JPS62161273A - Original detection device - Google Patents

Abstract

PURPOSE:To attain the detection of a tip oof an original area by providing an original area tip detection means detecting it as the tip reference of the original area for a position when density information changes from a while level into a black level and then changes into a white level again. CONSTITUTION:The tip 7b of an original scale is discriminated by using a black pattern 7c provided at the rear side of an original scale 7. In scanning the original scale 7 in the sub scanning direction shown in the arrow and when a CCD 20 crosses the original pattern 7, the output signal of the CCD 20 goes to a white into a black level and then a white level again, the position is the boundary of the black pattern 7c of the original 6. Thus, in adding a distance till the original scale tip 7b know in advance to the position (reference point), the position of the original scale tip (original tip) 7b is decided and the scanning of the original area is started from said position.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Industrial Application) The present invention relates to a document reading device capable of detecting the position of a document when reading the document with an imaging probe.

(Prior Art) In a document reading device using an imaging probe, a document to be read is placed on a document glass, and is further covered with a document cover. The original is illuminated by an exposure source, and its density is detected by a one-dimensional image sensor (
main scan). The image signals are sequentially binarized and then stored in a storage device. The image of the entire manuscript is read by sequentially scanning (sub-scanning) the Kyoto surface.

(Problems to be Solved by the Invention) When sub-scanning a document surface with a one-dimensional image sensor, it is necessary to determine the starting point of scanning. Conventionally, 1. In the r2 example, a document edge detection switch is provided at the tip of the area where the document can exist (document area), and this switch is activated by the r movement of the scanning optical system, etc., and detects the 7.2 points. Use f as the starting point for sub-scanning. Alternatively, a fixed position switch is provided, and the starting point of the sub-scanning is a point a predetermined distance from the point at which the switch is closed. In these methods, the position of the switch must be determined with high precision.

Furthermore, in the image reading device disclosed in Japanese Patent Publication No. 60-28182, a marking stamp material that is capable of f hyperactivity and is colored to have a predetermined reflection density is placed along the edge of the document glass. Two are provided in the sub-scanning direction of the document. When reading an image, marking portions are set at the start and end points of sub-scanning, and an image corresponding to the marking portion is detected to control the start and end of image reading. This method requires a mechanism for attaching the marking part,
Further, it is necessary to adjust the position of the marking member each time.

In Japanese Patent Application No. 60-85031, the present applicant has disclosed an image reading apparatus which detects the edge of a portion of a document containing information.

An object of the present invention is to provide an image reading device that can detect the leading edge of a document area using a method applicable to such image reading devices.

(Means for Solving the Invention) An image reading device according to the present invention includes a density detection means for detecting the density of an original on an original glass by an image sensor, and a scanning device for sub-scanning the original on the original glass by the density detection means. a 5-variable storage means for storing the density detected by the density detection means; and a plate having a black pattern of a predetermined density provided at the end of the document placement standard on the scanning start side of the document glass;
The black pattern is detected by the density detection means during the sub-scanning, and the document area leading edge detection means detects the position where the density information changes from the white level to the black level and then changes back to the white level as the leading edge reference of the document area. It is characterized by comprising:

(Function) When performing sub-scanning across the temporarily provided black pattern, the point where the density level detected by the image pickup device changes from white to black is detected as the reference point at the leading edge of the document area.

(Embodiments) Hereinafter, embodiments of the present invention will be described in the following order with reference to the accompanying drawings.

a: Image reader b, internal structure of the image reader C, document scale tip detection d, image information reading flow e Document detection (a) Image reader FIG. The exposure source is the Haloken lamp 2; the original 6 placed on the original glass 4 is irradiated.The original 6 is placed along the original slider provided at the edge of the original glass The Haloken lamp 2 is equipped with a reflector 8, an infrared light source, and a filter IO.

1! On the back side of the 71 Hanawa scale 7, various patterns are f
As shown in Figure 1, the two halves of the manuscript scale 7 are white, and are used precisely for kneading. ,
The black pattern 7c, which has a width of -100 and 4f, is missing. This black pattern 7c is used to detect the tip of the document scale (the tip of the document area). In this embodiment, black patterns 7a, 7a are also provided at both ends, and these black patterns 7a, 7a are used to determine the magnification from the measurement of the distance between the two.

The reflected light from the original 6 is transmitted to the first mirror I2 and the second mirror I.
4 and third mirror I6, the light passes through a lens 18 and enters a one-dimensional CCD (imaging device) 20.

The CCD 20 is held by a COD holding section 22, and its position and angle are adjusted. Further, the CCD holding section 22 and the lens 18 are attached to a moving table 24 as shown in FIG.

The magnification can be adjusted using the movable mechanism (not shown) on movable platforms 2 and 1.
This is done by moving in the direction of the optical axis.

Focus adjustment is carried out by moving the CCD fs marking section 22 in the direction of the optical axis using a motor (not shown) mounted on the B moving table 2.1.

When scanning one draft 6, as is well known, the light source 2 and the mirror 12 . +4. IG is moved t< in the scanning direction.

(b) Internal structure of the image reader FIG. 3 is a block diagram of the circuit for detecting the density of the original.

The tarock generation circuit 40 generates the S required for the image sensor 20.
It provides an H (Sample Hold) signal, is also connected to the CPU 42, and is used as a clock signal.

The image sensor 20 converts an optical signal into an electrical signal. Also,
Its output level is controlled by a signal from the CPU 42. The A/D converter 44 converts the analog output of the image sensor 20 into a digital signal. The conversion level is C
It is controlled by a signal from PU42. The shading circuit 46j is used to correct the unevenness of light in the main scanning direction and the variation among image sensors, so the CPU 12
The timing is given by The output of shading circuit 46 is input to comparator circuit 48 and line R to M50. The comparison circuit 48 compares the image signal corrected by the shape ink circuit 46 and the signal selected by the selector 52, and outputs the result.

The output circuit 5.1 outputs one image's image information and effective image signal (synchronization signal) to the outside. Output this. Line R,=
＼〜150 is nee de ink! +li Correctness Itl This signal is stored in the memory for one scan. This slow signal is C
It is output from the PU 12, and the CPU 112 obtains -line worth of image information by referring to the lines RA to 150. The attribute information stored in the RAM 56 is
The information written on line n A M 50 is determined by the attribute of 2 or dither or CI) U = 12.
During data transfer (scanning), the selector 52 is changed based on this attribute and output to the comparison circuit 48.

The pattern generation circuit 58 generates a threshold value at the time of dither selection, and the threshold value is generated by 7 trixes of (mXn).

The selector 52 selects threshold data to be sent to the comparison circuit 18 from the threshold information and attribute information from the pattern generation circuit 58 and RAM 56. If the attribute is dither, the data from the pattern generation circuit 5g is input to R. If the attribute is binary, R
Threshold information from AM56 is sent to comparison circuit 48. CPU
42 controls the entire system from the above signals, motor signals, lamp signals, fixed position signals, etc., and command signals.

(c) Original scale tip detection The tip 7b of the original scale 7 is determined using a black pattern 7c provided on the back side of the corrupted manuscript scale 7, as shown in FIG. Now, the original scale 7 is scanned in the sub-scanning direction indicated by the arrow and detected by the CCD 20. When crossing the original pattern 7, the position where the output signal of the C0D 20 changes from white to black and then becomes white again is the boundary of the black pattern 7C on the original 6 side. Therefore, by adding a previously known distance to the document scale tip 7b to this position (reference point), the position of the document scale tip (that is, document tip) 7b is determined. Therefore, scanning the imperial manuscript area should start from this position. In order to reliably detect the change in density level from white to black to white, white level portions are provided on both sides of the black pattern 7c.

The draft scale 7 is provided with magnification detection patterns 7a, 7a, pins), adjustment patterns (not shown), and the like.

FIG. 4 shows the flow of document scale tip detection. Scanning of the original scale 7 in the sub-scanning direction is started. And ■
A degree information of the element (or elements may be plural) at a position where the black pattern 7C of the dimensional CCD 20 can be detected is always detected. First, it is determined whether the density information of this element is at the white level or not, and if not, the process waits until it reaches the white level (step P25). Next, it is determined whether or not the a degree information of this element has reached the black level, and if no, it waits until rj reaches the black level (step P26). Next, it is determined whether the density information of this element has reached the white level again, and if not, it waits until it reaches the white level (Step I] 27).
Next, the distance between the white level point (reference point) and the tip of the original scale, which is known in advance, is determined.

((j) The flow of image information f (5 reading flow) is shown in Figure 1 below. 4 - According to the flowchart,
)1 Yue Ming”-ro.

First, a command signal is input (step P1).

It is determined whether this command signal is a start command (step P2). If the start command has been input, the process moves to the following scanning operation. Note that the magnification of the scanning system and the like are set according to each usage mode.

First, turn on the lamp for exposure (step PIO
). Next, it is determined whether the scanning system (slider) has returned to its home position (step pH).

If it has returned, the process advances to step P20. If not, the motor is driven in the return direction to return to the home position (step P12), and it is determined whether the scanning system has returned to the home position (step PI3). If not, return to step PI2. If it has returned, stop the motor (step P
14). Next, the level of C0D20 is adjusted (step P15).

Next, a noeding process is performed. Since the slider is at the reference level, the digital ink signal is turned on (step P20), and it is determined whether a sample and hold (SH) signal, which is a synchronizing signal for the output of the CCD 20, is input (step P2+).

SI (Wait for the signal to be input manually, turn off the navigation signal, and start preliminary scanning for document detection (step P22).

First, the tip 7b of the original scale is detected (step P
23. (see section (c)). Next, in step P30, S
It is determined whether the H signal is manually input or not.

When the S H signal is input, the current output of C0D20, which has been subjected to shading correction, is written to lines R and AM50 (step P31). It is determined whether or not the next S H signal has been input (step P32), and when the next S H signal has been input, the write signal of the line RAM 50 is turned off (step P33). Next, the position of the document is detected from the data written in the line RAM 50 (step P34).
). This detection will be explained in detail later. next,
Based on the data (density distribution, density peak, etc.) written in the line RAM 50, the attribute of the data processing, ie, whether it is binary or dither, is determined and written to the selector 52 (step P35). Next, it is determined whether the scan is completed (step P36). If not, the process returns to step P30. Next, step P37 of driving the motor in the return direction to return the scanning system to the home position.
), it is determined whether the scanning system has returned to its home position (step P38). If not, the process returns to step P37.

Next, perform the main scan. The motor is driven in the scanning direction, and scanning for data output is started (step P40). The tip 7b of the document scale is detected (step P41. See section (c)). Next, wait until the leading edge of the document reaches the position measured during scanning or preliminary scanning (step P42). Note that the size of the margin is determined based on the detected size and resistance of the document, and the paper feed timing or synchronization signal timing of the printer is changed. When the leading edge of the draft is reached, turn on the effective image signal in step P43.
), the output signal is used as an aid to activate the output device (printer).

Rear end of the crown (step P.1 to determine whether it has reached the device or not)
1) When the trailing edge of the draft is reached, that is, when scanning is completed, the enhancement image signal is turned off and the output signal is invalidated (step P45).

Next, clarify the lamp for exposure (step P50
). Step P51 to drive the motor in the return direction.
), it is determined whether the scanning system has returned to its home position (step P52). If it has returned, the motor is stopped (step P53). Then, return to step P1 and prepare for the next start command.

(e) Document Detection Next, the flow for detecting the position of the text/figure information area in the document from the data stored in the line RAM 50 during preliminary scanning (step P34) will be described with reference to the example in FIG. This will be explained with reference to the flowchart shown in FIG. The position of the edge of the document is detected at the point where the white level changes to the black level.

In the example shown in FIG. 1, the Roman letter A is written in large letters on the manuscript 6. In this embodiment, the text/graphic information area in the document is detected as a rectangular area surrounding the text/graphic information, as indicated by a dotted line. Here, dashed line a (document edge 1) and dashed line b (document edge 2) are the leftmost point A and the rightmost point A2 of character A in the direction perpendicular to the main scanning direction.
These are the lines that are tangent to each. Also, the broken line C (document leading edge position) and the broken line d (document trailing edge position) are the uppermost point A3 and the lowermost point A, A2 of the character A in the direction parallel to the main scanning direction.
These are the lines that are tangent to each. Note that the character/graphic information area boundaries a to d are parallel or perpendicular to the main scanning direction because they are detected in the main scanning direction as described later.

Detection of information on a document by COD is performed by sequentially moving the scanning position in the front and back direction of the document (sub-scanning). The document detection flow is performed once for each scan at each scan position.

First, a pointer for reading image sensor information is placed at a position corresponding to the tip (left end) of the image sensor (step P6
0). Initial values (the leading edge and trailing edge of the image sensor) are also set for document edge 1 and document edge 2.

Next, the image sensor information pointed to by the pointer is extracted (step P61). It is determined whether the image sensor information is a black level (step P62). If the black level is
Since there is no content of the original, the pointer is updated to the next position (step P63). (The pointer sequentially moves from left to right.) Next, it is determined whether the pointer has exceeded the document edge 2 (step P64), and if not, step P61 +: Return.

If it exceeds the limit, it is determined that there is no content in the manuscript. Next, it is determined whether or not there was a document in the previous document detection (step P65), and if there is, the current scan position is set as the trailing edge position of the document (step P66), and the process returns. If not, simply return.

When it is determined in step P62 that the image sensor information pointed to by the pointer is not at the black level, it is then determined whether or not it is at the white level (step P70).

If it is the white level, the pointer is updated to the next position (step P71). Next, it is determined whether the pointer has exceeded the edge 2 of the document, and if not, the process returns to step P70.

If it exceeds the black level, it is determined that there is no document content, and the process proceeds to step P65, where the process continues until the black level is reached.

If it is determined that the white level is not at step P70, that is, if a change to the black level is detected, then it is determined whether the original edge I is smaller than the previous position of the document edge I (whether it is on the left side or not). (Step P80).

If so, the pointer position at that time (the position where black was detected) is set as the document edge l from this time onwards, and if not, the process continues to the next step. Through the above steps, the document edge 1 is determined.

Next, document edge 2 is determined. First, the pointer is placed at a position corresponding to the rear end of the image sensor (step P90).

Next, the image sensor information not indicated by the pointer is extracted (step P91). It is determined whether the image sensor information is a black level (step P92). If it is the black level, there is no content in the document, so update the pointer to the next position (
Step P93). Next, it is determined whether the position of the pointer is smaller than the position of document edge 1 (step P911).
), if not, the process returns to step P91. Small i,
If the content of the document is J', ζ, -, 1-), the process proceeds to step P65. If it is determined in step P92 that the image sensor information pointed to by the pointer is not at the black level, Next, it is determined whether the white level is reached (step P100). If it is the white level, the pointer is updated to the next position (step PIOI). Next, it is determined whether the pointer is smaller than document edge 1 (step PI
02), otherwise the process returns to step P100. If it exceeds, it is determined that there is no document content, and the process advances to step P65.

If it is determined in step P100 that the white level is not high, that is, if the black level is detected, then it is determined whether the position of the pointer is smaller than the position of the document edge 2, and if so, the pointer is (the position f where blackness is detected) is set as document edge 2, and if not, the process continues to the next step. With the above steps, the document edge 2 is determined.

Next, when the previous document was detected, it is determined whether the draft is in the middle or not (progressive pH2). When the original is detected for the first time, the current scanning position No. 1 is set as the leading edge of the prefectural manuscript (step PI+3 if left) and the process returns. If not, simply return.

(Effects of the invention) One piece that has been used in the past [the tip of the draft is unnecessarily tattered. The sub-scanning of the Kyoto manuscript ('J,, 7 The pattern provided on the back side of the manuscript scale, which is the active part of the manuscript placement, starts from the temporary tip, so the positional accuracy of the manuscript scale is 1.

[Brief explanation of drawings]

FIG. 1 is a diagram showing the positions of character areas to be detected in a draft. FIG. 2 is a sectional view of the image reader. FIG. 3 is a block diagram of the internal configuration of the image reader. Figure 1 shows the flowchart 1 of detecting the tip of the draft scale.
It is. FIG. 5 is a flowchart for reading the shrine. Figure 6 of Kaya is a flowchart of draft detection. 2...Light source, 6...Manuscript, 7.Document scale, 7a, 7a...Black pattern, 7b., Tip of manuscript scale, 7c...Black pattern, 20.Photographing tide.

Claims (1)

[Claims] (1) Density detection means for detecting the density of the original on the original glass using an image sensor; scanning means for sub-scanning the original on the original glass by the density detection means; and density for storing the density detected by the density detection means. a storage means; a plate having a black pattern of a predetermined density provided at a document placement reference end on the scanning start side of the document glass; 1. A document detection device comprising document area leading edge detection means for detecting a position where a level changes from a black level to a white level again as a reference to the leading edge of the document area.