JPH0223475A - Logarithm converting circuit - Google Patents

Abstract

PURPOSE:To make concentration conversion highly accurate by providing an A/D converting means to change a conversion gain with the level of an analog input signal and a logarithm converting table to switch reference data according to the switching of the conversion gain. CONSTITUTION:When an analog input inserted to an input terminal A of a comparator 106 is smaller than a direct current voltage divided by voltage- divided resistances 109 and 110 inserted to an input terminal B, the output of the comparator 106 becomes low, a selector 103 selects the input A, an A/D converter 104 A/D-converts the high gain, the output selects a high gain converting LUT by a logarithm converting LUT 105, and executes the logarithm conversion. On the other hand, when the input terminal A of the comparator 106 becomes higher than the input terminal B, the output becomes high, and the selector 103 selects the input B. Consequently, the A/D converter 104 A/D- converts the standard gain, and the output selects a LUT for the standard gain by the logarithm converting LUT 105, and executes the logarithm conversion.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to signal processing of a document reading device, and particularly to a logarithmic conversion circuit that performs brightness-density conversion, that is, logarithmic conversion.

Conventional technology Conventionally, in image reading devices, a document is irradiated with white light or monochromatic light, and the intensity of the reflected or transmitted light is detected by a photodiode or phototube to obtain a reading output of the document point. After that, logarithmic transformation is performed to obtain C, M,
A Y density signal is obtained. FIG. 7 shows the configuration of logarithmic transformation using the conventional method. In this example, first, an input analog signal is A/D converted by an A/D converter 701 based on a reference signal, and then input to a logarithmic conversion table (LUT) 702, and the output is converted to a logarithm for transmittance linear data. Concentration information is obtained. The characteristics of LUT 702 are shown in FIG. The LUT conversion characteristics in Figure 8 are normally as follows: If the input is Y1 and the output is D, then D = -1ogxo Y (However, Y is 0 to 1Ω
normalized value). In the case of such a conventional example, if the concentration resolution of the output signal (2) is set to 8 bits, for example,
Due to the non-linearity of the logarithmic curve, the input data Y requires a density resolution of 8 bits or more. For example, to ensure 6 bits (64 gradations) in the density data, the input reflectance data must be 1
It is known that 12-bit input reflectance data is required in the case of 0 via) strong and 8-focus density (as described in Journal of the Japan Printing Society, Vol. 25, No. 3, 1988, p. 175).

Therefore, the A/D converter 701 shown in FIG. 7 often has a high resolution of 10 bits or more and converts it into a desired concentration signal.

Problems to be Solved by the Invention However, with conventional methods, when attempting to achieve high-speed and high-precision image reading, a high-speed and high-precision A/D converter is required, which makes the device expensive or does not meet the purpose. Matching A/D
The problem was that a converter could not be realized.

In view of the need for a high-speed, high-precision A/D converter during concentration conversion, the present invention improves the conventional method. It is possible to perform highly accurate concentration conversion. A second purpose is to perform highly accurate concentration conversion using a plurality of low resolution A/D converters.

Means for Solving the Problems In order to achieve the above object, the technical solution of the present invention is as follows:
First, the conversion gain and concentration conversion table of the A/D converter are switched according to the reflectance or transmittance linear analog signal level. This is intended to increase the number of conversion bits of the A/D converter isometrically by increasing the conversion gain of the converter.

The second method is to provide a plurality of A/D converters with different conversion gains and density conversion tables corresponding to each one, and to switch according to the analog signal level to obtain the same effect as the first method. It is something.

Effects of the present invention Firstly, when the input level to the A/D converter is small, by increasing the conversion gain of the A/D converter, a high conversion gain can be obtained without saturation. This is apparently equivalent to conversion with an increased number of bits in the A/D converter.

When the input level becomes high and exceeds the level at which saturation of the A/D converter occurs, the conversion gain of the A/D converter is restored (
(lower), it is converted using the number of bits of a normal A/D converter. By passing the data converted with a high conversion gain through a high conversion gain table, the data is density-converted using a number of bits that is substantially greater than the number of bits of the A/D converter used. Logarithmic transformation requires finer data when the input value is small, and when the input data is large, the fineness of the input data may be coarser than the data after conversion. Therefore, sufficient effects can be obtained by simply increasing the number of bits of the A/D converter only when the input is low as described above.

Second, instead of switching the gain of a single A/D converter, a plurality of A/D converters are used and the gain of each A/D converter is set to a different value, and the input level is changed. By switching the A/D converter according to the change and switching the contents of the conversion gain table accordingly, the same effect as in the first example can be achieved.

EXAMPLE A first example of the present invention will be described below with reference to FIG.

A linear reflectance or linear transmittance signal from a document reading sensor (not shown) is input to an analog input terminal 101 . Direct analog input terminal 10 to selector 103
1 and analog inputs divided by dividing resistors 107 and 108 are respectively input. On the other hand, a DC voltage is supplied to the reference signal input terminal 102 to a Vref terminal that determines the A/D conversion gain to the A/D converter 104. Further, the DC voltage divided by the voltage dividing resistors 109 and 110 is input to the comparative human power B of the comparator 106, and the other comparative human power A is input directly from the analog input terminal 101. Here, the voltage dividers 107.108 and 109.111 do not need to have the same voltage division ratio. Here comparator 10
When the input A of 6 is smaller than B, the comparator output becomes low, the selector 103 selects the A input, the A/D converter 104 performs high gain A/D conversion, and the output is converted to high gain by the logarithmic conversion LUT 105. A conversion LUT is selected and logarithmic conversion is performed. On the other hand, when the analog input terminal voltage becomes higher and the comparator input terminal A becomes higher than B, its output becomes NO, and the selector 103 selects the B input. Therefore, the A/D converter 104 has a standard gain of A/D.
A D conversion is performed, and the output thereof is subjected to logarithmic conversion by selecting a logarithmic conversion LUT for standard gain. The LUT 105 is connected to the A/D converter 10 by the output of the comparator 106.
It operates by switching between two types of tables depending on the high gain or standard gain operation of 4.

FIG. 2 shows the analog input, A, in the embodiment of FIG.
FIG. 3 is a diagram showing the relationship between /D output and LUT output. 201 is an A/D conversion straight line during standard conversion, 202 is an A/D conversion straight line during high gain conversion, 203 is a logarithmic conversion curve during standard conversion,
204 shows a logarithmic conversion curve during high gain conversion (in the figure, the broken line portion does not occur due to switching of the A/D conversion gain and LUT depending on the input signal level). For example, when the input value is a, high gain conversion is performed as indicated by the arrow in the figure to obtain a converted output of a', and when the input value is b, standard gain conversion is performed to obtain a converted output of b'.

Next, a second embodiment of the present invention will be described.

In Fig. 3, unlike Fig. 1 in which the input level was changed to change the gain of the A/D converter, the conversion reference voltage Vre
f is switched depending on the input level. In this figure, 301 to 308 correspond to 101 to 108 in FIG. The DC voltage at the reference signal input terminal 301 is divided by voltage dividers 307 and 308, and after being selected by the selector 303,
It is used as an A/D conversion reference voltage of the A/D converter 304 to switch the conversion gain. The switching signal is obtained by comparing the signal at the analog input terminal 301 and the above-mentioned divided DC voltage using a comparator 306, and then switching the gain of the A/D converter 304 and switching the logarithmic conversion LUT 305. There is.

Next, a third embodiment of the present invention will be described using FIG. 4.

FIG. 4 is an improved version of FIG. 2 in which the gain of the A/D converter is switched in two stages, and the gain of the A/D converter is switched in multiple stages.

The input analog signal is the reference input signal and the second A/D
The converter 406 performs A/D conversion, and the output is converted to the first A/D.
The reference voltage V r e f of the D converter 404 and the density conversion LUT 405 are switched in multiple stages. The D/A converter 404 uses a second input reference voltage 402 as a reference.
A reference voltage is generated in multiple stages based on the output of the A/D converter 406. FIG. 5 shows how density conversion is performed in multiple stages in the embodiment of FIG. 4. It shows how the levels of input analog signals at points a, b, and c are converted to points a', b', and 07, respectively, using a multi-step conversion curve.

A fourth embodiment of the present invention will be described below with reference to FIG. As shown in FIG. 6, a linear reflectance or linear transmittance original reading output signal from an analog input terminal 601 is input to two A/D converters 603 and 604. The reference voltage of the A/D converter 603 is input from the reference signal input terminal 602, and the reference input of the A/D converter 604 is inputted from the voltage of the reference signal input terminal 602 through voltage dividers 608 and 60.
The voltage divided by 9 is input, and the A/D converter 604
A/D conversion is performed with a conversion gain higher than that of the /D converter 603. When the A/D converter 604 becomes saturated and near the bell, the switching signal generator 610 switches the A/D converter 603
．． Buffers 605 and 606 connected to the output of A/D converter 604 are switched, and the output of A/D converter 603 becomes L for logarithmic conversion.
Input to UT607. The LUT 607 outputs two A/D converters 603 by the output of the switching signal generator 610.
Since we select the logarithmic conversion table corresponding to and 604,
The same effect as the first embodiment shown in FIGS. 1 to 5 can be obtained.

As described in detail, the effects of the present invention include the ability to perform luminance-density, ie, logarithmic conversion, in a document reader, etc., using an A/D converter and L.
When configured with a UT, highly accurate brightness-to-density conversion can be performed using an A/D converter with a bit length coarser than the required bit length of the A/D converter, and an inexpensive logarithmic converter can be configured.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a logarithmic conversion circuit according to a first embodiment of the present invention, FIG. 2 is a characteristic diagram of the embodiment of FIG. 1, and FIG. 3 is a logarithmic conversion circuit according to a second embodiment of the present invention. 4 is a block diagram of a logarithmic conversion circuit according to the third embodiment of the present invention; FIG. 5 is a characteristic diagram of the embodiment of FIG. 4; and FIG. 6 is a diagram of the fourth embodiment of the present invention. FIG. 7 is a block diagram of a logarithmic conversion circuit in the embodiment, FIG. 7 is a block diagram of a conventional logarithmic conversion circuit, and FIG. 8 is a characteristic diagram of the same. 101.301.401.601... Analog input terminal, 102.302.402.602... Reference signal input terminal, 103.303.403... Selector, 10
4.304.404.406.603.604...A
/D converter, 105.305.405.607...L
UT, 107.108.109.110.307.3
08.608.609... Voltage divider, 605.606.
- Buffer, 610... switching signal generator. Name of agent: Patent attorney Shigetaka Awano and 1 other person Book 101 Acho Log 8man 5 / 306 Ko:, Nog L-pno / 604% Kokyo Drink\ Za06nohe Sofa

Claims (2)

[Claims] (1) An A/D converter that receives an analog signal and changes the conversion gain depending on the level of the analog input signal, and switches the data to be referenced in accordance with the switching of the conversion gain of the A/D converter. A logarithmic conversion circuit comprising a logarithmic conversion table for performing the logarithmic conversion. (2) A plurality of A/D conversion means that input an analog signal and have different conversion gains, and a reference signal corresponding to the conversion gain of each A/D converter according to the level of the analog input signal. A logarithmic conversion circuit comprising a logarithmic conversion table for switching data.