KR100267997B1 - Apparatus and method for scanning conversion - Google Patents

Abstract

PURPOSE: A scan converter is provided to detect a vertical transition edge of characters in a down scan converting process and to apply a filtering, so as to prevent a blink generated while converting a sequential scanning signal by an interlaced scanning. CONSTITUTION: A scan converter initializes a histogram array(S51). If a screen is turned on(S52), the scan converter stores the first line of the screen(S53). If the next line starts(S54), the scan converter reads an optional pixel gray level of the first line(S55). A present-inputted line reads and stores the optional pixel gray level(S56). The scan converter detects a vertical transition edge of characters according to a case flag decided by using the pixel gray level of the first line and a pixel gray level of the present-inputted line(S58).

Description

Apparatus and method for scanning conversion

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a scanning apparatus and method of a TV, and more particularly, to a scanning converter and a method for converting a sequential scanning signal of a PC into an interlaced scanning method and displaying the same on a TV screen.

Recently, in consideration of the trend of gradually combining PCs and TVs, and consumers requiring a more easy-to-use PC, multimedia TVs combining PCs and TVs that are familiar to users are being developed.

The multimedia TV combined with the PC and the TV is down-scan converted from the sequential scanning method of the PC to the interlaced scanning method of the TV in order to display the signal of the PC (Video Graphics Array) card on the TV screen. It needs to be displayed on the screen.

Hereinafter, a method of converting from a sequential scanning method to an interlaced scanning method according to the related art will be described with reference to the accompanying drawings.

1 is a view for explaining a basic conversion method of the conversion method from the sequential scanning method to the interlaced scanning method according to the prior art, Figure 2 is a weighted average of the conversion method from the sequential scanning method to the interlaced scanning method according to the prior art It is a figure for demonstrating the method using a filter.

There are two types of conversion from the sequential scanning method to the interlaced scanning method: a basic method and a method using a weighted average filter.

First, the basic scheme configures one field using only odd-numbered lines of the sequential scanning video frame as shown in FIG. 1A, and changes only the even-numbered lines of the sequential scanning video frame as shown in FIG. Configure one field.

In addition, the method using the weighted average filter forms a field using only the odd-numbered lines by weighted average filtering the sequential scan data of the three scan lines as shown in FIG. 2 (a), and the three scan lines as shown in FIG. The weighted average filtering of sequential scan data of s constitutes another field using only even-numbered lines.

This method is flattened over several interlaced scan lines, so no flicker after conversion.

The basic method of converting from the sequential scanning method to the interlaced scanning method according to the prior art has a problem that there is a flickering effect on a long horizontal direction line of one line thickness, causing severe flickering on the entire screen.

In addition, a method of using a weighted average filter causes a problem in that it is difficult to recognize a character because vertical transition edges of the character spread upward and downward.

The present invention has been made to solve the above problems by separating the vertical transition boundary region of the character, applying a clear filtering method to the vertical transition boundary region of the character, and applying a filtering method with less flicker in other regions. It is an object of the present invention to provide a scanning converter and a method thereof.

1 is a view for explaining a basic conversion method of the conversion method from the sequential scanning method to the interlaced scanning method according to the prior art

2 is a view for explaining a method using a weighted average filter of the conversion method from the sequential scanning method to the interlaced scanning method according to the prior art

3 is a block diagram showing a configuration of a scan conversion device for performing scan conversion based on boundary detection using two lines according to the present invention;

Figure 4 is a block diagram showing the configuration of a scan conversion device for performing a scan conversion according to the boundary detection using three lines in accordance with the present invention

5 is a flowchart of a method of determining a case flag value using a histogram according to the present invention.

FIG. 6 is a diagram illustrating pixel position notation for describing FIG. 5. FIG.

FIG. 7 is a diagram for describing quantization of gray levels of FIG. 5. FIG.

FIG. 8 is a diagram illustrating initialization of the histogram array of FIG. 5. FIG.

FIG. 9 is a diagram showing the types of case flags representing the form in the region of FIG.

FIG. 10 is a flowchart for explaining determination of a case flag and a character value of FIG. 5. FIG.

11 is a flowchart of a scan conversion method for performing scan conversion according to another boundary detection according to the present invention.

12 is a flowchart of a scan conversion method for performing scan conversion based on boundary detection using another two lines according to the present invention.

13 is a diagram for explaining boundary detection using three lines according to the present invention.

Explanation of symbols for main parts of the drawings

31: VGA card 32: Y detection unit

33: ADC 34, 42, 43: line memory

35 control unit 36 CH7001

37: TV monitor 41: delay unit

A feature of the scan converter according to the present invention is that the luminance signal detector receives a video signal of a sequential scan method, converts the digital signal into a digital signal after detecting the luminance signal indicating a gray level, and the controller is configured to output one line of the luminance signal detector. The digital signal and the digital signal of the previous line stored in the line memory unit are input to determine the vertical transition boundary, and then the control signal according to the determination result is output to the scan converter to convert the weighted average filter mode.

A characteristic of the scanning conversion method according to the present invention is that when the screen starts, the first line of the screen is stored, and when the next line starts, the gray level of the pixel of the currently input line is read and the gray level of the pixel of the currently input line is read. It is to detect the vertical transition boundary according to the duration and the change amount and the case flag determined in comparison.

Hereinafter, a scan converter and a method thereof according to the present invention will be described with reference to the accompanying drawings.

3 is a block diagram showing a configuration of a scan converter for performing scan conversion according to boundary detection using two lines according to the present invention, wherein the scan converter using two lines is R, G, B for a video signal of a PC. A VGA card 31 that outputs a color signal, and a Y detector 32 that receives the R, G, and B signals output from the VGA card 31 and outputs a luminance signal Y indicating a gray level of the current input signal. And an analog / digital converter (ADC) 33 for converting the analog luminance signal output from the Y detector 32 into a digital signal, and a line memory for storing the digital signal for one line output from the ADC 33. (34) and the digital signal for one line output from the ADC 33 and the digital signal currently output from the ADC 33 are input to determine whether the vertical transition boundary, and then the control signal (MS 0, MS according to the determination result). 1) a control unit 35 for outputting, R, G, output from the VGA card 31, CH7001 (36) and CH7001 () for receiving the B color signal and outputting the luminance signal and the color signal filtered by the weighted average filter method selected during the conversion from the sequential scanning method to the interlaced scanning method according to the control signal output from the control unit 35; And a TV monitor (37) for displaying the luminance signal and the color signal outputted at 36).

Here, the Y detection unit 32 inverts and amplifies the R, G, and B color signals input according to the input resistors Rr, Rg, and Rb having a predetermined ratio and the feedforward resistor R0 at a predetermined ratio. An inverting amplifier 321 and a second inverting amplifier 322 for inverting and amplifying and outputting the luminance signal inverted and amplified by the first inverting amplifier 321.

4 is a block diagram showing a configuration of a scan conversion apparatus for performing scan conversion based on boundary detection using three lines according to the present invention, in which a scan conversion apparatus using three lines uses scan conversion using two lines of FIG. A delay unit 41 receives the R, G, and B signals outputted from the VGA card 31 and outputs them after a predetermined time delay, and stores a digital signal for one line currently output from the ADC 33. After that, the first line memory 42 outputs to the controller 34 and the second line memory 43 stores a digital signal for one line output from the first line memory 42 and outputs the digital signal to the controller 34. To add more).

Referring to the accompanying drawings, the operation of the scan converter configured as described above is as follows.

The VGA card 31 outputs the R, G, and B color signals of the PC video signal to the Y detector 32 and the CH7001 36, and the Y detector 32 receiving the R, G, and B color signals receives luminance. The signal is detected and output to the ADC 33.

The ADC 33 converts the input analog luminance signal into a digital signal and outputs it to the control unit 35 and the line memory 34, and the line memory 34 stores and outputs the digital signal for one line.

The controller 35 determines whether it is a vertical transition boundary by using the digital signal currently output from the ADC 33 and the digital signal of the previous line stored in the line memory 34.

When the vertical transition boundary is detected by the controller 35, the control signals MS 0 and MS 1 for selecting a filter for sharpening the vertical transition boundary are output to the CH7001 36.

CH7001 (36) is an encoder that converts VGA signal into NTSC / PAL signal.It is equipped with four weighted average filters for converting sequential scanning method to interlaced scanning method and converting scanning method. The selection is selected from MS 0 and MS 1 output from the control unit 35.

In the weighted average filter method when the scanning method is converted through the control signals MS 0 and MS 1, MS 0 and MS 1 may be 0 0, 0 1, 1 0, 1 1.

When MS 0 and MS 1 are 0 0, the filter mode is 0: 1 0 averaging, flicker is very severe, and the sharpness of the vertical transition boundary is very sharp.

When the MS 0 and MS 1 are 0 1, the filter mode is 1: 3: 1 averaging, and the flicker is more than when the MS 0 and MS 1 are 1 1, and the vertical transition boundary The sharpness is clearer than when MS 0 and MS 1 are 1 1.

When the MS 0 and the MS 1 are 10, the filter mode is 1: 2: 1 averaging, there is almost no flicker, and the sharpness of the vertical transition boundary is very blurry.

Finally, when MS 0 and MS 1 are 1 1, the filter mode is 1: 1: averaging, there is little flicker, and the sharpness of the vertical transition boundary is blurred.

Accordingly, the weighted average filter method of 0: 1: 0 is significantly more flickering than the other three weighted average filter methods, while the vertical transition boundary of the object is clear.

In the other three filters, the 1: 2: 1 weighted average filter method has the least flicker and the vertical transition boundary of the object is blurry, but there is no significant difference between the three methods.

Therefore, CH7001 (36) uses the 0: 1: 1 weighted average filter method when converting the sequential scanning method to interlaced scanning as a control signal for the vertical transition boundary of the controller 35. For the rest of the frame, only the other weighted average filter method needs to be applied.

That is, the CH7001 36 performs interlaced scanning of the sequential scanning R, G, and B color signals output from the VGA card 31 according to a control signal according to whether the vertical transition boundary of the character of the controller 35 is determined. The signal is converted and output to the TV monitor 37.

In addition, in order to determine the vertical transition boundary using three lines, the control unit 35 stores the signals of two previously input lines in the first and second line memories 42 and 43, respectively. If necessary, the controller 35 outputs the signal to the controller 35, during which the delay unit 41 delays the R, G, and B signals input from the VGA card 31 to the CH7001 34.

Here, when the control unit 35 detects the vertical boundary using two lines or detects the vertical boundary using three lines, the control unit 35 digitally converts only the G signal output from the VGA card 31 and converts the ADC 33 into an ADC 33. The vertical transition boundary may be detected using the digital G signal currently output from the digital signal and the digital G signal of the previous line stored in the line memory 34 or the first and second line memories 42 and 43.

A detailed method for outputting the control signal according to the vertical transition boundary detection of the controller 35 will be described with reference to the accompanying drawings or flowchart.

FIG. 5 is a flowchart illustrating a method of determining a case flag value using a histogram according to the present invention, FIG. 6 is a diagram illustrating pixel position representation for describing FIG. 5, and FIG. 7 is a gray level diagram of FIG. 5. FIG. 8 is a diagram illustrating initialization of the histogram array of FIG. 5, FIG. 9 is a histogram representing a shape in the region of FIG. 5, which shows types of case flags, and FIG. 10 is a diagram of FIG. 5. This is a flowchart for explaining the determination of case flags and character values.

In the vertical transition boundary detection method of the control unit 35, the boundary detection method according to the gray level as shown in FIG. 5, the boundary detection method according to the duration length (clrl) in the current pixel as shown in FIG. There are three boundary detection methods according to the gray level variation, and each can be applied at the same time.

First, referring to FIG. 5, a vertical transition boundary detection method will be described. As shown in FIG. 6, the screen is represented by x and y coordinates, and the pixel corresponding to the coordinate (x, y) is represented by pixel (x, y), and pixel ( The pixel of the previous line having the same position on the x axis as x, y) is represented by pixel (x, y-1).

As shown in FIG. 8, the histogram of the gray level of the pixels of one line of the image includes six elements having a field for frequency and a gray level field as subfields.

Histogram array elements are histo [0], histo [1],... , histo [5], and the fields for the frequency of the subfield are histo [0] .freq, histo [1] .freq,... , histo [5] .freq, and the gray level fields are histo [0] .g_level, histo [1] .g_level,... , histo [5] .g_level.

Here, histo [0] .g_level divides 256 gray levels into a predetermined range as shown in FIG. 7, and gray levels between 0 and 25 are 0, gray levels between 26 and 76 are 51, gray levels between 77 and 127, and so on. Gray levels 102, 128 and 178 are 153, gray levels between 179 and 229 are 204, gray levels between 230 and 225 are 255, and gray levels are represented by six representative values.

Histo [0] .freq, histo [1] .freq,... Of the histogram array. , histo [5] .freq are all initialized to 0 (S51), and when the screen starts (S52), the first line of the screen is stored in the line memory 34 (S53).

It is determined whether the next line starts (S54). When the next line starts, the gray level of the pixel (x, y-1) one line before is read from the line memory 34 (S55).

The gray level of the current line pixel (x, y) is read, and the gray level of pixel (x, y) is stored in the line memory 34 (S56).

By updating the character run-length of the image in pixel (x, y) (S57), after completing the histogram of the representative value for a line, the element of the highest frequency is located in histo [0]. Arrange the histogram so that histo [1] has the second highest frequency element.

In the alignment of the histogram, a vertical transition boundary is detected according to a case flag value indicating a form of a region for a line and a character value indicating a gray level of a character in the corresponding line (S58).

As shown in FIG. 9, the case flag value has values of case 0, case 1, case 2, and case 3, and in this case, control signals MS 0, 1 are 0 0, 0 1, 1 0, 1 is output to CH7001 (36).

Case 0 of the case flags is the area where the gray level of the background is larger than the gray level of the character, case 1 is the area where the gray level of the background is smaller than the gray level of the character, and case 2 contains only the bright pixels with the same gray level. Case 3 is an area where only dark pixels with the same gray level exist.

In the case 0 and case 1, the charater related to the gray level of the character in the corresponding line has a value within 25 of the histo [1] .g_level value among 1, 51, 102, 153, 204, and 255.

On the other hand, if the case flag of one line is case 2, the gray level of the character in the next line is predicted to be less than 128, and in case 3, the gray level of the character in the next line is predicted to be 128 or more.

The frequency histo [0] .freq of the gray level value of the array of histograms is continuously updated for each pixel of one line (S59), and when the line is finished (S60), the histogram is analyzed to determine the case flag value (S61).

After aligning the histogram array in the step S58, vertical transition boundary detection is performed according to the case flag value and the character value for the line on which the histogram is created. This will be described with reference to the accompanying flowchart.

FIG. 10 is a flowchart illustrating the determination of the case flag and the character value of FIG. 5. First, it is determined whether histo [1] .freq of the histogram array is 0 (S100), and the determination result (S100) histo [1]. If .freq is 0, it is determined whether histo [0] .g_level is 128 or more (S101).

As a result of the determination (S101), if histo [0] .g_level is 128 or more, the case flag (case_flag) corresponds to case 2 (S102). If histo [0] .g_level is not 128 or more, case_flag corresponds to case 3.

If the histo [1] .freq is not 0, the determination result (S100) determines whether histo [0] .g_level is 128 or more (S104). If histo [0] .g_level is 128 or more, the gray level (g_level) is less than 128. In step S105, it is determined whether there is a histogram array element having a frequency other than zero (freq).

As a result of the determination (S105), if the duration is not 0 and there is a histogram array, g_level of the element having the highest frequency among the elements having a g_level of less than 128 is a character value, and case_flag corresponds to case0 (S106).

If the gray level (g_level) is less than 128 and there is no histogram array element having a non-zero persistence (freq), the character is histo [1] .g_level (S107) and histo [1] .g_level. It is determined whether is less than histo [0] .g_level (S108).

As a result of the determination (S108), if histo [1] .g_level is not less than histo [0] .g_level, case_flag corresponds to case 1 (S109), and if histo [1] .g_level is less than histo [0] .g_level, case_flag Corresponds to case 0 (S110).

In addition, if the histo [1] .freq is not 0 and histo [0] .g_level is not 128 or more, the histogram in which the gray level (g_level) is 128 or more and the duration (freq) is not 0 is determined. It is determined whether there is an array element (S111).

As a result of the determination (S111), if there is a histogram array having a non-zero persistence freq, g_level of the element having the highest freq among g_level of 128 or more is a character value, and case_flag corresponds to case1 (S112).

If the gray level (g_level) is 128 or more and there is no histogram array element having a non-zero persistence (freq), the character is histo [1] .g_level (S113) and histo [1] .g_level. It is determined whether is less than histo [0] .g_level (S114).

As a result of the determination (S114), if histo [1] .g_level is less than histo [0] .g_level, case_flag corresponds to case 0 (S115), and if histo [1] .g_level is not less than histo [0] .g_level, case_flag Corresponds to case 1 (S116).

In addition, another vertical transition boundary detection method of the control unit 35 will be described with reference to the accompanying drawings.

FIG. 11 is a flowchart of a scan conversion method for performing scan conversion based on boundary detection using another two lines according to the present invention. First, it is determined whether one line starts (S200), and when the line starts, the variable x is 0. In step S201, the current analog / digital converted (ADC) signal is set as the gray level of the currently input pixel (S202).

After that, it is determined whether the variable x is 0 (S203). If x is 0, the current line run length (clrl) is set to 1 (S204), and if the determination result (S203) is not 0, It is determined whether the gray level of the currently input pixel is the same as the gray level of the previous pixel (S205).

If the gray level of the currently input pixel is not the same as the gray level of the previous pixel, the step S204 is repeated. If the gray level of the currently input pixel is the same as the gray level of the previous pixel, the current input pixel is determined. Increase the gray level by one (S206).

After performing the above steps S204 and S206, the gray level of the currently input pixel is set to the gray level of the previous pixel (S207), and it is determined whether clrl is 1 (S208).

If the determination result (S208) clrl is 1, the control unit 35 outputs a control signal to the CH7001 (36) to operate in the 0: 1: 1 mode among the four methods of the weighted average filter (S209).

On the other hand, if the determination result (S208) clrl is not 1, it is determined whether clrl is more than a predetermined threshold (S210).

In general, the duration of a character in an image is about 10 pixels on average, and in the case of a large character over 10 pixels, the character can be sufficiently identified without a 1: 1: 0 weighted average filtering from a portion beyond 10 pixels.

In addition, in the case of a long continuous line, the part that exceeds 10 pixels is filtered in a blurry manner to distinguish from the characters so that flickering in the entire screen is reduced.

That is, in the step S210, when detecting the vertical transition boundary, clrl is limited to 10 pixels or less to detect only pixels having a duration of 10 pixels or less.

If the determination result (S210) clrl is less than 10 pixels, the control unit 35 outputs a control signal to the CH7001 (36) to operate in the 1: 1: 0 mode of the four methods of the weighted average filter (S211).

On the other hand, if the determination result (S210) clrl is more than 10 pixels, or if the mode is set to 1: 1: 0, the variable x is increased by 1 (S212), and it is determined whether the line is finished (S213). The process is repeated from the step S202, and if the line is finished, the process is repeated from the step S200.

In addition, another vertical transition boundary detection method of the control unit 35 will be described with reference to the accompanying drawings.

FIG. 12 is a flowchart of a scan conversion method for performing scan conversion according to boundary detection using another two lines according to the present invention. First, it is determined whether a screen starts (S52). Is stored in the line memory 34 (S301).

It is determined whether the next line starts (S302). When the next line starts, the gray level of the pixel (x, y-1) one line before from the line memory 34 is read (S303).

The gray level of the current line pixel (x, y) is read, and the gray level of pixel (x, y) is stored in the line memory 34 (S304).

The gradient amount is detected using the gray levels of the pixel (x, y) and the pixel (x, y-1) read in the above step (S305).

Looking at the step (S305) in more detail, the amount of gradient (Gradient) is an important amount when detecting the boundary.

The magnitude of the gradient may be expressed by Equation 1 below.

Equation 1 represents the maximum increase rate per unit length of f (x, y) in the ∇f (x, y) direction as the magnitude of the gradient.

는

로 근사화될수 있는데, 수직 천이 경계만을 검출하는 경우,

만을 사용하여

가 일정 문턱값(Threshold)이상일 때 (x,y)지점을 수직 천이 경계로 검출할 수 있다.remind

Is

Can be approximated by

Using only

When x is greater than or equal to a certain threshold, the (x, y) point can be detected as a vertical transition boundary.

Meanwhile,

As it is a digital image

Can be approximated by Where g (pixel (x, y)) represents the gray level of pixel (x, y).

가 일정 문턱값 이상일 경우 제어부(35)는 pixel(x,y)를 수직 천이 경계로 검출할 수 있어 CH7001(36)로 가중 평균필터의 4가지 방식중 0 : 1 : 0 모드로 동작하게 하는 제어신호를 출력한다(S307).therefore,

Is greater than or equal to a predetermined threshold, the control unit 35 can detect pixel (x, y) as the vertical transition boundary, and control the CH7001 (36) to operate in 0: 1: 1 mode among four methods of the weighted average filter. The signal is output (S307).

가 일정 문턱값 미만일 경우 제어부(35)는 CH7001(36)로 가중 평균필터의 4가지 방식중 1 : 1 : 0 모드로 동작하게 하는 제어신호를 출력하고(S308), 라인의 끝인지를 판단한다(S309).And, the determination result (S306)

Is less than a predetermined threshold, the control unit 35 outputs a control signal to the CH7001 36 to operate in a 1: 1: 0 mode of the four methods of the weighted average filter (S308), and determines whether it is the end of the line. (S309).

If the determination result (S309) is not the end of the line, the operation is repeated from the step (S303), and if the end of the line is terminated.

As described above, three boundary detection methods using two lines can be applied simultaneously.

In the case of using three of them, the boundary detection method for the three will be briefly described.

FIG. 13 is a diagram for explaining boundary detection using three lines according to the present invention. When performing boundary detection using three lines, pixel (x, y) corresponds to the current pixel in time but is a pixel in the past in time. It is determined whether or not the boundary for (x, y-1) is determined.

Then, the gray level of the pixel of the previous line (pixel (x, y-1)) and the pixel of the previous line (pixel (x, y-2)) having the same position on the x-axis of the pixel (x, y) currently input. The gray level of) is read in the first and second line memories 42 and 43, respectively.

First, the boundary detection method using three lines is almost the same as the boundary detection method using two lines in FIG. 10, and there is a difference in only generating a histogram to which pixel (x, y-2) belongs. The control signal for boundary detection is output by determining a case flag value and a character value.

In addition, the boundary detection method using two lines uses a character duration length (clrl) of pixel (x, y) as shown in FIG. -1) is judged whether or not the vertical transition boundary using the character duration of the belonging.

에 대한 변화량 검출하여 일정 문턱값과의 비교결과에 따른 필터 모드를 설정하였으나 3라인을 이용하여 경계검출 하는 방법은

와

값과 일정 문턱값과의 비교결과에 따라 CH7001(36)의 필터 모드를 설정함으로써 PC의 순차 주사 방식의 데이터를 TV 모니터(37)에 표시해주기 위한 비월 주사 변환 방식으로 변환된다.In addition, the boundary detection method using two lines as shown in FIG.

Although the filter mode is set according to the result of comparison with a certain threshold by detecting the change amount of, the boundary detection method using 3 lines

Wow

By setting the filter mode of the CH7001 36 in accordance with the comparison result between the value and the constant threshold value, it is converted into an interlaced scanning conversion method for displaying data of the sequential scanning method of the PC on the TV monitor 37.

The scanning conversion apparatus and the method according to the present invention interlaced a sequential scanning signal by detecting a vertical transition boundary of a character and applying adaptive filtering during down scan conversion for displaying a signal of a VGA card of a PC on a TV screen. It is possible to reduce the lumping phenomenon caused by flickering or the filtering of characters generated by the conversion, and can be implemented with only the line memory without using the frame memory, so that the image quality of the multimedia TV can be implemented at a relatively low cost.

Claims (9)

Initializing the histogram array, saving the first line of the screen when the screen starts, Reading any pixel gray level of the stored previous line when the next line starts, Reading and storing the arbitrary pixel gray level in a currently input line; And detecting a vertical transition boundary of a character according to a case flag determined by using the gray level of the pixel of the previous line read and the gray level of an arbitrary pixel of the currently input line. The pixel of claim 1, wherein the case flag includes a case 0 in which the gray level of the background is larger than the gray level of the character, a case 1 in which the gray level of the background is smaller than the gray level of the character, and a bright pixel having gray levels. And case 3, which is a region where only the pixels exist, and case 3, which is an region in which only dark pixels having gray levels exist. The scanning conversion method of claim 1, wherein the histogram region is divided into six regions of one line of pixels. The method of claim 1 or 2, wherein the detecting of the vertical transition boundary is Creating a histogram array by detecting a frequency of the gray level of the pixel currently input using the gray level of the pixel of the previous line and the gray level of the pixel of the currently input line; Determining if there is a second high frequency histogram array, and if the gray level of the first high frequency histogram array is greater than 128, And determining that the gray level of the first histogram array having the highest frequency is 128 or more, and the case flag case 2 if the gray level of the histogram array having the first large frequency is not 128 or more, , Determining that the gray level of the first high frequency histogram array is greater than or equal to 128 if there is no histogram array having the second largest frequency; If the gray level is 128 or more, the gray level is less than 128, and if it is determined whether there is a non-zero histogram array, the gray level of the element whose gray level is less than 128 and the maximum duration is the character value. Determining that it is case0, If the gray level is less than 128 and there is no histogram array whose frequency is non-zero, the gray level and the first gray level of the histogram array having the second largest frequency are set as the gray level of the histogram array having the second highest frequency. Comparing the gray levels of the histogram array with high frequency, And the case flag is case 0 if the gray level of the histogram array having the second largest frequency is smaller than the gray level of the histogram array having the first large frequency, and the case flag is case 1 if the gray level is large. The method of claim 1, wherein detecting the vertical transition boundary is And a vertical transition boundary is detected according to a case flag determined by using the read gray level of the pixel of the previous line and the pixel gray level of the previous line of the previous line. Determining whether the first line is input, and setting the duration of the current input pixel to 1 when the first line is input, and comparing the gray level of the previous pixel to the gray level of the currently input pixel if the first line is not included. If the gray level of the previous pixel is equal to the gray level of the currently input pixel, the duration length of the currently input pixel is set to 1; if the gray level of the previous pixel and the gray level of the current input pixel are different, the gray level of the current pixel is different. Adding 1 to The gray level of the currently input pixel is set as the gray level of the previous pixel, and it is determined whether the duration length of the currently input pixel is 1, and if it is 1, it operates in 0: 1: 1 mode of the weighted average filter, and if it is not 1, a certain threshold Determining whether the value is within a value and setting a mode of the weighted average filter. 7. The method of claim 6, wherein setting the mode of the weighted average filter is If the determination result is within a certain threshold value, it operates in a different mode of the weighted average filter other than 0: 1: 0 mode, and if the predetermined threshold value is exceeded, the next pixel is inputted, and the gray level is repeated. Scan conversion method. The method of claim 6, wherein the determining of whether the predetermined threshold value is within And determining whether the character duration is within 10 pixels. 7. The method of claim 6, wherein setting the mode of the weighted average filter is It is determined whether the duration of the previous line pixel is 1, and if it is 1, it operates in 0: 1: 1 mode of the weighted average filter, and if it is not 1, it is determined whether the weighted average filter mode is within a certain threshold. Scanning conversion method.

Images