JPH01144773A - Original reader - Google Patents

Abstract

PURPOSE:To prevent the malfunction due to dirt or the like of an original platen by dividing read image information into unit areas having a prescribed size independently of the moving speed of the read position and detecting coordinates of an original according as the result of decision of the image information in each divided unit area is all white or not. CONSTITUTION:Since the number of encoder pulses of a DC motor 8-18 which moves an optical unit is proportional to the speed of an optical system, the size of the original of the unit area is not changed even in the case of the change of the speed of the optical system at the time of varying magnifications to recognize the original without the malfunction if a RAM 8-5 is refreshed based on encoder pulses of the DC motor 8-18. Unit area having a fixed size can be used as an original presence/absence detection area independently of the moving speed of the optical unit in this manner. Thus, the malfunction of original detection due to dirt of the original platen is eliminated in the case of expansion as well as reduction.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image reading device used in a copying machine or the like.

[Conventional technology]

A function that positions the image information in the center of the recording medium for a document placed in any position on the document table, a function that automatically determines the image magnification based on the dimensions of the document and the size of the recording medium, etc. In order to respond to multifunctional image processing, a document reading device is required that has a function of detecting the position and size of a document placed on a document table.

In such a function, it is necessary to prevent malfunctions in document recognition due to dirt on the document table or dust on the document table. The third way to prevent malfunctions in document recognition due to this dust is to
There are ideas shown in Figures 4 and 4.

In FIG. 3, S is a document table, M is a document, and N is, for example, dirt on the document table. Further, in a laser copying machine, for example, X is the scanning direction of the laser (hereinafter referred to as the main scanning direction), and Y is the moving direction of the optical system (hereinafter referred to as the sub-scanning direction).

When recognizing a manuscript like the one shown in Figure 3, the small square l in the figure
If a reading element such as a CCD line sensor reads an area in which all pixels in that area are white, it will recognize that there is a document in this area, as shown in Figure 3. When reading the original, the result is the fourth one.
A manuscript like the one shown in the figure will be recognized. In other words, according to this method, the stain N in FIG. 3 can be ignored.

In Figure 4, Kl (the point where the X coordinate is the minimum while the Y coordinate is the minimum), K2 (the point where the X coordinate is the maximum while the Y coordinate is the maximum), K3 (the point where the X coordinate is the minimum and The position and size of the document can be detected by determining and selecting four points: K4 (the point with the maximum Y coordinate among the points) and K4 (the point with the minimum Y coordinate among the maximum X coordinates).

That is, if nine pixels of 3×3 are considered as one unit area for document recognition, when each pixel is read in black and white as shown in FIG. 5 by the CCD, document recognition will be performed as shown in FIG. 6. If you increase the number of pixels in this unit area, the recognition accuracy will decrease, but you can remove large stains.If you decrease the number of pixels, the recognition accuracy will increase, but there is a risk that even small stains will be recognized as documents. There is sex. The number of pixels in such a unit area must be appropriately selected depending on how the device is used, the performance of the line sensor such as COD, etc.

[Problem that the invention is trying to solve] However,
The above detection method had the following drawbacks. This is generally achieved by changing the scanning speed of the optical system during enlargement or reduction, especially in the sub-scanning direction, and by changing the moving speed of the line sensor reading position.When the scanning speed in the sub-scanning direction changes, For example, when enlarging, the speed becomes slower, so if 9 pixels of 3 x 3 are considered as 1 unit area when the scanning speed is reduced to 2/3, 6 pixels of 3 x 2 will become 1 unit area when the scanning speed is reduced to 2/3. Become. In other words, when enlarging, the dirt on the document table is stretched out in the sub-scanning direction, and there is a risk that it will not be possible to skip the dirt. For example, if there is a stain like A in Figure 7 on the document table and it is divided into areas as shown in Figure 7 when magnified at 100% magnification, the stain will be skipped at 100% magnification, but when magnified, 2/3 of the stain will be read out by the optical system. The area is reduced in the sub-scanning direction Y as shown in Figure 8.
As a result, the dirt will be recognized as a document.

However, this drawback can be overcome by using separate optical systems for document recognition and document image reading, for example, by keeping the scanning speed constant regardless of enlargement or reduction during document recognition. However, a dedicated time is required for document recognition. Further, when it is necessary to perform document recognition and image reading at the same time, for example, when masking an area outside the document with white, especially when the document is placed diagonally, it is necessary to perform it in one scan. This recognizes the original area at the same time as reading the main scanning line, and then scans the next one.
When writing a line, it is necessary to output the area outside the document as white based on the document area information of the previous line. Furthermore, in the case where a page memory is not provided and each main scanning line is processed in real time, it is necessary to recognize the document area and read the image in one scan.

[Means for solving problems]

The present invention has been made in view of the above, and includes a means for reading image information placed on a document table while moving the reading position of a line sensor for reading line by line, and a means for reading the image information placed on the document table while moving the reading position, and a means for reading the image information placed on the document table while moving the reading position. It has means for dividing into unit areas of a predetermined size regardless of the size, means for determining whether or not all the image information in the divided unit areas is white, and means for detecting the coordinates of the document according to the determination result. The present invention provides a document reading device. Thereby, even if the moving speed of the reading position changes, such as during enlargement or reduction, it is possible to perform excellent document detection.

〔Example〕

Hereinafter, the present invention will be described in detail based on embodiments shown in the drawings.

FIG. 2 is a simplified configuration diagram of a document reading device to which the present invention is applicable.

A document placed with the reading side facing down on a document table 59 is illuminated with a fluorescent lamp 52, and a reflection mirror 53,551 optical lens 56
An original image is formed on a CCD sensor l via a CCD sensor l. The dotted line in the figure indicates the optical path of reflected light from the original.

An optical unit consisting of a fluorescent lamp 52 and reflective mirrors 53 and 55 is moved along a guide rail 58 by a DC motor (not shown) and scans an original on an original platen 59.

The CCD sensor 1 converts the strength of the reflected light from the original into an electric signal for each predetermined pixel, line by line, to read the original image.

When enlarging the image, the optical unit moves at a constant speed compared to when the image is magnified to the same size, and when reducing the image, the optical unit moves faster than when the image is magnified to the same size. Thereby, it is possible to achieve image reading at a desired magnification in the sub-scanning direction.

FIG. 1 is a circuit block diagram for processing image reading signals.

The image information on the document is captured by the CCD sensor 8-1 (see Figure 2).
After being read by the sensor (corresponding to sensor 1) and photoelectrically converted,
It is sent on line 58-1 to shift register 8-2. The shift register 8-2 is synchronized with the video clock 58-2 corresponding to the pixel and sequentially sends information for four consecutive pixels to the NOR circuit 8-3. Here, if the image information is "0" for white and "B" for black, the output of the NOR circuit 8-3 will be "B" if all the information of four consecutive pixels is white, and vice versa. If black information is included, it becomes "O".

Considering the case where the output of the OR circuit 8-9 is 1, if the information of four consecutive pixels is white, the gate 8-9
4 opens, and the main scanning address counter 8-5 is stored in the RAM 8-5.
1 is written at the address corresponding to the main scanning address sent from 6. In this way, image information for one main scanning line is written into RAM'8-5. Next, the same process is repeated for the next main scanning line, but in this case, the same information of the main scanning address of the previous line is output from Dout of the RAM 8-5, and the OR circuit 8
-9 to gate 8-4. In other words, RAM8-
The information input to 5 is "l" when four pixels in the main scanning direction are white and two pixels in the sub-scanning direction are white.

By repeating such an operation four times, information indicating whether or not all of the main and sub 4×4 pixels are white is written.

At this point, line 58-
7, "B" is output for one main scanning line, and the RAM
8-5 will be refreshed. This refresh timing determines the size of the unit area in the sub-scanning direction.

That is, a DC motor 8-1 for moving the optical unit.
The number of encoder pulses of 8 is proportional to the speed of the optical system, so if you refresh RAM 8-5 based on the encoder pulse of DC motor 8-I8, the speed of the optical system will change when changing the magnification. Even in such cases, the size of the original in the unit area does not change, and the original can be recognized without malfunction.

Further, the DC motor 8-18 is controlled by a motor driver 8-16, and the CPU 8-17 reads the encoder pulse of the DC motor 8-18, and the motor driver 8-16 applies feedback. In addition, the relationship between the number of encoder pulses and the moving distance of the optical system is, for example, about 0.05 mm per pulse at the same magnification, so 20 pulses per 1 mm. If you think about it, encoder pulse is line 58-2.
R is synchronized with the video clock and divided into 1/4.
It may be used as an AM8-5 refresh signal. This frequency dividing operation is performed by a frequency dividing circuit 8-19.

In this way, regardless of the moving speed of the optical unit, a unit area of a constant size can be used as the document presence/absence detection area. Therefore, even in the case of enlargement or reduction, it is possible to eliminate malfunctions in document detection due to dirt on the document table. In addition, the RAM output gate 8-8 is opened for each unit area, and the line 5
If all the signals are white through 8-8, a clock is output, and the main scanning and sub-scanning addresses at that time are latched into latches 8-10 and 8-11.

Incidentally, 8-12 to 8-15 are coordinate determination units, each of which ultimately determines the coordinates K l near the four corners of the document in FIG.
, K2, K3, and K4, and if document detection and image reading are performed separately, this information will be used as the basis for image reading. Also CPU8
-17 controls the motor driver 8-16 of the optical unit and can access document detection information. The main scanning address counter 8-6 is initialized by a video enable signal on line 58-4, and counts main scanning addresses in synchronization with the video clock on line 58-3. Also, the sub-scanning address counter 8-7 is on the line 88-5.
It is initialized by the original start signal of line 58-4.
The sub-scanning address is counted in synchronization with the video enable signal.

In this embodiment, the unit area in the sub-scanning direction is fixed, but for example, the operator can arbitrarily select the size of the unit area in the sub-scanning direction by selecting an arbitrary value. The unit area in the direction can also be configured to be arbitrarily selected by the operator. Also, by simple modification, it can be determined that there is no document when, for example, 2 out of 16 pixels are black.

In this way, for example, the operator can select the accuracy of document detection, and can realize the accuracy of document detection that is suitable for the device.

In addition to the configuration in which the optical unit moves, the present invention is also applicable to a configuration in which the original is passed through a fixed reading position at a speed matched to the reading magnification and the reading position is moved.

〔Effect of the invention〕

As explained above, according to this embodiment, even if the scanning speed of the optical system changes due to enlargement or reduction, the accuracy of document detection (malfunction due to dirt on the document table, etc.) is prevented from decreasing, and good document detection is achieved. It becomes possible.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of an image processing circuit to which the present invention is applied. FIG. 2 is a configuration diagram of an image reading device. FIG.
The figure shows the result of document detection in Figure 3, Figure 5 is a supplementary figure to Figure 3, Figure 6 shows the result of document detection in Figure 5, and Figures 7 and 8 are supplementary figures. FIG. 7 is a diagram showing that dirt is detected as a document by enlarging the unit area in the scanning direction. In the figure, 8-1 is a line sensor, 8-2 is a shift register, 8-17 is a CPU, 8-19 is a frequency dividing circuit, 8-6 is a main scanning address counter, and 8-7 is a sub-scanning address counter. . Figure 3 Figure 4 Figure 5 II l l 11 Figure 6

Claims (1)

[Claims] A means for reading image information placed on a manuscript table line by line while moving the reading position of a line sensor, and a means for dividing the read image information into unit areas of a predetermined size regardless of the moving speed of the reading position. and means for determining whether all image information in the divided unit area is white.
A document reading device comprising means for detecting the coordinates of a document according to the determination result.