JPS6284658A - Image processor - Google Patents

Abstract

PURPOSE:To eliminate the variance of an output and to obtain an output image which becomes neither blurred nor thin by providing a means which detects the ratio of a part which has a larger or smaller reflection factor than a specific value and performs quantization by using the 2nd reference value generated by adding an offset to a reference value. CONSTITUTION:Either of the binary-coding result of a comparator 15 based upon an N% slice level and the binary-coding result of a comparator 17 based upon an (N+alpha)% slice level is outputted according to the decision result of a comparator 16. When an image signal is lower than a voltage determined by a resistance R11, the output of an amplifier 14 which uses the (N+alpha)% slice level is outputted to an output point OUT 3. When the image signal is higher than the level of a rate determined by a resistance R1, the binary coding result of the comparator 15 based upon the normal N% slice level is outputted to the output point OUT 3.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Application Field] The present invention relates to an image processing apparatus, and more particularly to an image processing apparatus that quantizes and outputs an image signal read by an image reading means.

[Prior Art] Devices such as those described in item 1 are widely used for digitally processing images, such as facsimile machines and image file devices.

FIG. 3 shows the configuration of an image reading device used in a facsimile machine or the like. As shown in the figure, a document 1 to be read is conveyed in the direction of the arrow by a conveying means (not shown), and is read by a reading sensor 2. The reading position by the reading sensor 2 is illuminated by a light source 4 such as an LED array, and the reflected light is irradiated onto the reading sensor 2 through the lens 3.

The reading sensor 2 is composed of a CCD reading sensor or the like, and has light receiving elements s1 to Sn arranged at a density of, for example, about 8 pixels/mm, as shown in FIG. The lens 3 is an array type lens such as a SELFOC lens (trade name), and provides optical information on a scanning line to each of these light receiving elements.

The analog image signal read by the reading sensor 2 is subjected to predetermined quantization, but in most cases it can be converted into black and white binary information by comparing it with a predetermined slice level using a circuit such as a comparator. many.

[Problems to be Solved by the Invention] However, especially in a device as shown in FIG. 3, which transports and scans a document, variations in the document in the conveyance path and deviations in the angle relative to the sensor due to wrinkles in the document, etc. Therefore, variations may occur in the amount of reflected light. In addition, signals with delicate levels that are output as black or white near the aforementioned slice level may be determined to be white or black even if the image information has a certain reflectance, depending on the noise level of the power supply and ground lines, etc. There may be cases where Therefore, for the reasons mentioned above, uniform binarization results cannot be obtained even when scanning an original with a constant density, and when scanning a text original, the edges of characters that should be black may be output as white. , characters became thin or parts of ruled lines became blurred.

For example, as shown in FIG. 5, when a document l has a margin part and a completely black part, and the boundary line BL exists in the scanning line direction,
Assume that when this document is conveyed in the direction of the arrow to the light receiving element S1 (shown enlarged) of the reading sensor 2, this boundary portion is read at one of the positions a to d of the element.

Then, as the position of the boundary line moves from a to d, the proportion of the black portion on the light receiving element Sl increases, so
The scanning output of the element gradually decreases as shown in FIG. When this scanning output is binarized at slice level 5 corresponding to N% of the reflectance of the white original, white information is output when the boundary line BL is read by the light receiving element Sl at positions a and b. is read at positions c and d and output as black information.

Here, suppose that the boundary line BL is read at position b, and the reading sensor 2
If the output signals of all the elements vary above and below the 8% line of 5, the binarization determination becomes unstable. Therefore, if the image information of the boundary line BL is output to a recording device or the like, it will be recorded in a jagged manner.

[Means for Solving the Problems] In order to solve the above-mentioned problems, the present invention provides an image processing device that quantizes and outputs an image signal read by an image reading means. means for detecting a proportion occupied by a portion of reflectance above or below; first means for quantizing the image signal using a first reference value; and means for adding an offset amount to the first reference value. A configuration is adopted in which a means for quantizing the image signal using a second reference value is provided, and the first and second quantization means are switched depending on the output of the detection means.

[Function] According to the above configuration, even if the ratio of white or dot portions on the reading means fluctuates in the vicinity of the quantization standard due to variations or wrinkles in the read original, the quantization standard is moved and the output is adjusted. Variations can be prevented, all-white information can be output as all-white information, all-black information can be output as all-black information, and an output image without blur or thinning can be obtained.

[Embodiments] The present invention will be described in detail below based on embodiments shown in the drawings.

FIG. 1(A) is a block diagram showing an embodiment of an image processing apparatus according to the present invention.

The output of the reading sensor 2, which has the same configuration as the conventional example shown in FIG. be done.

The output of the peak value detection circuit 8 is input to a D/A converter 10.12, and data read from the memory 7.9 is input to each of these two D/A converters as a reference value. . Among these, memory 7 is a digital memory,
The image output when the Kanayama original is read by the reading sensor 2 shown in FIG. 3 is stored as a digital quantity. Memory 9 is a similar memory, and stores image output when an all-black original is read. Both memories store image information for at least l scanning lines at addresses corresponding to read bit positions.

The information in the memory 7.9 is formed from an image output obtained by scanning a known standard white (black) plate or the like using the reading sensor 2 of the apparatus. Therefore, the information in the memory 7.9 reflects variations in the amount of light emitted by the light source and variations in element sensitivity.

The peak value detection circuit 8 is composed of a known integrating circuit, a diode, etc., and is used to store the peak value within one scanning line when reading a standard white plate.

The D/A converters 10 and 12, together with the peak value detection circuit 8, are circuits similar to those widely used for shading distortion correction and the like in this type of image processing apparatus.
The peak value output from the peak value detection circuit 8 is set as a reference voltage v,
EF converts the memory output into analog.

The D/A converters 10 to 12 are, for example, as shown in FIG.
Configure it like this.

The reference voltage VnEF output from the peak value detection circuit 8 is divided by a resistor circuit 36 weighted by a resistor R12R, and a switch 38 switches each divided voltage according to 4-bit information output from the memory 7.9. By applying this to the operational amplifier 39, the memory contents are output as an analog value.

The amplification factor of the operational amplifier 39 is determined by the ratio of each input resistance to the feedback resistance RR. The offset (H) of this amplification factor can be adjusted by the resistor 37.

Now, the D/A converter 10.12 converts the input value into an analog value with the above configuration and outputs it to the output points OUT l, 0UT2.
Output to. These output points are connected by a resistor 11, and the variable partial voltage at the midpoint is led to the output of a comparator 16.

The resistor 11 (R1) is for outputting an arbitrary level between the gold mine information and the all black information stored in the memory 7.9.

Here, the output voltage at the output point OUT1.0UT2 is shown in FIG. 2(A). The horizontal axis in the figure indicates the read bit position of l-scan by the read sensor 2. Also, resistor 11
The voltage division by (R11) is a level obtained by dividing the voltage of 0UT1.0UT2 at a fixed ratio.

Further, the voltage at the output point OUT1 is applied to the resistor 13 (R2), and the potential at the desired intermediate tap is guided by the power of the non-inverting amplifier 14, which is composed of an operational amplifier or the like. This resistor 13 determines the slice level corresponding to N% of the reflectance of the all-white original (or standard white plate) stored in the memory 7. If the reflectance of all white is 100%, a level of 50 to 80% is generally used as the slice level, and the slice level of N% determined by the resistor 13 is a value within this general range.

Therefore, the comparator 15 binarizes the image signal at the slice level of N% determined by the resistor 13 (R2).

On the other hand, the non-inverting amplifier 14 is for adding an offset amount of α% to this N% slice level and outputting it with the same polarity.The gain of the amplifier 14, that is, the offset amount of α% is determined by resistors R3 and R4. be done. As is well known, the gain can be set by (1+R4/R3). The output of non-inverting amplifier 14 is led to the output of comparator 17.

Therefore, the comparator 17 converts the image signal into (N+α)
The image signal is binarized at a slice level of %, and the output is input manually to the AND gate 18. and gate 1
The output of the comparator 16 is led to the other human power of 8. In addition, the output of the comparator 15 is
1 is input. The output of the comparator 16 inverted by the input gate 20 is led to the remaining inputs of the AND gate 21. Therefore, 8% by comparator 15
Depending on the determination result of the comparator 16, either the binarization result based on the slice level or the binarization result based on the (N+α)% slice level by the comparator 17 is output.

18. The logical sum of the outputs of 21 is output to the output point 0UT3 via the OR gate 19, and is stored in the memory.
Receives various processing such as facsimile transmission.

The comparator 16 calculates the proportion of the black area on the reading sensor 2 as explained in FIGS.

This is to realize a function that automatically makes a judgment by comparing the image signal output from the reading sensor 2 with a predetermined reference voltage determined by taking into account the partial pressure of 0UT2, that is, shading.

Next, the operation in the second configuration will be explained.

For example, assume that the output of the tenth light receiving element S10 of the reading sensor 2 of the third younger brother is read out and input to the circuit of FIG. 1(A).

Since the reading of the memory 7.9 is performed in synchronization with the image output of the reading sensor 2, the information of the 10th and 11th bits is manually input to the D/A converter 10.12, so that the output is Point 0UTl, 2 has memory 7.9 (
7) A voltage obtained by D/A-converting the information of the 10th value by using the output of the peak value detection circuit 8 as a reference voltage is output.

As a result, the voltage level generated by the resistance division ratio determined in advance by the resistor 11 is input to the ten input terminal of the comparator 16.

At this time, as shown by the symbol B in FIG. 2(A), the image signal is smaller than the voltage determined by the resistor R11 (at this time, the boundary !J!BL in FIGS. (as in the case where the proportion occupied by black is greater than the value determined by the resistor R11), the comparator 16 outputs a high level. This opens AND gate 18 and closes AND gate 21.

As a result, the output of the amplifier 14 using the slice level of (N+α)% is output to the output point 0UT3. That is, the image signal is binarized at a slice level that is α% higher than normal.

On the other hand, when the image signal is larger than the level determined by the resistor R1 (in this case, the area it occupies on the light receiving element is large), as indicated by the symbol W in FIG. 2(A), the comparator 16 outputs a low level, the AND gate 21 is opened, and the AND gate 18 is closed, and the 2(fi conversion result by the normal N% slice level comparator 15) is output to the output point 0UT3.

Here, FIG. 2(B) shows the state of binarization when the slice level is automatically changed as described above. In FIG. 2(B), the line 6 is the slice level offset by the amplifier 14 when the image signal level is smaller than the ratio determined by the resistor R1, which is a slice level that is α% higher than the normal N%. It becomes.

Therefore, when the image signal level is smaller than the ratio determined by the resistor R1, that is, when the proportion of black light on the light receiving element -1° is large, the slice level is increased by α% to the normal N% slice level ( Subtle image signals near symbol 5) can also be reliably determined to be black.

For example, in the conventional example, when the boundary line BL between all white and all black in FIGS. 5 and 6 is at position b on the sensor, it is reliably output as black information.

The above configuration can be applied as a binarization device for various devices such as a facsimile device and an image file device.

[Effects of the Invention] As is clear from the above description, according to the present invention, in an image processing device that quantizes and outputs an image signal read by an image reading means, the image signal is quantized on the reading means. a first means for quantizing the image signal using a first reference value; and a second means for quantizing the image signal by using the first reference value and Since a configuration is adopted in which a means for quantizing the image signal using the basic value is provided and the first and second quantizing means are switched depending on the output of the detecting means, the read original Even if the proportion of white or point portions on the reading device fluctuates near the quantization standard due to fluttering or wrinkles, the quantization base is moved to prevent variations in the output. information, all-black information can be output as all-black information, and an image processing device that can obtain an output image without blur or thinning can be provided.

[Brief explanation of drawings]

FIG. 1(A) is a block diagram showing one example of the configuration of an image processing apparatus according to the present invention, and FIG. 1(B) is a block diagram showing an example of the configuration of an image processing device according to the present invention.
Circuit diagram showing an example of the structure of a /A converter, Figure 2 (A)
, (B) are explanatory diagrams showing the operation of the device in FIG. 1(A), FIG. 3 is an explanatory diagram schematically showing the structure of the image reading device of the image processing device, and FIG. FIGS. 5 and 6 are explanatory diagrams showing the structure of the reading sensor shown in the figure, respectively, and are explanatory diagrams showing the drawbacks of conventional image processing. 2...Reading sensor 7,9...Memory 1O11
2...D/A converter 15-17...Comparator

Claims (1)

[Claims] An image processing device that quantizes and outputs an image signal read by an image reading means, comprising: means for detecting a proportion of a portion of a reflectance above or below a predetermined value on the reading means; and a first reference value. and a means for quantizing the image signal using a second reference value obtained by adding an offset amount to the first reference value, and an output of the detection means. An image processing device characterized in that the first and second quantization means are switched and used depending on the situation.