JPH0423467B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image reading device that quantizes an image signal read by scanning an original image.

Conventionally, a solid-state scanning device has a large number of solid-state scanning elements such as CCDs arranged in an array, and the reading width of each line in the main scanning direction is divided to read the document image in parallel by multiple CCD arrays. In the scanning device, the peak level of the image signal is detected for each CCD array, and the white level reference is detected for each CCD array.

However, if the white level reference detection of the image signal read in each CCD array is performed, for example, as shown in Fig. The reading range of the CCD array is in that part B.
If all of the above is input, the detected white level reference (peak level detection value) in that part B will be different from that in other parts, and therefore the image signals read in parallel by each CCD array will be different. If normalization or quantization is performed at the peak level, portion B and other portions of document A will be processed using different white level standards, resulting in uneven density and poor quality in the reproduced image. It has the disadvantage of being In addition, Figure 2 shows the main scanning line L of a document A having a portion B with high background density in the state shown in Figure 1, and two
This shows each image signal when read in parallel by the CCD array CCD1 and CCD2, and the peak level of the image signal read by CCD1 is the CCD
The peak level of the image signal read by No. 2 is lower than the peak level of the image signal read by No. 2. At that time, for example, CCD
1. A/D of image signals read by CCD2
If the conversion is performed independently using 4 bits (16 values), if each white level standard has a density of 0.05 and 0.2,
04 Concentration information is output from A/D conversion on CCD1 side.
10, and the A/D conversion output on the CCD2 side will show 7. Even if the image information has the same density, the output after each A/D conversion will be different data, causing density unevenness in the reproduced image. This will occur.

The present invention has been made in consideration of the above points, and
In an image reading device that uses multiple CCD arrays to read image information in parallel by dividing the scanning width of each line in the main scanning direction of a document image, the CCD arrays are Even if the entire reading range contains an image part with a background density different from other parts, the images are read by each CCD array so that density unevenness does not occur in the reproduced image. An object of the present invention is to provide an image reading device capable of quantizing an image signal.

Hereinafter, one embodiment of the present invention will be described in detail with reference to the accompanying drawings.

Figure 3 shows two CCD arrays CCD1 and CCD2.
was used to divide the reading width of each line in the main scanning direction of the original image and read the image information in parallel, and each of the read image signals was amplified by amplifiers AMP1 and AMP2. Later, when the A/D converters A/D1 and A/D2 quantize the amplified image signals BS1 and BS2, the respective A/D converters A/D
1 and A/D 2 are provided with a white level reference detection circuit 1 according to the present invention which provides a common normalized signal, that is, a white level reference signal _Vref .

The white level reference detection circuit 1 includes a peak detector 2 that detects the peak level of the image signal BS1 read by the CCD 1, and a peak detector 2 that detects the peak level of the image signal BS2 read by the CCD 2. 3, and a maximum value selection circuit 4 which selects the larger signal among the peak level signals detected by each of the peak detectors 2 and 3 and outputs it as the white level reference signal _Vref . It is structured as follows.

Note that FIG. 4 shows an example of the configuration of the peak detectors 2 and 3, including an operational amplifier OPA and a diode D.
and a capacitor C, and is configured to detect an increase in the input signal by an operational amplifier OPA, and deposit it cumulatively in the capacitor C to perform peak detection of the input signal. Also,
FIGS. 5a and 5b show examples of the configuration of the maximum value selection circuit 4, with FIG. 5a showing one using a diode OR gate system, and FIG. 5B showing one using a FETOR gate system. In the case of the diode OR gate method, each diode D1,
The output voltage decreases due to the forward voltage drop of D2, but in the case of the FETOR gate method, by appropriately selecting the value of resistor R1,
It is possible to make the input and output voltages the same magnitude.

In this configuration, first, the image signals photoelectrically converted by each CCD1 and CCD2 are sent to the respective amplifiers AMP1 and AMP2.
and each amplified image signal BS1, BS
2 peak levels are detected by peak detectors 2 and 3, respectively. Since the outputs of CCD1 and CCD2 are proportional to the amount of incident light on which the image information of the original is written, the peak value becomes the signal of the whitest part of the line read by CCD1 and CCD2. Next, the maximum value selection circuit 4 selects the larger signal among the peak level signals respectively detected by the peak detectors 2 and 3, and selects it as the white level reference signal V _ref to the A/D converter A/D converter A/D. D1 and A/D2 respectively, but at that time each A/D2 is given as shown in Figure 6.
The D converters A/D1 and A/D2 divide the value of the white level reference signal V _ref equally according to a predetermined number of bits (for example, in the case of 4 bits, it is divided into 16 stages) and divide it into each input. In order to quantize the signals BS1 and BS2, respectively, the image information of the same density level read by the CCD1 and CCD2, respectively, is converted into digital values having the same value. Therefore, when an image is recorded based on each quantized digital image signal, a reproduced image of good quality can be obtained without causing a difference in density between the right and left half image parts. You will be able to do this.

FIG. 7 shows another embodiment of the white level reference detection circuit according to the present invention. In this case, the maximum value selection circuit 5 uses the image signals BS1 and CCD2 read by the CCD1. The larger one (maximum value) of the read image signals BS2 is selected, and its peak level (maximum value) is held and detected by the peak detector 6 to obtain the white level reference signal _Vref . This has the advantage of simplifying the circuit configuration.

Note that the white level reference detection circuit according to the present invention is as follows:
This is not limited to the case where one line in the main scanning direction is divided by multiple CCD arrays and the image signals read in parallel are quantized.
The gain of the voltage-controlled amplifier that receives each image signal as input normalizes the image signal read by the CCD array according to the deviation signal between the sensor output that detects the amount of incident light on each CCD array and the white level reference signal. In some cases, the peak level value is subtracted from each image signal by using a subtracter after detecting the peak level of each image signal read by multiple CCD arrays. It goes without saying that this method can be similarly applied to cases where the background density level of a document is removed by subtraction.

As described above, in the image reading device according to the present invention,
One line of image information in the main scanning direction is divided and read in parallel by a plurality of scanning element arrays, and each read image signal is sent to an A/D converter provided for each scanning element array. It has the excellent advantage of being able to set a common normalization signal to each A/D converter when quantizing them, making it easy to perform optimal processing without causing density level errors. ing.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an example of an original image, FIG. 2 is a diagram showing each image signal when the original image is divided and read by two CCD arrays, and FIG. 3 is an example of an embodiment of the present invention. A block configuration diagram when the white level reference detection circuit according to the example is applied to an image signal quantization device, FIG. 4 is an electric circuit diagram showing an example of the configuration of the peak detector in the same example, and FIGS. 5 a and b 6 is an electric circuit diagram showing an example of the configuration of the signal selection circuit in the same embodiment, FIG. 6 is a diagram showing the characteristics of the A/D input range of each CCD array with respect to the original density in the quantization device of FIG. 3, and FIG. The figure is a block diagram showing another embodiment of the present invention. 1... White level reference detection circuit, 2, 3, 6...
Peak detector, 4, 5... Maximum value selection circuit.

Claims (1)

[Claims] 1. Image information for each line in the main scanning direction is divided and read by a plurality of scanning element arrays, and each read image signal is provided corresponding to each scanning element array. A peak detector that detects the peak level of each image signal divided and read by each scanning element array, and each detected peak level. and a maximum value selection circuit for selecting the maximum value of the signals, and the output of the maximum value selection circuit is given to each of the A/D converters as a common normalized signal and read by each scanning element array. An image reading device characterized in that the image reading device is configured to perform quantization of each image signal. 2 Each line of image information in the main scanning direction is divided and read by a plurality of scanning element arrays, and each read image signal is sent to an A/D converter provided corresponding to each scanning element array. A maximum value selection circuit that selects the maximum value of the image signal divided and read by each scanning element array, and a maximum value holding circuit that holds the selected output of the maximum value selection circuit. circuit, and supplies the holding output of the maximum value holding circuit to each of the A/D converters as a common normalization signal to quantize each image signal read by each scanning element array. An image reading device characterized in that: