black runs of black data in the case of binary data, black data equal to or higher than a certain threshold in the case of multivalue data, the starting Y coordinate BLS, and the ending Y coordinate BLE.
Calculated as

From this table, first, the correction amount in the main scanning direction is calculated. In this example, the length BLL of this black run should increase monotonically at a constant rate defined by the wedge angle. The position of the column where the monotonically increasing continuity ends due to a decrease or an extreme increase is the sensor junction. As shown in FIG. 5, No. Assuming that data from one sensor has started, if the position is corrected correctly, the length of the black run in the n+1st column is BLLn+1,
If the length of the black run in the nth column is BLLn, then BLLn+
The relationship 1=BLLn+α1 (α1 is the rate of increase) should hold true. Now, BLLn+1 in column n+1 is BLL
BLLn+1+ of column n+1+m, smaller than n+α1
Assuming that m is closest to BLLn+α1, -m is the correction value in the main scanning direction. That is, sensor No. At the end of the output data of No. It is better to bring it closer to the beginning of 1. Conversely, BLLn+1 in column n+1 is larger than BLLn+α1, and BLLn+1-m in column n+1-m is BLL.
Assuming that it is closest to n+α1, +m is the correction value in the main scanning direction.

Next, the amount of correction in the sub-scanning direction is calculated. The BLS of the starting X coordinate of the black run and the BLE of the ending coordinate, whichever has a larger amount of change depending on the column, is adopted as the reference position. This is because continuity can be checked more accurately by using the side with a larger slope as a reference between the upper and lower sides of the wedge shape. Let us now assume that BLS is the standard. The BLS should monotonically increase or decrease at a constant rate defined by the set angle of the test chart. If the X coordinate at which the black run in the n+1 column starts is BLSn+1, and the X coordinate at which the black run in the n column starts is BLSn, then the following relationship should hold: BLSn+1=BLSn+α2 (α2 is the rate of change). In the case of the above example, the correction value in the sub-scanning direction is (BLSn+α2)-(BLSn+1+m
), and in the case of the latter example, (BLSn+α2)−(BLS
n+1-m). The correction values in the two directions thus obtained are written into the position correction register. The multiplexer offsets the start position of reading data from the line buffer of each sensor by the amount indicated by the register, thereby making it possible to automatically correct the position. That is, the automatic position correction mode is stopped, and the above-mentioned offset amount is sequentially corrected for the read data in the normal reading mode.

0010

[Effects of the Invention] As explained above, according to the present invention, C
An image scanner device that has a PU, a means for switching to automatic adjustment mode, a rewritable position correction register, and a frame memory that can store test chart data, has an angle of more than a certain level.
The wedge shape between straight lines can be automatically calculated by reading a clearly printed test chart that straddles the line sensor joint.
Adjustment work can be automated, eliminating the need for visual inspection. Furthermore, since the test chart can be set roughly, the efficiency of adjustment work is significantly improved. Furthermore, since a skilled worker is not required, even a user can easily make adjustments, and since there is no variation among workers, it is easy to maintain the accuracy of the joints.

[0011]

[Brief explanation of drawings]

FIG. 1 is an explanatory diagram of an image scanner device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram of a test chart according to an embodiment of the present invention.

FIG. 3 is data obtained by scanning data output from each sensor in FIG. 1 in the sub-scanning direction.

FIG. 4 is a data table that is created.

FIG. 5 is an explanatory diagram of the action.

0012

[Explanation of symbols]

1 Paper 2 Cuneiform 3 Test chart

Claims (1)

[Claims] Claim 1: An image scanner device that reads images using a plurality of line sensors arranged in parallel,
Using a specially printed test chart in the shape of a triangle or inverted triangle with continuous differential coefficients, and a frame memory that stores the digital data obtained by reading this chart, we can calculate the amount of deviation in the reading position of each sensor. A reading position automatic correction method for an image scanner device using a plurality of line sensors, in which the amount is calculated in advance and the data is automatically corrected as an offset amount during general data reading.