JPH01106184A - Image signal correcting device - Google Patents

Abstract

PURPOSE:To execute highly accurate shading correction for gamma correction or the like by varying reading density in order to read out gamma characteristics in each bit of an image sensor and storing reading data in each bit and in each reading density to form a correcting look-up table. CONSTITUTION:The image of an original (a) and reference density generated by a density generating means (b) are read out by an image sensor (c). A 1st address generating means (d) generates a gradation address corresponding to the reference density. A 2nd address generating means (e) generates a picture element address corresponding to each picture element in the image sensor (c). A storage means (g) successively stores density data in the image sensor reading out the reference density by using the gradation address and the picture element address as writing addresses. An arithmetic means (h) converts the stored density data density-corrected data for continuous gradation in each picture element of the image sensor (c).

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image signal correction device for an image reading device using an image sensor.

[Conventional technology]

Conventionally, gradation correction (for example, shading correction, etc.) of image signals obtained from an image sensor in an image reading device includes two methods: one for correcting bit variations in darkness and the other for correcting bit variations in light and dark. . This correction is generally expressed as the following equation (1).

Here, 086w is corrected data, and Din is corrected data. α is the amount of variation in darkness (correction amount)
where D□8 is the maximum output value of the image data of the corresponding part to be corrected, and Max is the maximum output value of this reading system.6 FIG. 8 shows the relationship of the above equation (1).

Furthermore, since the above-mentioned correction is based on the assumption that the gradation characteristic of the image signal of the image sensor is linear, gamma (γ) correction is conventionally performed before these corrections.

(Problem to be Solved by the Invention) However, in the above-mentioned conventional example, the value of γ in the γ correction is constant; Many bits (
In the case of an image sensor composed of pixels of bits, the value of γ differs for each bit depending on the image sensor manufacturing process, transfer efficiency, etc. For this reason, there is a drawback that the gamma correction is not completed completely, resulting in unevenness in the image signal in the image sensor reading width direction.

Particularly, as shown in FIG. 9, when reading a document 2 by relatively moving the image sensor 1 in the X direction and the Y direction with a predetermined reading width of the image sensor 1,
As shown in Figure 10, there is a difference in the image signal at the joint 3 of the pixels of the image sensor I, which has the disadvantage that even though a continuous image is read, the image signal becomes discontinuous. .

Furthermore, when performing the above-mentioned correction, when performing correction calculations digitally, the image data to be handled is, for example, 8 bits.
In the case of 8-bit calculation at t, the number of output bits is often 8 bits. In this case, there is a drawback that the error becomes large, especially when the image signal is small.

Furthermore, after performing correction calculation processing for each bit, a lookup table etc. of correction data is created, so it takes a considerable amount of time to create the correction data. However, there was a drawback in that it took a considerable amount of time to read the second image.

The present invention eliminates the above-mentioned drawbacks, makes it possible to perform detailed and highly accurate correction for each pixel of an image sensor and each density, and thereby eliminates image unevenness caused by image reading. An object of the present invention is to provide an image signal correction device.

Means for Solving Problem C] In order to achieve the above object, the present invention provides an image signal correction device for an image reading device that uses an image sensor to read a document image. density generating means for generating a variable reference density; first address generating means for generating a gradation address corresponding to the reference density generated by the density generating means; a second address generation means for generating a pixel address; a storage means for sequentially storing the density data of the image sensor that has read the reference density as a writing address; A calculation means converts the data into continuous tone density-corrected data for each pixel of the image sensor, and when reading the original image, reads out the density output of the image sensor and the pixel address of the second address generation means. [Function] The present invention is characterized by comprising an address control means for reading out density-corrected data from the storage means as an address. [Operation] In the present invention, means for changing the reading density for reading the γ characteristic of each bit of the image sensor; It also has a memory that stores the data read by the image sensor in bit units and read density units, and uses the read data to create a look-up table for correction using arithmetic means. It is possible to perform γ correction of signals and shape ink correction for variations in dark conditions.

Furthermore, by ensuring that the number of data bits after correction is equal to or greater than the number of bits of input data, correction errors associated with correction can be minimized.

Furthermore, instead of creating the look-up table for the above-mentioned correction for each copy, it can be done only when the power is turned on or when the operator wants to perform the correction operation using the operation unit. By doing so, processing time can be shortened.

〔Example〕

Embodiments of the present invention will be described in detail below with reference to the drawings.

FIG. 1 shows the basic configuration of an embodiment of the present invention. In this figure, a is a document, b is a density generating means, and an image sensor C reads the image of the document a and the reference density generated by the density generating means. The density generating means generates a continuously variable reference density via a reference density plate. d is a first address generating means that generates a gradation address corresponding to the reference density already generated, and e is a first address generating means that generates a pixel address corresponding to each pixel of the image sensor C during reading scanning. This is the second address generation means.

g is a storage means that sequentially stores the density data of the image sensor that has read the reference density as a gradation address and a pixel address as a write address, and h is a storage means that stores the density data stored in the storage means g for each pixel of the image sensor C. This is a calculation means that converts each image into continuous tone density-corrected data. f is an address control means, which uses the density output of the image sensor C and the pixel address of the second address generation means e as read addresses when reading the original image;
Address control is performed to read the density-corrected data from the storage means g.

FIG. 2 shows a schematic configuration of an embodiment of the present invention.

In this figure, 100 is an original to be read or a reference correction plate that can continuously generate a reference density for performing the correction of the present invention, and 101 is an original or a reference correction plate 100.
It is a lamp that illuminates the Reference numeral 102 denotes an image sensor such as a CCD that receives reflected light from the original or the reference correction plate 100 and reads the image. 103 is an amplifier for amplifying the image signal photoelectrically converted by the image sensor 102, and 104 is for converting the analog image signal amplified to a predetermined value by the amplifier 103 into a digital image signal.
(analog to digital) converter.

105 and 106 are memory 10 that stores correction data.
The former multiplexer 105 receives the digital image signal or gradation address A1 from the A/D converter 104, and the latter multiplexer 106 receives the output of the multiplexer 105 and the output from the main counter. Count value A2 and CPU address ^
3 is selected and input. 111 performs arithmetic processing to convert the image data stored in the memory 107 into correction data via the bus transceiver 110;
This is a main control unit that corrects the amount of light of the controller 1 and selects a control mode by exchanging signals with the operation unit 112.

108 is a D-type flip-flop circuit (F/F) connected to the output terminal of the A/D converter 104;
The output of type F/F 108 is sent to memory 107. 1
09 is a D-type F/F connected to the output side of the memory 107.

In the above configuration, first, in order to create correction data, a certain density level must be actually read by the image sensor 102. For convenience of explanation, if the A/D converter 104 has an 8-bit resolution, the resolution of the image will automatically become 256 gradations, and the density read by the image sensor 102 will be 256 levels. Further, if a 256-pixel, three-color image sensor is used as the image sensor 102, the data format of the memory 107 will be as shown in FIG. 3, for example.

In other words, since the number of pixels stored in the memory 107 is 768, if density data of 256 gradations is to be stored in the memory 107 for each pixel, it is sufficient that the memory 107 has a storage capacity of 256 bytes per pixel. However, in order to create correction data in the main control unit 111 based on this stored data and to reduce calculation errors, more than 8 bits is required, so 2 bits are added to each gradation. Let us allocate a memory space of bytes.

Therefore, first, assuming that the O level of 256 density levels is the darkest image level, first switch to the correction data input mode and correct (adjust) the light amount of the lamp 101 using the main control unit ill so that it becomes the 0 level. ), the gradation address AI is made to correspond to the density level of the reference density, and the value A2 of the main counter is made to correspond to the pixel of the image sensor 102. As a result, the gradation address Al and main counter A2 are input as the write address of the memory 107,
The image data from the image sensor 102 at that time is A/
From the D converter 104 to the D type F/Fl (through 18 to the memory 1
87. At this time, the image data is used as a gradation data sample for the gradation address ^l,
Every other item is written.

Based on the image data for a constant density written in the memory 107 in this way, the main control unit 111 performs arithmetic processing so that the density curve becomes a straight line for each pixel. The correction data obtained through the arithmetic processing is sequentially loaded into the table to complete the creation of the correction data lookup table.

Next, when correcting the actual image data based on this correction data, the image data from the A/D converter 104 is sent to the multiplexer 105 instead of the gradation address A1 when reading the correction data. By entering the image sensor 102 as the read address of the memory 107,
Input the value A2 of the main counter corresponding to the pixel and its image data, read out 2 bytes of image correction data from the memory 107, and transfer the read image correction data to the D type F/
It is received at F109 and the image data is output to the outside.

FIG. 4 shows the above-mentioned operating procedure.

The flowchart in FIG. 4 is divided into five steps, SPI to SP5.

First, in SPl, by switching the correction data input mode, the multiplexer 105 selects the gradation address A1 in order to sample gradation data, and the subsequent multiplexer 106 selects the gradation address A1 of the previous multiplexer 105 and the main counter A2. and select memory 10
7 is given a write address corresponding to the pixel and density of the image sensor 102, and set so that image data can be taken into the memory 107 from the image sensor 102 through the D-type F/F 108. In the next SF3, the main counter A
2 and the gradation address ^1, the image data is actually taken into the memory 107 as a gradation data sample. This S
Through the two steps PI and SF3, image data for a constant density of the actual image sensor 102 can be obtained from the memory 107.

Next, in SF3, the multiplexer 106 selects CPU address A3 by switching the CPU access mode so that the memory 107 can be accessed by the main control unit 111, and then, in SF3, the input image stored in the memory 107 is The main control unit 111 calculates so that the output image data is linear in density with respect to the data,
The calculation result is set in a predetermined area of the memory 107 as a lookup table.

Further, in SF5, the multiplexer 105 selects the image data input from the A/D converter 104,
Furthermore, by selecting the image data selected by the multiplexer 105 and the main counter A2 at the multiplexer 106 as the read address of the lookup table in the memory 107, it becomes possible to perform gradation correction on the image data input from the D-type F/F 108. .

The correction in this embodiment is performed for each pixel for 256 gradation levels, so the time required for this correction is long. It takes too much time. Therefore, a control signal (command) is sent from the operation section 112 to the main control section 111 so that the correction operation shown in FIG. This makes it possible to operate it.

FIG. 5 shows an example of the operating procedure using the operating section 112, and FIG. 6 shows an example of the display screen of the operating section 112 at that time.

First, when you turn on the power, the correction data is still stored in memory 10.
7, so please make sure to make the 6th one to perform the correction operation.
By displaying a message such as 61 in the figure on the display section of the operation section 112, instructions are given to the operator (SPI
I).

However, if the operator presses the copy switch (copy start switch) (not shown) before inputting the correction key for automatic image correction (SPI2.5P13), the operation will be canceled as shown at 62 in FIG. After displaying on the display section 112 that the image is being automatically corrected (SPI4) and executing the correction operation shown in FIG. 4 in automatic fishing (SPI5),
A copying operation is performed (5P16).

As a result, when the power is turned on, the correction operation shown in FIG. 4 is always performed regardless of whether or not the correction key is input, thereby preventing defective copies.
It is possible to display on the display unit that the correction is in progress to notify the operator of the status of the image reading device.

Moreover, even after that, correction can be performed by pressing the correction key (SP17 to 5P21), and correction can be performed according to the current temperature and power supply voltage of the device, making it possible to obtain even better images.

Note that the image reading device of the embodiment shown in FIG. 2 can also be used as a pattern generator. In this case, first, an image pattern to be output is written in memory 107 in advance. The address at the time of outputting this image pattern is given by the gradation address AI and the main counter ^2 by selecting the gradation address AI with the multiplexer 105 and selecting the main counter ^2 with the multiplexer 106. Here, the difference from when correcting the image data described above is that the gradation address AI is input as the readout address instead of the image data, and by changing this gradation address ^l, various patterns can be created. It can be generated sequentially.

In this way, by using the image reading device of this example as a pattern generator, it is possible to check the subsequent image processing circuit and the processing coefficients in the image processing circuit, and also to check the output of the printer, etc. Equipment can also be inspected.

FIG. 7 shows the main structure of another embodiment of the present invention. In this figure, reference numeral 71 denotes a correction light source whose light intensity can be varied, and light from the light source 71 passes through a condensing lens 72 and forms an image on an area sensor 73. The image signal obtained from the area sensor 73 is amplified by an amplifier 74, then converted to a digital image signal by an A/D converter 75, and inputted to a correction section 76 similar to that shown in FIG. After being corrected, it is output as a video output. The correction light source 71 is adjusted by a correction section 76.

In this way, the present invention can perform the same correction as the embodiment shown in FIG. 2 even in a video camera using the area sensor 73. That is, by attaching a surface light source (correction light source) 71 with variable brightness in front of the lens 72 in order to apply a uniform amount of light to the area sensor 73 of the video camera, the same correction as in the embodiment shown in FIG. 2 can be completely performed. It becomes possible to do so.

Furthermore, sampling image data for all density levels as described above takes time, so data is sampled at certain intervals of density, and the density that has not been sampled is approximated by data interpolation. By doing so, it becomes possible to shorten the processing time. For example, in the example shown in Fig. 2, data sampling for density correction was performed for all gradations, but since the actual linearity of the image sensor does not change drastically at all gradations, data sampling for density correction is performed for all gradations. For example, by approximating the gradation data for which no sampled data has been obtained by dropping the sampling data, for example, by approximating it with a straight line, the same effect can be obtained, and the processing time for correction can be shortened. Depending on the number of resulting gradation levels, it is also possible to execute correction data creation processing for each copy.

(Effects of the Invention) As explained above, according to the present invention, since correction data can be obtained for each pixel of the image sensor and each density, it is possible to perform detailed and highly accurate correction, and The effect of almost completely removing unevenness can be obtained.

Furthermore, processing time can be shortened by employing an approximation method based on data interpolation for creating correction data according to the present invention, and by executing correction processing only when necessary.

Furthermore, by using the device of the present invention as a pattern generator, it is possible to easily adjust printers and video output devices.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the basic configuration of an embodiment of the present invention, FIG. 2 is a block diagram showing the circuit configuration of an embodiment of the present invention, and FIG. 3 is an address map showing the memory configuration of FIG. 2. FIG. 4 is a flowchart showing the procedure of the correction operation of the embodiment of the present invention shown in FIG. FIG. 5 is a flowchart showing an example of an operation procedure regarding the operating section of FIG. 2, FIG. 6 is an explanatory diagram showing an example of a display displayed on the operating section of FIG. 2, and FIG. A block diagram showing the circuit configuration, FIG. 8 is a characteristic diagram showing the relationship between the image sensor read image density and the corrected output value, and FIGS. 9 and 10 are explanatory diagrams explaining the scanning state of the image sensor. be. 101--Reference correction plate, original, 101--Lamp, 102--Image sensor, 103--Amplifier, 104--A/D converter, 105.106--Multiplexer, 107--・Memory, 108.109-D type F/F (flip-flop),
11G... Bus transceiver, 111... Main control unit, 112... Operation unit. (7 dress map of memory of 4 frames Figure 3 Flowchart of actual eyebrow M column Nenan' positive movement I) Figure 4 Flowchart of A-direction I of award method arrangement Figure 5 Figure 8 Figure 1O figure

Claims (1)

[Scope of Claims] 1) a) An image signal correction device for an image reading device that reads a document image using an image sensor, including: b) a density generating means that continuously generates a variable reference density via a reference density plate. c) a first address generating means for generating a gradation address corresponding to the reference density generated by the density generating means; and d) generating a pixel address corresponding to each pixel of the image sensor during reading scanning. e) storage means for sequentially storing the density data of the image sensor that has read the reference density with the gradation address and the pixel address as write addresses; g) calculating means for converting the stored density data into continuous tone density-corrected data for each pixel of the image sensor; g) when reading an original image, the density output of the image sensor; 2. An image signal correction device comprising: address control means for reading out the density-corrected data from the storage means using the pixel address of the address generation means as a readout address. 2) The image signal correction apparatus according to claim 1, wherein the calculation means sets the number of bits of the density-corrected data to be larger than the number of bits of the density data. 3) The image signal correction device according to claim 1, wherein the calculation means creates the density-corrected data by approximate calculation using data interpolation. 4) The image signal correction device according to claim 1, wherein the storage means inputs a pattern image in synchronization with the image sensor instead of the reference density. 5) An image signal correction device according to claim 1, characterized in that it has display of a correction instruction message and correction instruction means, and is capable of performing a correction operation at an arbitrary timing apart from a reading operation.