JPS6367862A - Eimage input device - Google Patents

Abstract

PURPOSE:To reproduce image information faithful to an original image even by manual scanning operation by finding scanning picture element positions on each scan and writing image information of scanned picture elements in corresponding addresses of a storage part. CONSTITUTION:For example, when a character image on a form placed on a digitizer tablet 10 is read by moving an image reader 12 by a hand, the image reader 12 outputs image information obtained by decomposing the image into picture elements. Induced signals generated at exciting coils 14 and 16 provided to the image reader 12 are led to a table controller 18 to detect coordinates on the tablet 10 which are indicated by the exciting coils 14 and 16. A processor 20 finds picture element positions on the image from the indicated coordinates of the exciting coils 14 and 16 by main scans and store image information of corresponding picture elements in addresses of the memory 22 corresponding to the picture element positions. Consequently, a character image pattern in the same shape with the original image is stored in the memory 22 regardless of movement variation such as the snaking of the image reader 12.

Description

[Detailed Description of the Invention] <Industrial Application Field> The present invention relates to an image input device, and particularly to a hand belt scanner that reads image information of an image by manually moving an image reader over the image. The present invention relates to a type of image input device.

<Prior art> A hand belt scanner, which is one of the image input devices, is
It is an easy-to-operate input device that allows you to input a desired image such as a figure while holding the image reader in your hand and moving it across the surface of the paper.

Usually, the image reader of a hand belt scanner has a linear image sensor (referred to as a line image sensor) using a CCD or the like. performs sub-scanning and reads the still image into pixels while outputting image information.

The output of this image reader is input to a computer and subjected to predetermined processing.

<Problems to be Solved by the Invention> However, in this handbelt scanner, sub-scanning for image reading is performed manually, so fluctuations in manual feeding speed (sub-scanning speed) and manual feeding direction (sub-scanning) may occur. As a result, image information faithful to the original image cannot be obtained due to unavoidable bending fluctuations (direction). For example, when attempting to read a character image and perform character recognition, the character pattern after reading may be deformed, making it difficult to reduce the character pattern size. Even after repair, the curved deformation remains, resulting in a high rate of misrecognition and unrecognizability of characters.

The present invention was made in view of the drawbacks of the prior art,
To provide an image input device capable of reproducing image information faithful to an original image even in manual scanning.

<Means for Solving the Problems> FIG. 1 is a block diagram showing a system for inputting image information such as characters according to an embodiment of the present invention.

10 is a digitizer tablet on which an image is placed; 12 is an image reader that decomposes the image into pixels and outputs image information; 14 and 16 are two excitation coils for specifying coordinates integrated into the image reader; 18 is an image reader that outputs image information; A tablet controller 20 outputs the coordinate data specified by the excitation coils 14 and 16, and a processor (CPU) 20 organizes and stores image information in the RAM 22.

Function> When reading, for example, a character image on a sheet of paper placed on the digitizer tablet 10, while moving the image reader 12 by hand, the image reader 12 repeats main scanning and sub-scans the image as it moves. Outputs image information by decomposing the image into pixels.

Meanwhile, during image reading, the induction signals generated in the two excitation coils provided in the image reader 12 are individually sensed by the digitizer tablet 10, and guided to the tablet controller 18 to activate each excitation coil 14,16. The specified coordinates on the digitizer tablet 10 are sequentially detected according to the position of the image reader 12.

Here, the image information related to one main scan read by the image reader 12 at a certain point during movement uniquely corresponds to the scanning position on the image from the indicated coordinates of the two excitation coils 14 and 16 at that point. Can be attached.

On each main scanning side, the processor 20 calculates a pixel position on the image corresponding to the main scanning by performing a predetermined calculation on the indicated coordinates of the two excitation coils 14 and 16 at the time of performing the main scanning, and calculates the pixel position on the image corresponding to the main scanning. The image information of the pixel is stored at the corresponding address of the memory 22.

As a result, a character image pattern having the same shape as the original image is stored in the memory 22 regardless of movement fluctuations such as bending of the image reader 12.

<Example> An example of the present invention will be described below with reference to FIGS. 1 to 4.

FIG. 1 is a block diagram showing an image information input system according to the present invention, which includes a digitizer tablet 10 on which an image is placed, and an image reader 12 that scans the image while manually moving it and outputs image information. And Otsu's image reader 12
The first and second digital quizzes are provided at both ends of the
The DigiQuiza tablet 10 is directed by the excitation coil 14.16 and the first and second excitation coils 14.16.
Tablet controller 1 that detects the above two coordinate values
8, and a computer 19 for image processing, for example, text 31fi.

A grid coil 30 for the X coordinate and a grid coil 32 for the Y coordinate are embedded in the digital and isa tablet 10. These grid coils 30 , 321 are connected to an analog multiplexer 34 , 36 which is controlled by the tablet controller 18 .

The image swatch reading unit 12 includes a CCD 38 as a line image sensor in which a large number of elements are arranged one-dimensionally, a CCD driver 40 that drives the C0D 38 and repeats scanning (main scanning) at a predetermined period, and the C0D 38. an AD converter 44 that binarizes the output of this amplifier at a predetermined threshold value and converts it into white or black image information (image data);
, and coil drivers 46, 48 that alternately drive the second excitation coils 14, 16 at a predetermined frequency.

The tablet controller 18 has a microcomputer configuration, and controls the switching of the coil drivers 46 and 48 as well as the analog multiplexers 34 and 36, and controls the output of each AD converter 50 and 52. , the indicated coordinates P on the digitizer tablet 1° indicated by the second excitation coil 14.16, (X# y) p
Q, (xp y) is detected. Figure 4 [al is a switching signal to switch the coil driver 46, 48, Figure 4 (
b) (This is a waveform diagram of a switching signal that switches each analog multiplexer 34, 36, and the first
The excitation coil 14 is excited, and the second excitation coil 16 is excited during the period T2.

The analog multiplexer 3 detects the X coordinate value of the excitation coil 14 during the period Tlx of the period T1.
A regrid select signal is applied from the tablet controller 18 to the excitation coil 14 during the period T17.
A grid select signal is applied to an analog multiplexer 36 that detects the Y coordinate value of . In addition, a tablet controller is connected to the analog multiplexer 34 that detects the X coordinate value of the excitation coil 16 during the period T2 of the period T2.
A grid select signal is applied from the roller 18, and T2
A grid select signal is applied to the analog multiplexer 36 which detects the Y coordinate value of the excitation coil 16 during the Y period.

The computer 19 has an interface 54 and a ROM 5.
6, RAM 22, processor 20, etc. are connected by bus. The ROM 56 stores various programs for causing the processor 20 to perform image processing, such as character recognition processing, and according to these programs, the processor 20 processes image information input from the AD converter 44 via the interface 54. Coordinate information p, (Xy y) p Q, (X# y
) is assigned an address and stored in a predetermined area of the RAM 22, and then feature extraction, matching, etc. are performed to perform character recognition.

Next, the overall operation of the above embodiment will be explained.

When the image reader 12 is placed at the beginning of the desired character reading area on the character image placed on the digitizer tablet 10, the first and second excitation coils 14 and 16 are activated at a predetermined period as shown in FIG. be energized alternately. First excitation coil 1
When the first excitation coil 14 is excited, the induction signal generated by the excitation coil is sensed by the grid coil 30.32 which is scanned by the analog multiplexer 34.36. 34, analog multiplexer 36 for Y coordinate value detection
, TABLEF 1 - CONTROLLER 1 via AD converter 50
8, the peak of the induced voltage is estimated using a predetermined method by this tablet controller 18, and the χ, Y coordinate values j (x, y) on the digitizer tablet 10 where the excitation coil 14 is located are )(i=o) is detected, and the 'Jb8 voltage is similarly input to the tablet controller 18 via the analog multiplexer 34, 36 and the AD converter 52 for the excitation m coil 16 of measure 2, and the 'Jb8 voltage is inputted onto the digitizer tablet 1o. The coordinate value Q1 (
X,y)(1-o) is detected.

Next, when the image reader 12 moves over the decimal tablet 10, the tablet controller 18 repeatedly excites the image reader 12 at a predetermined period by applying the first and second excitation to the coil 1.
Since the indicated coordinates of 4.16 are detected, P, (X, y) and J(
Xyy) are sequentially output to the computer 19 (see FIG. 2).

On the other hand, regarding images, when the image reader 12 is placed at the beginning of the desired character reading area on the character image placed on the digitizer tablet 10, the CCD 38 decomposes the character image into pixels in the main scanning direction and generates an image signal. After being output and amplified by an amplifier, the dot string signal is digitized by an AD converter 44 and sent to the computer 19.

Then, when the image reader 12 moves over the digitizer cubelet 10, since the C0D 38 repeatedly performs main scanning at a predetermined period, the image scanning signal IMSi changes depending on the position 1 of the image reader 12. are sequentially output to the computer 19.

Now, the processor 20 on the computer 19 side processes the first Po output from the tablet controller 18 after the image reader 12 is placed on the digitizer tablet 10.
(x, y) and Qo (x, y) (see FIG. 3) to determine each element AI, A2 . ...
Pixel position X on the image corresponding to the n-th element An of AN. , Y, is calculated using the following formula % (11). However, N is the number of image reading elements in the CCD (for example, 1024), ("PC+'ypo)"
are the X and Y coordinate values of the position P0 (x, y) of the first excitation coil, (x, ., yQo) are the position Q0 (X
, y).

Then, when the image reader 12 inputs the image scanning signal IMSO at the position i-0 to the computer 19,
The processor 20 stores the n-th dot data (n=1.2...N) in the image scanning signal IMSO in the RAM 22 (]) and the X, Y address (x, . . . yn).

Then, after a certain period of time, the tablet controller 18 outputs P, (x) as the position coordinate value of the image reader 12.

y) and Q s (xp y ) are input, the processor 20 of the computer 19 executes calculations similar to equations (1) and (2) to obtain the pixel positions Xn, Y, on the image, Thereafter, each dot data in the image scanning signal IMSI when the image reader 12 is at the position i=1 is obtained by the above calculation in the RAM 22 and sequentially stored in the i:x,y address.

Thereafter, while the image reader 12 is moving, P,
(Xp y) p Q, (XP y) ('=1p
2p "p...) position data input, X and Y address calculations according to equations (1) and (2), input of image scanning signal IMSi, and storage in the X and Y addresses of the RAM 22 are repeated.

As a result, dot pattern data is accurately stored in the RAM 22 regardless of the degree of curvature of the in-image sensor.

In this way, images can be freely input, and if necessary, the processor 20 performs subsequent character recognition processing such as compression and feature extraction according to known methods.

In the above embodiment, a digitizer using an electromagnetic induction method was used, but other methods such as a magnetic field phase method may be used to
, lxp Y) p Q, CXp '/) (of course).

<Effects of the Invention> As explained above, according to the present invention, the original image can be faithfully reproduced and image information can be obtained. There is no reading error (even if the image is scanned), and high-precision image input is possible with simple manual operations.For this reason, character input, which was difficult in the past, can be performed easily.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing a system for inputting image information such as characters according to an embodiment of the present invention, FIG. 2 is an explanatory diagram showing a movement path of an image reader, and FIG. 3 is an illustration of an image of each element of the COD. FIG. 4 is a time chart showing the timing of switching control of each coil driver and multiplexer. 10... Digitizer tablet, 12... Image reader, 14. 16... Excitation coil, 18... Tablet controller, 19... Computer, 20... Processor, 22... RAM

Claims (1)

[Claims] An image reading unit that sequentially scans the image while manually moving the image and outputs image information, a digitizer bullet on which the image is placed, and coordinate instructions for two digitizers provided in the image reading unit. a coordinate detection unit that detects the designated coordinates on the digitizer tablet designated by the coordinate designation unit; a storage unit that stores image information; What is claimed is: 1. An image input device comprising: an image information writing section for determining the position of a scanning pixel on an image from the image data and writing image information of the scanning pixel at a corresponding address in a storage section.