JP2748825B2 - Image processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus,
In particular, the present invention relates to an image processing device suitable for image recognition and character recognition.

[0002]

2. Description of the Related Art Generally, in an image processing apparatus, a plurality of CCDs (Charge Couples) arranged two-dimensionally are used.
The image processing data is obtained by photoelectrically converting a grayscale image such as a character, a figure, or a photograph using a solid-state imaging device such as an ed device.

[0003] However, CCDs have variations in sensitivity due to manufacturing problems. Therefore, if image processing is performed using photoelectric conversion signals as they are, erroneous processing results may be obtained.

Therefore, in order to correct variations in the performance of individual CCDs, for example, Japanese Patent Application Laid-Open No. 63-132579 discloses a method of correcting the sensitivity between pixels by using a previously stored sensitivity correction value for each pixel. To compensate for variations,
In Japanese Patent Application Laid-Open No. 2-72774, non-uniformity of an image signal due to variation in dark voltage of each light receiving element is corrected by driving with a different gate voltage for each frame period.
Further, in Japanese Patent Application Laid-Open No. Hei 2-303276, high-frequency distortion is corrected by comparing adjacent pixel data.

[0005] On the other hand, as the image processing apparatus is used for various purposes, a demand for an improvement in image processing speed has increased.

In order to transfer electric charges at high speed, Japanese Patent Application Laid-Open No. 3-123181 discloses a method of transferring electric charges transferred by a plurality of vertical transfer means arranged in a vertical direction to a plurality of horizontal transfer means provided in a horizontal direction. CCD to transfer to transfer means at the same time
In the above, the electric charges are simultaneously read from two adjacent pixels arranged in the vertical direction of the vertical transfer means, and the electric charges are transferred to the corresponding horizontal transfer means for each pixel.

That is, the grayscale information from the CCD is simultaneously output in a double readout system in which a shift register of two systems, that is, a shift register for an even-numbered pixel column and a shift register for an odd-numbered pixel column is used.

[0008]

However, in the conventional example disclosed in Japanese Patent Application Laid-Open No. 3-123181, there are two shift registers for outputting density information from the CCD, that is, a shift register for even-numbered pixel columns. Because it is divided into a register and a shift register for odd pixel columns,
Variations occur in the output of shading information between the shift register for the even-numbered pixel column and the shift register for the odd-numbered pixel column for the same image due to a temperature change or the like, which has the disadvantage of adversely affecting the quality of the image processing result. .

[0009]

SUMMARY OF THE INVENTION An object of the present invention is to improve the disadvantages of the prior art, and in particular, to improve the quality of image processing results due to the fact that a shift register for outputting grayscale information from an image sensor is divided into two systems. An object of the present invention is to provide an image processing apparatus capable of preventing deterioration.

[0010]

Therefore, according to the present invention, there is provided an image sensor for photoelectrically converting a gray-scale image such as a character, a figure, or a photograph, and gray- scale information output from the image sensor.
A / D converter for converting to a digital signal, and the A / D converter
Image processing based on the digital signal output from the converter
And a processing unit for performing the processing. In addition, the processing unit
Of grayscale information input from an even pixel row
And store the image signal as even pixel column information
Digital grayscale information input from the odd pixel array of the sensor
Image processing mask for storing signals as odd pixel column information
And the pixel to be processed by the processing unit
In some cases, the digital signal of the even pixel row of the image processing mask
Output to the processing unit and image processing target by the processing unit
If the pixel is odd pixel column information, the image processing mask
Output control to output digital signal of odd pixel column to processing unit
And a means are provided . This aims to achieve the above-mentioned object.

[0011]

The image sensor converts a grayscale image such as a character, a graphic, or a photograph into an analog electric signal for each pixel.

Then, the grayscale information from the even-numbered pixel column of the image sensor is read and input to the AD converter.

Further, the grayscale information from the odd-numbered pixel column of the image sensor is read and input to the AD converter .

The AD converter converts the grayscale information of the even-numbered pixel row into a digital signal and outputs the digital signal to an image processing mask.

[0015] Further, the AD converter is capable of shading the odd-numbered pixel columns.
The information is converted to a digital signal and output to an image processing mask.

The image processing mask stores the even-numbered pixel row density information and the odd-numbered pixel row density information from the AD converter.

The output control means selects and outputs the even-numbered pixel row shading information from the image processing mask if the pixel to be image-processed is in the even-numbered pixel row, and outputs from the image processing mask if the pixel to be image-processed is in the odd-numbered pixel row. The odd-numbered pixel row density information is selectively output.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to FIGS.

The embodiment of FIG. 1 shows an image sensor 5 in which a plurality of two-dimensionally arranged CCDs for photoelectrically converting a grayscale image such as a character, a figure or a photograph, and grayscale information from an even-numbered pixel array of the image sensor 5 are provided. , An odd pixel row density information output section 7 for reading density information from an odd pixel row of the image sensor 5, and a correction for the gray level information from the even pixel row density information output section 6. And a first correction unit 8 for performing the correction.

A second correction means 9 for correcting the grayscale information from the odd-numbered pixel row grayscale information output means 7 and a first AD conversion for converting an analog signal from the first correction means 8 into a digital signal. AD converter 1 for converting an analog signal from the second correcting means 9 into a digital signal
1 and an image processing mask 3 for storing the even-numbered pixel row density information from the first AD converter 10 and the odd-numbered pixel row density information from the second AD converter 11.

Further, in the present embodiment, the density information control means 4 for storing and controlling the density information of the image sensor 5 in the image processing mask 3 for each pixel row, and the density information of the even pixel row or the odd pixel from the image processing mask 3. The image processing apparatus includes an output control means 1 for selecting and outputting density information of a column and a processing means 2 for performing image processing based on density information from the image processing mask 3.

Next, the operation of this embodiment will be described.
Here, it is assumed that the image sensor 5 is composed of m columns × n rows of CCDs (C11 to Cmn) as shown in FIG.

[0023] When a gray-scale image such as a character, a figure, or a photograph is input to the image sensor 5, each CCD converts an analog electric signal proportional to the gray scale of the input image, that is, density information.

[0024] The density information control means 4 outputs the density information from the CCDs in the even-numbered pixel rows of the image sensor 5 to the even-numbered pixel row density information output means 6 and the density information from the CCDs in the odd-numbered pixel rows of the image sensor 5 to the odd-numbered pixel rows. Output to the density information output means 7 simultaneously.

[0025] The even-numbered-pixel-row density information output means 6 outputs the first-correction means 8 the density information from the CCD in the even-numbered pixel row of the image sensor 5. Here, as shown in FIG. 2, first, the CCD at the head position of the second column,
That is, from the density information of C12, C22, C32,.
Are output in that order.

The odd pixel row density information output means 7 outputs the density information from the CCD in the odd pixel row of the image sensor 5 to the second correction means 9. Here, as shown in FIG. 2, first, the CCD at the head position of the first column, that is, C21, C31,...

[0027] The first correction unit 8 corrects a variation in sensitivity of each pixel by synchronizing a signal from the even-numbered pixel column density information output unit 6 with correction information for each CCD stored in advance. That is, the correction is performed so that “true white” becomes 0 [V] and “true black” becomes [2] V.

The second correction means 9 corrects the sensitivity variation in each pixel by synchronizing the signal from the odd-numbered pixel row density information output means 7 with correction information for each CCD stored in advance. That is, the correction is performed so that “true white” becomes 0 [V] and “true black” becomes 2 [V].

[0029] The first AD converter 10 converts an analog signal from the first correction unit 8 into a digital signal.
That is, the analog input of 0 to 2 [V] is changed from 00H to FF.
Convert to H digital signal. Here, 00H means “true white” and FFH means “true black”. That is, the density is decomposed into 256 gradations.

The second AD converter 11 converts an analog signal from the second correction means 9 into a digital signal. That is, the analog input of 0 to 2 V is converted into a digital signal of 00H to FFH. Here, 00H means “true white” and FFH means “true black”. That is, the density is decomposed into 256 gradations.

[0031] The density information control means 4 stores the signal from the first AD converter 10 at a predetermined even column position of the image processing mask 3, and stores the signal from the second AD converter 11 in a predetermined position of the image processing mask 3. Store at odd column position. That is, as shown in FIG. 3, the signal from the CCD of Cij is S
Stored at the location of ij.

[0032] The output control means 1 selects and outputs, from the image processing mask 3, the density information of the even-numbered pixel to the processing means 2 if the pixel to be processed is in the even-numbered pixel row, and outputs the image processing mask if the pixel to be processed is in the odd-numbered pixel row. 3 to output the odd-numbered pixel row density information to the processing means 2. Therefore, this implementation
In the example, when performing image processing, the same
Image processing such as binarization processing with reference to the pixels in the
Will do. Then, the shading information from the image sensor
Report output is split into two shift registers
Quality degradation of image processing results due to
It works. This means that the clock is independent in each system
Therefore, there is a shift in the density change between the even pixel row and the odd pixel row.
However, this clock is used between even pixel columns or between odd pixel columns.
This is because there is no fluctuation in density change due to independence.

[0033] The processing means 2 performs affine transformation, Fourier transformation,
Further, predetermined image processing such as binarization processing, statistical processing and recognition processing is performed.

[0034]

Since the present invention is constructed and functions as described above, according to this, it is possible to perform image processing only with the output from the same shift register, and therefore, it is possible to output the grayscale information from the image sensor. There it is possible to prevent the image processing results of the quality degradation due to being divided into a shift register of the two systems, further secondary effects
By performing such image processing,
And adjustment of the density information from the CCD by the second correction means.
There is no need to be strict, and therefore the system
This makes it possible to provide an excellent image processing apparatus which has not been achieved conventionally.

[Brief description of the drawings]

FIG. 1 is a configuration diagram showing one embodiment of the present invention.

FIG. 2 is a detailed configuration diagram showing details of an image sensor in FIG. 1;

FIG. 3 is a detailed configuration diagram illustrating details of an image processing mask in FIG. 1;

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 output control means 2 processing means 3 image processing mask 4 density information control means 5 image sensor 6 even pixel row density information output means 7 odd pixel row density information output means 8 first correction means 9 second correction means 10 first A / D converter 11 Second A / D converter

Claims (1)

(57) [Claims] An image sensor for photoelectrically converting a grayscale image such as a character, a figure, or a photograph, and an output from the image sensor.
A / D converter that converts the obtained grayscale information into a digital signal, and a digital signal output from the A / D converter.
And an image processing device that performs image processing based on the
In this case, the image data is input to the processing unit from an even pixel row of the image sensor.
The digital signal of the input grayscale information is used as the even pixel row information.
And from the odd pixel row of the image sensor.
The input grayscale information digital signal is
Image processing mask to store as image and image by the processing unit
When the pixel to be processed is the even pixel column information,
Processing the digital signal of the even pixel column of the image processing mask
Output to the processing unit and the image processing target pixel by the processing unit
Is odd pixel column information,
An output for outputting a digital signal of an odd pixel column to the processing unit
An image processing apparatus comprising a control unit and a control unit .