JPS59231970A - System for changing magnification of picture for image input device - Google Patents

Abstract

PURPOSE:To obtain optional variable magnification and also to prevent deteri oration in picture quality by sampling a quantizing signal of an envelope in a fixed period, collecting the signal for the fixed period and applying sampling to the selected quantizing signal in the period in response to a desired magnification. CONSTITUTION:An output envelope signal of a comparator 5 is sampled by a sub-clock having a period of 1/2 and outputted from an AND circuit 6. A judging circuit 7 collects this sampling value series is collected at each period the same as the sampling period of an FF 8 so as to judge a correct sampling vlaue from the result of collection by means of majority decision. A comparator 12 outputs a value of logical ''1'' when the count value of a counter 11 is larger than a compared value 1.5 and outputs a value of logical ''0'' when smaller. Thus, the logical output signal is applied to the FF 8 and sampled by a main clock having a period of 1.5 time in order to obtain a display signal displaying the shrinked picture of 1/1.5 time and a desired picture display output signal is obtained.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention provides an image magnification conversion method for an image input device, particularly an image input device that prevents image quality from deteriorating during image reduction due to erroneous information caused by chipping or dirt on an object to be scanned. This relates to the image magnification conversion method of the device.

In an image input device using an image sensor such as a CCD or a BBD, it may be desired to arbitrarily enlarge or reduce the output image due to limitations of the original to be scanned or limitations of the output device. As a method for enlarging or reducing, we focus on the fact that the peak values of adjacent pulsed signals of the quantized output signal from the image sensor generally change slowly, and extract the envelope from the quantized output signal of the image sensor. And after quantizing this envelope.

The sampling period corresponds to the desired image magnification, that is, when obtaining a 1 times enlarged image, the period is 1/n times the period of the pulsed output signal of the image sensor, and when obtaining a 1/n times reduced image. It is conceivable to obtain a quantized image output signal that displays an image at a desired magnification by sampling at a period twice as large as 1.

However, according to this method, if the object to be scanned, such as a document, has chips or stains, incorrect information will be obtained if the envelope portion corresponding to the chips or stains is sampled at the time of sampling. As a result, there arises an inconvenience that the image quality deteriorates, especially when the image is reduced. This situation is the first
This will be explained in more detail based on the figures.

FIG. 1(b) shows an envelope extracted from a quantized output signal of a primary image sensor composed of a plurality of pixels (bits). This envelope is shown as having a falling edge 1 corresponding to a stain on the document and a rising edge 2 corresponding to a chip in the document. An envelope like this,
For example, to obtain a two-tone image, slice level VB
Figure 1 (C) is obtained by dividing and quantizing the data and sampling it with a sampling clock as shown in Figure 1 (h).
A binary image display output signal as shown in is obtained.

In this signal, logic "0" part 3 and logic "1°.

The outside 4 is a portion generated by sampling the undesired falling edge 1 and rising edge 2 of the envelope.

Therefore, these parts 3 and 4 are false information,
If such erroneous information is included in the image output signal,
In particular, a problem arises in that the image quality deteriorates when the image is reduced.

In order to solve this problem, the present invention converts the quantized signal obtained by quantizing the envelope into a period 1. The above-mentioned erroneous information was eliminated by sampling at a frequency that is at least twice as long as Image magnification conversion for an image input device that obtains a quantized signal and samples the quantized signal at a sampling period corresponding to a desired image magnification to obtain an image output signal that does not degrade image quality when reducing the image. The purpose is to provide a method.

The following two inventions will be explained in detail based on the drawings.

FIG. 2 is a circuit diagram showing an embodiment of the method of the present invention.

This embodiment shows a case where an image output signal is obtained with a reduction ratio of 1/1.5 times and a two-tone image. Further, FIG. 3 is a signal waveform diagram for explaining the operation of the circuit shown in FIG. 2. - In FIG. 2, 5 is a converter constituting the quantization section, 6 is an AND circuit constituting the first sampling section, 7 is a judgment circuit section, 8 is a flip-flop constituting the second sampling section, and 9 is a flip-flop constituting the second sampling section. Branch unit.

lO is a delay circuit.

For example, the period is 1. FIG. 3 (α) shows the envelope extracted from the quantized output signal of the primary line sensor, which is composed of a zero-noise signal train. It is assumed that this envelope has a falling portion 1 corresponding to dirt on the document and a rising portion 2 corresponding to chipping of the document. This envelope is supplied to the ten terminals of the comparator 5 in order to be divided and quantized at the slice level v8.

The slice level vs is supplied to one input terminal of this comparator. Congreta 5 is.

The value of the envelope is compared with the slice level v8, and if it is larger than the slice level, a logic l+1 is generated, and if it is smaller than the slice level, a logic ``ol+'' is generated.

The output signal of the comparator 5 is shown in FIG. 3(b).

In this signal, the logic "01" portion 1' and the logic "1" portion 2' correspond to the undesired falling edge 1 and rising edge 2 of the envelope.

The output signal of the comparator 5 is supplied to one input terminal of an AND circuit 6, and the other input terminal of this AND circuit is
Period of pulsed output signal of image sensor 1. 1
A sub-clock with a cycle of /2 or less, for example 1/2 to (0 cycles M) is supplied. This sub-clock is
Shown in c). In AND circuit 6.

The quantized output from the converter 5 is sampled at the subclock, and a sampling value sequence as shown in Fig. 3 (4) is obtained at its output terminal. 1111100 from the left side
111...... becomes. In this sampling value sequence, the "00" part IM and the "11" part 2' are false information caused by the falling edge part 1 and the rising edge part 2 of the envelope.If these pieces of information are false υ, In order to judge and remove, 1 judgment circuit section 7 is required.

The sampling value sequence is aggregated every period that is the same as the sampling period of the 7-lip-flop 8 of the second sampling section, which will be described later, that is, every 1.5to.

Based on the total results, the correct sampling value within the period is determined by majority vote. FIG. 3 ω shows the sampling value sequence shown in FIG.
The results are shown for each period. The numbers represent the number of sampling values 11'' in each period. Since the total number of sampling values in the period is three,
If the number of sampled values ``11'' is 2 or more by majority vote.

The correct sampling value within that period is @I I+.

1 if the number of sampling values ll01' is 2 or more
is the correct sampling value within that period l □
1 and outputs the correct sampling value.

In order to execute the above operations, 1 judgment circuit section Ma is used.

A counter 11 and a comparator 12 are provided. The counter 11 receives the sampling value "1" from the AND circuit 6.
” Count the number of noculus representing “.

This counter has a period of 1.5 as shown in FIG. 3 (#).
It is reset by the 5to reset pulse and the count is returned to "zero". As a result, the counter 11.

The count values shown in FIG. 3 ω are output in time series. This count value is supplied to the ten input terminal of the comparator 12, and a comparison value of 1.5 is supplied to one input terminal of the comparator 12 in order to take a majority vote of each count value. The converter 12 generates a logic 1111' when the count value from the counter 11 is greater than the comparison value 1.5, and generates a logic 1111' when the count value from the counter 11 is also smaller than the comparison value.
Generates 01. The output signal of the comparator 12 is
Shown in Figure (g). As is clear, this output signal corresponds to the quantized signal shown in FIG. 3(h) with the undesired erroneous information portions 1' and 2' removed. This output signal is supplied to the flip-flop 8 to obtain a quantized image output signal for displaying a 1/1.5 times reduced image.

Sampling is performed using the main clock having a period of 1.5to. The image output signal obtained in this way is
It is clear that it does not contain any erroneous information.

The period of the main clock is the same as the period of the reset pulse (FIG. 3(C)) that resets the counter 11. In this embodiment, this main clock is obtained by dividing the sub-clock whose period is 1/2 to 1/3, which was used to sample the output signal of the converter 5, into 1/3 in the frequency divider 9. This can be obtained by delaying the main clock obtained by using two or more reset pulses by a delay circuit 1o for a certain period of time. The reason for the delay is to shift the reset time of the counter 11 from the pulse generation time of the AND circuit 6 to make the counting operation of the counter appropriate.

The image output signal from the flip-flop 8 obtained as described above is processed by, for example, a microcomputer, and a reduced image of good quality can be displayed on one display device.

As described above, according to the method of the present invention, a quantized signal from which erroneous information caused by dirt or chipping of the object to be scanned is removed is sampled, so a clear image can be obtained without deteriorating the image quality when reducing the image. can be obtained. Furthermore, since an envelope is once extracted from the quantized output of the image sensor and sampling is performed from this envelope at a sampling period according to a desired magnification, it is possible to arbitrarily select a variable magnification. Furthermore, since it can be configured with a simple electronic circuit, an inexpensive conversion system can be obtained.

Although the above has been explained based on a particularly simple practical example in order to explain the present invention, the present invention is not limited only to the above-mentioned embodiment, and those skilled in the art will be able to use various examples within the scope of the present invention. Of course, modifications and changes are possible.

For example, in order to obtain an image output signal for displaying an image with three or more gradations, this can be achieved by providing parallel stages of the circuit shown in FIG.

In addition, in order to judge erroneous information, the sampling period in the first sampling circuit section must be set to 1/2 or less of the period of the pulsed output signal of the image sensor, but this increases the judgment accuracy in the second judgment section. In order to achieve this, it is desirable to reduce the sampling period and increase the number of sampling values obtained. Also.

In addition to obtaining the main clock by frequency-dividing the sub-clock, it is also possible to provide a separate main clock generation circuit.

Furthermore, it is clear that the present invention can be applied not only to a primary line sensor but also to a two-dimensional area sensor.

[Brief explanation of drawings]

FIG. 1 is a signal waveform diagram for explaining the generation of erroneous information due to chipping or staining of a document, and FIG. 2 is a block circuit diagram showing an embodiment of the method of the present invention. FIG. 3 is a signal waveform diagram for explaining the operation of the circuit shown in FIG. 2. In the figure, 5 is a comparator, 6 is an AND circuit, M is a judgment circuit, 8 is a flip-flop, 9 is a frequency divider, 10 is a delay circuit, 11 is a counter, and 12 is a comparator. Patent applicant: Usatsuk Electronic Industry Co., Ltd. Representative Patent Attorney Mori 1) Hiroshi (and 2 others)

Claims (1)

[Claims] From the image sensor at a constant period 1. 1/L (rL is an integer) from the envelope extracted from the pulsed signal generated by
In an image magnification conversion method of an image input device that obtains a quantized image output signal for displaying a scaled-down image, a quantization unit quantizes the envelope. The quantized output signal from this quantizer is l/2t. A first sampling circuit section samples at the following sampling period, and the sampling value from this first sampling circuit section is measured every 3 to periods. A judgment circuit unit that judges and outputs the correct sampling value for each period by majority vote from the aggregated sampling values for each period; and a judgment circuit unit that samples the output signal from this judgment circuit unit at the same period nto as the period. A second sampling circuit section is provided. The judgment circuit section removes erroneous information caused by chipping or dirt on the object to be scanned. An image magnification conversion method for an image input device, characterized in that a quantized image output signal is obtained without deterioration in image quality during image reduction.