JPH0430670A - Picture reader - Google Patents

Abstract

PURPOSE:To realize the correction of a black level with high accuracy and to obtain a uniform picture data by using a subtractor and a gamma correction circuit in common to independently correct a dispersion in a black level respectively to each image sensor. CONSTITUTION:A correction data calculation device 4 calculates a black level correction data for each image sensor based on a digitized black level by an A/D converter 3. The integral part of the calculated correction data is set to a subtractor 5. Moreover, plural gamma correction tables subject to black level correction are prepared to a table selector 7 for each image sensor section. Then the fraction part of the above correction data is used and the table selector 7 selects an optimum gamma correction table for each image sensor. Thus, an uniform data is obtained. Then picture processing such as magnification or reduction is applied in a picture processing section 9 and the resulting picture is outputted.

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to an image reading device for converting an image into an electronic signal in a facsimile machine, a copying machine, etc. The present invention relates to an image reading device that can be easily implemented.

[Prior art] Figure 7 is from NEC Technical Report Vol. 1.41 No. 3/198
8 is a circuit block diagram of the conventional image reading device shown in FIG. In FIG. 7, 26 is a CCD sensor, 27 is an amplifier for amplifying the output of the CCD sensor, 28 is an A/D converter for A/D (analog/digital) conversion of the signal, and 29 is a shading correction, noise Signal processing unit such as reduction, 30, image reduction unit, 31
32 is an edge enhancement section, 32 is a ring buffer, 33 is a color correction section, 34 is an error diffusion section, 35 is a control section for controlling the entire circuit, 36a to 36e are line memories, and 37 is where color correction coefficients are stored. This is a certain memory section.

Next, the operation will be explained. The image signal obtained from the C0D 26 is amplified by the amplifier 27 and sent to the A/D converter 2.
8 into a digital signal.

The image signal digitized by the A/D converter 2 is converted into a noise-free, uniform signal by the signal processing section 29.

Next, the image reduction section 309 and the contour enhancement section 31. The image signal that enters the color correction section 33 is subjected to conversion processing to improve color reproducibility, and if further conversion to 24tL is required, it is binarized and outputted at the error diffusion section 34. Ru.

Next, signal processing will be explained. - When fishing on a boat, the following image distortion occurs due to the optical characteristics of the image reading device.

(1) Image distortion due to sensitivity differences between individual pixels due to variations in the photoelectric conversion characteristics of the image sensor's constituent elements.

(2) Distortion between image sensors due to differences in spectral intensity characteristics of individual image sensors when multiple image sensors are used.

(3) Shading distortion due to non-uniformity of light source illuminance between the center and edges, and insufficient light intensity at the edges due to lens characteristics.

In order to correct these distortions and variations, data at the white level is read in advance, and correction is performed in the signal processing section 29 using the white level data as a reference voltage.

As a method for correcting the variation in the black level, a method similar to the above-mentioned method for correcting the white error can be used. That is, a common method is to perform correction using a subtracter using black level data read in advance.

Therefore, in the conventional method, the black level can only be corrected within the range of integer values in digital values.

However, the output value of the black level value is quite low, so correction of only integer values cannot be said to be accurate, and even if correction is performed to the decimal part, a certain amount of hardware is required. do.

[Plot B to be solved by the invention] Since the conventional image reading device is configured as described above, the correction for the black level is only within the range of integers, and the correction for the black level with a low output value is There were problems in that the accuracy was insufficient for correction, and it was difficult to correct the decimal part.

This invention was made to solve the above-mentioned problems, and it is possible to accurately correct the black level with a low output value.
Moreover, the object is to obtain an image reading device that can be easily realized.

[Means to solve the problem]

The image reading device according to the present invention includes a light receiving unit that includes a plurality of image sensor units that convert reflected light from a document into an electrical signal, and a light receiving unit that reads several lines of black level data obtained by the light receiving unit in advance. The first means for creating correction data for each image sensor section is a subtracter that corrects the integer part of the black level using the integer part of the correction data for each image sensor section when reading an actual image, and a second means having a plurality of T correction tables that perform black level correction for each sensor section and selecting the T correction table for each image sensor section using a decimal part of the correction data; A T correction unit is provided for correcting the T characteristic of the output of the subtracter and for correcting the fractional part of the black level based on the selected T correction table.

(Function) The black level data obtained from the plurality of image sensor units constituting the light receiving means in this invention is
The correction data is calculated for each image sensor section, and when the actual image is read, the integer part of the black level is corrected using the integer part of the correction data using a subtracter. A T correction table is selected for each image sensor unit using the decimal part of the data, and the γ correction unit corrects the decimal part of the black level using the selected γ correction table.

〔Example〕

Embodiments of the image reading device of the present invention will be described below with reference to the drawings. The first part is a block diagram showing the configuration of one embodiment.

In FIG. 1, reference numeral 1 denotes a light receiving means composed of a plurality of image sensor sections that convert reflected light from the surface of a document into electrical signals; 2 an amplifier for amplifying the output signal from the light receiving means 1; 3 is an A/D converter for digitizing the output signal of the amplifier 2; 4 is an A/D converter that reads the black level signal of each image sensor section in advance and provides black level correction data for each image sensor section; This is a correction data calculator for calculation.

Further, during actual image reading, the subtracter 5 subtracts the read data for each image sensor section and the integer part of the correction data in order to correct the black level for each image sensor section. The integer part of the black level is now corrected.

For this subtracter 5, for example, JP-A-62-22006
A device using the concentration reference plate shown in FIGS. 3 and 4 of Publication No. 0 is shown.

That is, in this publication, a density reference plate with controlled white density and unevenness is provided on the edge of the contact glass on the back side of the original scale, and the density reference plate is optically scanned prior to scanning the original to obtain an image. The data obtained by the sensor is used as a density reference value when gamma-correcting a multi-value quantized digital screen of a document to be read later.

This density reference plate is provided with three reference surfaces having different reflection densities, and each of these reference surfaces is selectively sampled.

Here, the explanation returns to FIG. 1. The output of the subtracter 5 described above is output to a signal processing section 6. This signal processing section 6 is adapted to perform corrections on the white level obtained by the light receiving means 1, color corrections, and the like.

Furthermore, the correction table selector 7 selects γ correction tables that are the same in number as the number of image sensor units and that have undergone black level correction.
A table selector 7 selects this γ correction table.

This table selector 7 uses the fractional part of the correction data calculated by the correction data calculator 4 to select the optimum γ correction table for each image sensor section.

Further, after the white level correction and color correction are performed in the signal processing section 6, the T characteristic is corrected for each image sensor section, and at the same time, the black level of the decimal part is corrected, and the image processing section 6 It is designed to output to .

This image processing section 9 is for scaling and adjusting image quality.

Next, the operation will be explained. The black level data obtained from the light receiving means 1 having a plurality of image sensor sections is
The signal passes through an amplifier 2 and is digitized by an A/D converter 3.

Based on the black level data digitized by the second A/D converter 3, a correction data calculator 4 calculates black level correction data for each image sensor section.

At this time, accurate correction data is calculated by reading black level data for a plurality of lines and taking an average value for each image sensor section.

The integer part of the correction data calculated by the correction data calculator 4 is set in the subtracter 5.

Further, the table selector 7 is provided with a plurality of γ correction tables in which black level correction is performed for each image sensor section. Then, using the decimal part of the correction data, the table selector 7 selects the optimum γ correction table for each image sensor.

When actually reading image data, the subtracter 5 corrects the integer part for each image sensor section.
The signal processing section 6 performs white level correction and color correction. Thereafter, in the γ correction section 8, the T characteristic is corrected for each image sensor section, and at the same time, the black level of the decimal portion is further corrected to obtain uniform data. Thereafter, the image processing section 9 performs image processing such as enlargement/reduction, and outputs the resultant image.

Figure 2 shows the γ correction curve after black level correction.
Let the value of the decimal part of the black level correction data be ΔX,
By translating the T correction curve lO in the X-axis direction, a new γ correction curve 11 with black level correction is obtained.

A plurality of γ correction curves 11 are created for each image sensor section, and the table selector 7 shown in FIG. 1 selects the optimum γ correction curve correction table. Make corrections.

FIG. 3 shows a flowchart of the procedure for creating black level correction data and setting it in each circuit.

In step S1, perform initial settings for reading black level data, in step S2 turn on the light source, read the reflected light from the document with light receiving means 1 and convert it into an electrical signal, and wait 1 second in step S3. , the light source is turned off in step S4, the reading of the output of the light receiving means 1 is started in step S5, the output is amplified by the amplifier 2, digitized by the A/D converter 3, and the output is digitized by the correction data calculator 4 in step S6. , black data is collected, and after reading N947 minutes in step S7, accurate correction data is calculated by a method such as taking an average value for each image sensor section when calculating correction data in step S8.

The integer part of the correction data obtained by the correction data calculator 4 is set to the subtracter in step S9.

Furthermore, a γ table is selected for each image sensor unit in step SIO based on the decimal part of the correction data.

This series of operations from step 51 to step 510 can be performed before each image reading.

Furthermore, when reading black level data, the greater the number of lines N, the better the accuracy.

FIG. 6 shows a diagram of a multi-chip image sensor composed of four image sensor sections as an example of the light receiving means l having a plurality of image sensor sections. 38~ in the diagram
Reference numeral 41 indicates each of the first to fourth image sensor sections, and the above-mentioned γ correction table is prepared for each image sensor section, and black level correction is performed independently for each image sensor section. be exposed.

FIG. 4 is a block diagram showing the configuration of another embodiment of the invention. Portions indicated by 1 to 35 to 9 in FIG. 4 are the same as in FIG. 1.

In FIG. 4, the memory 19 and CPU 20 are the correction data calculator 4. of FIG. It is used in place of the table selector 7.

In FIG. 4, black level data obtained from the CCD 1 is stored in the memory 19 for several lines.

The CPU 20 reads the data in the memory 19 and calculates accurate correction data by, for example, taking an average value for each image sensor section. The integer part of the calculated correction data is set in the subtracter 5.

Furthermore, using the decimal part of the correction data, γ shown in FIG.
A correction curve is calculated and created for each image sensor section by the CPU 20, and is set in the .gamma. correction section 8 as a .gamma. correction table.

FIG. 5 is a flowchart showing the procedure of the series of processes described above.

In this FIG. 5, the processing from steps 31 to S7 is the same as the processing in the flowchart of FIG. In FIG. 5, the processing of 5ll-315 is different from that in FIG.

That is, the black level data is read in the memory 19 for N lives in step S7, and when the reading is completed in step S21, the data stored in the memory 19 is read out in step S22 and transferred to the CPU 20.
Based on this data, correction data is calculated in step S23, the integer part of the calculated correction data is set in the subtracter 5 in step S24, and T correction is performed based on the decimal part of the correction data in step S25. A γ correction table is created for each image sensor section.

As a result, the same effects as in the embodiment described above can be obtained. Moreover, there is no need to have a plurality of T correction tables in advance, and optimal correction can always be achieved.

〔Effect of the invention〕

As described above, according to the present invention, since the subtracter and the T correction are used together to correct variations in black level for each image sensor section independently, accurate black level can be achieved. This has the effect of easily realizing the correction, obtaining more uniform image data, and improving the reliability of the image quality.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an image reading device according to an embodiment of the present invention, FIG. 2 is a T correction curve diagram applied to the above embodiment, and FIG. 3 is a black level diagram in the above embodiment. FIG. 4 is a block diagram showing the configuration of another embodiment of the image reading device of the present invention, and FIG. 5 shows the flow of the operation of another embodiment of the same as above. 6 is a perspective view showing an example of a multi-chip image sensor having a plurality of image sensor sections, and FIG. 7 is a block diagram showing the configuration of a conventional image reading device. l... Light receiving means, 2... Amplifier, 3... A/D converter, 4... Correction data calculator, 5... Subtractor, 6
... Signal processing unit, 7... Table selector, 8...
T correction section, 9... Image processing section, 19... Memory, 2
0...CPU. Note that the same reference numerals in the figures indicate the same or equivalent parts. Agent Masuo Oiwa Figure 2 Figure 3 Figure 5

Claims (1)

[Claims] A light receiving means consists of a plurality of image sensor sections that convert light reflected from the document surface into electrical signals, and a black level signal is detected for each of the above image sensor sections using data obtained by reading the black level signal of the light receiving means in advance. means for calculating level correction data; a subtractor for correcting a black level signal using an integer part of the correction data for each of the image sensor sections when actually reading an image of the document; means for selecting the γ correction table for each of the image sensor sections using a decimal part of the correction data; and an output of the subtracter for each of the image sensor sections; means for correcting the γ characteristic of the image sensor, and selecting the γ correction table for each of the image sensor units based on the decimal part of the correction data, and the γ correction for each of the image sensor units selected by this means. An image reading device comprising: a γ correction unit that corrects the fractional part of the black level from a table.