JP2002152623A - Automatic contrast correction circuit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automatic contrast correction circuit that can be realized on a simple configuration. SOLUTION: In this automatic contrast correction circuit which converts luminance, equal to or higher than fixed luminance, on the higher luminance side of input image data into the highest luminance, also converts luminance, equal to or lower than fixed luminance, on the lower luminance side into the lowest luminance and linearly expands the luminance in the meantime from the lowest luminance to the highest luminance in accordance with the luminance, the luminance of the lower luminance side, the luminance of the higher luminance side, the number of neglected pixels on the lower luminance side and the number of neglected pixels are preliminarily set, the number of pixels that are equal to or lower than the set luminance of the lower luminance side and the number of pixels that are equal to or higher than the set luminance of the higher luminance side are counted among pieces of the input image data, the respective numbers of counted pixels are compared with the number of the neglected pixels, the luminance of the lower luminance side and the luminance of the higher luminance side are increased/reduced so as to make the respective numbers of the pixels to be the closest, and luminance at the time when an increase/reduction direction is inverted is respectively defined as the luminance of the lower luminance side and the luminance of the higher luminance side which are searched for.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to image processing,
In particular, it relates to a circuit for automatically correcting the contrast of an input image.

[0002]

2. Description of the Related Art When the range of luminance of an input image is smaller than the allowable display range of a display, the displayed image is difficult to see. In order to make a display image easy to see, it is known to perform contrast correction for expanding the luminance so that the luminance range of the input image matches the allowable display range of the display.

[0003] As one method of the contrast correction, a luminance lower than a certain luminance among input image data is set as a minimum luminance, a luminance above a certain luminance is set as a maximum luminance, and the luminance during this period is changed from the lowest luminance according to the luminance. There is known a method of linearly expanding luminance so as to have the highest luminance. However, when the contrast correction is performed by selecting the minimum luminance and the maximum luminance of the input image data as a constant luminance using this method, if even one pixel has the minimum luminance or the maximum luminance, it is dragged by the pixel and the desired contrast is obtained. There was a drawback that correction was not possible. In order to solve this drawback, it has been found that the luminance of the pixel having the lowest luminance or the highest luminance, which occupies a small number of pixels in one frame of the input image, may be ignored. (Distribution) method, a histogram for one frame of the input image is calculated, and the contrast correction is executed by setting the luminance that is equal to or less than a predetermined number of pixels as the minimum luminance and the maximum luminance.

FIG. 3 is a block diagram of a main part for executing contrast correction using this method. In FIG. 3, 4
Reference numeral 1 denotes a histogram creation unit that creates a histogram for each predetermined gradation from input image data. Reference numeral 42 denotes a ratio of pixels occupied by pixels having a specific brightness or higher in the input image data determined in advance from the histogram created by the histogram creation unit 41. A white level search unit for setting the specific luminance to a white level when the pixel ratio is closest to the predetermined ratio, the ratio of the number of pixels having a luminance equal to or lower than the specific luminance in the input image from the histogram created by the histogram creating unit 41; When it is closest to a predetermined ratio, the black level search unit 44 sets the specific luminance to a black level. The black level search unit 44 calculates the amplification factor from the white level from the white level search unit 42 and the black level from the black level search unit 43. A calculation unit for calculating an amplification factor; 45, a black level subtraction unit for subtracting the black level from the black level search unit 43 from the input image data; 6 is a multiplying unit for multiplying the amplification factor of the amplification factor calculating unit 44 to the data from the black level subtraction unit 45. The amplification factor is obtained by dividing a value obtained by subtracting the lowest luminance from the highest luminance by a value obtained by subtracting the black level from the white level.

Next, the operation of the contrast correction shown in FIG. 3 will be described. The ratios of the number of pixels to be ignored in the high luminance part and the low luminance part of the input image data are set in advance in the white level search unit 42 and the black level search unit 43, respectively. In this state, the histogram of the input image data is created by the histogram creating unit 41. The histogram is input to a white level search unit 42 and a black level search unit 43, and a white level and a black level that are closest to the set values are calculated. Subsequently, the gain is calculated by the gain calculator 44 as described above. The black level subtractor 45 subtracts the black level from the input image data. Thus, the input image data below the black level has the lowest luminance. Note that in this subtraction, the minimum luminance is zero and is not set to a negative number. The image data from which the black level has been subtracted is multiplied by the amplification factor in the multiplier 46, and the luminance is expanded so that the input image data between the black level and the white level has the lowest luminance to the highest luminance. In this multiplication, the maximum value is the highest luminance, and the maximum value is set not to exceed this value.
Thus, contrast correction is realized.

[0006]

As described above, the conventional contrast correction can be realized by creating a histogram, and the method is simple, but the number of pixels per frame increases, and When the number of bits per pixel increases, the circuit scale increases. SUMMARY OF THE INVENTION It is an object of the present invention to provide an automatic contrast correction circuit capable of easily calculating a negligible black level and white level and realizing it with a simple configuration.

[0007]

A first automatic contrast correction circuit according to the present invention is arranged so that the luminance higher than a certain luminance on the high luminance side of input image data is set to the maximum luminance, and the luminance lower than the predetermined luminance on the low luminance side. Is converted to the lowest brightness, and the brightness between the low brightness side, the brightness on the high brightness side, and the number of ignored pixels on the low brightness side in the automatic contrast correction circuit that linearly expands the brightness from the lowest brightness to the highest brightness according to the brightness. And the number of neglected pixels on the high-luminance side is set in advance, and the number of pixels below the set low-luminance-side luminance and the number of pixels above the set high-luminance-side luminance in the input image data are counted. Compare the number of pixels with the number of ignored pixels, increase or decrease the luminance on the low luminance side and the luminance on the high luminance side so that each pixel number is the closest, and obtain the luminance when the increasing / decreasing direction is reversed. And high brightness side It is characterized by brightness.

Further, the second automatic contrast correction circuit according to the present invention sets the luminance above a certain luminance on the high luminance side of the input image data to the maximum luminance, and the luminance below the constant luminance on the low luminance side to the minimum luminance. The automatic contrast correction circuit that converts and linearly expands the luminance between the minimum luminance and the maximum luminance in accordance with the luminance has the following configuration to obtain the constant luminance on the low luminance side and the constant luminance on the high luminance side. It is characterized by the following. (B) Predetermined constant luminance on the high luminance side, constant luminance on the low luminance side, the number of ignored pixels on the high luminance side, and the number of ignored pixels on the low luminance side are set. Count the number of pixels of input image data equal to or higher than a certain luminance on the high luminance side and equal to or less than a certain luminance on the low luminance side for one frame of the input image. (B) comparing the number of counted pixels on the high-brightness side with the number of pixels on the low-brightness side described in (b) according to the comparison result described in (c). (E) increasing or decreasing the luminance of each of the high-luminance side and the low-luminance side by a predetermined number. (E) The number of pixels of the input image data equal to or higher than the luminance on the high-luminance side and lower than the luminance on the low-luminance side according to (d). Count for one frame. (F) Ignored image on the high luminance side described in (a) Compare the number and the number of pixels to be ignored on the low-brightness side with the number of counted pixels on the high-brightness side and the number of pixels on the low-brightness side described in (e), respectively, according to the comparison result described in (g) and (f). (D) increasing or decreasing the luminance of each of the high luminance side and the low luminance side described in (d). (H) repeating the counting and comparison of the number of pixels until the direction of increase or decrease is reversed. Respective luminances for obtaining the luminance on the high luminance side and the luminance on the low luminance side [Detailed Description of the Invention]

According to the present invention, the luminance on the low luminance side, the luminance on the high luminance side, the number of neglected pixels on the low luminance side, and the number of neglected pixels on the high luminance side are preset, and the low luminance set in the input image data is set. Count the number of pixels below the luminance on the luminance side and the number of pixels above the luminance on the set high luminance side, compare the number of counted pixels with the number of neglected pixels, and compare the number of pixels with the number of ignored pixels on the low luminance And the luminance on the high luminance side are increased / decreased to obtain negligible luminance on the low luminance side and luminance on the high luminance side.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the drawings. FIG. 1 shows a black level detecting unit for calculating a black level having a negligible constant luminance according to an embodiment of the present invention. FIG. 2 shows a black level detector. Without a white level calculator that calculates a white level that is negligible luminance, it is not possible to calculate the neglected white level and black level, which is one of the conditions for contrast correction, but there is a difference on the block diagram. Since there is no such figure, it is omitted. Also, a black level subtraction unit required for contrast correction,
Since the amplification factor calculation unit and the multiplication unit are the same as those in the conventional example, the illustration and description are omitted.

[0011]

Embodiment 1 In FIG. 1, reference numeral 1 designates a predetermined black level as a comparison value. Further, in response to an increase / decrease signal 107 of the comparison value, the set comparison value is increased / decreased and the result is obtained. A comparison value register that sets the comparison value as a comparison value and outputs the set comparison value as it is as a comparison value 101. The comparison value register 2 compares the comparison value 101 with the input image data 102.
A first value comparison unit that outputs a comparison result signal 103 is a first value comparison unit.
When the input image data 102 is smaller in accordance with the comparison result signal 103, the number of pixels is counted, and the
4 is a pixel counter that outputs 4, and 4 sets the number of pixels to be ignored among the total number of pixels of one frame of the input image, and outputs an ignored value register that is directly output as the number of ignored pixels 105. And a second value comparison unit 6 that outputs a second comparison result signal 106
A comparison value increasing / decreasing unit that generates the increase / decrease signal 107 according to the comparison result signal 106.

Next, the operation of detecting the black level will be described with reference to FIG. First, a black level which is a predetermined constant luminance is set in the comparison value register 1 and the number of pixels to be ignored is set in the ignore value register 4. In this state, the input image data 102 is input to the first value comparison unit 2. In the first value comparison unit 2, the luminance of the input image data 102 is compared with the luminance of the comparison value 101. The first value comparison unit 2 outputs a logic “1” as the first comparison result signal 103 when the input image data 102 is smaller than the comparison value 101, and outputs a logic “0” otherwise.

The first comparison result signal 103 is the pixel counter 3
, And counts the number of pixels when the first comparison result signal 103 is logic “1”. This counting is performed for one frame of the input image. In this way, the number of pixels having a comparison level of 1 or less, which is the black level set to the comparison value 1 in the image of one frame, is counted. The counted pixel number 104 is sent to the second value comparing unit 5, where it is compared with the neglected pixel number 105. When the neglected pixel number 105 exceeds the counted pixel number 104 as the second comparison result signal 106, the logic “1” is output. , And outputs a logic “0” otherwise. The second comparison result signal 106 is sent to the comparison value increasing / decreasing unit 6, where the comparison value increase signal is used when the logic is "1", the comparison value decrease signal is used when the logic is "0", and the logic value is inverted when the logic is inverted. A signal that does not increase or decrease is output as an increase / decrease signal 107.

When the increase / decrease signal 107 is a comparison value increase signal, 1 is added to the comparison value set in the comparison value register 1, and this value is output as a new comparison value 101, and the above-described operation is executed. As a result, if the increase signal 107 is still a comparison value increase signal, 1 is added to the comparison value set in the comparison value register 1 and this value is added to the new comparison value 10.
1 is output, and the above-described operation is performed. This behavior is
The process is continued until the increase / decrease signal 107 changes to the comparison value decrease signal. When this change appears, the above-described operation is stopped, and the value of the comparison value 101 at this time is the desired black level.

Similarly, when the increase / decrease signal 107 is a comparison value decrease signal, 1 is subtracted from the comparison value set in the comparison value register 1, and this value is output as a new comparison value 101. Execute As a result, if the increase signal 107 is a comparison value decrease signal, 1 is further subtracted from the comparison value set in the comparison value register 1, and this value is output as a new comparison value 101, and the above-described operation is executed. . This operation is continued until the increase / decrease signal 107 changes to the comparison value increase signal. When this change appears, the above-described operation is stopped, and the value of the comparison value 101 at this time is the desired black level.

Next, the operation of detecting the white level will be described. Since the detection of the white level is essentially the same as the detection of the black level, it will be briefly described. First, a white level which is a predetermined constant luminance and the number of pixels to be ignored are set. In this state, the input image data is compared with the white level in the same manner as in the above-described black level detection, and the result is obtained.

As a result of the comparison, only when the input image data is larger, the number of pixels is counted for one frame of the input image. In this way, the number of pixels at or above the white level in the image of one frame is counted. This counted pixel number is compared with the negligible pixel number. If the measured pixel number is larger than the negligible pixel number, 1 is added to the white level, and the above operation is repeated as a new white level. This operation is repeated until the number of counted pixels becomes smaller than the number of ignored pixels, and the white level at this time is the desired white level.

Conversely, when the number of counted pixels is smaller than the number of ignored pixels, 1 is subtracted from the initially set white level, and the above operation is repeated as a new white level. This operation is repeated until the number of counted pixels becomes greater than the number of ignored pixels, and the white level at this time is the desired white level.

[0019]

Second Embodiment The configuration of the second embodiment is to speed up the convergence of the black level faster than the configuration of the first embodiment. In FIG. 2, reference numeral 21 designates a predetermined black level as a comparison value, and also receives a comparison value increase / decrease signal 211 to increase or decrease the set comparison value and set the result as a comparison value. The set comparison value is used as the comparison value 201
The first comparison value register 22 outputs the comparison value 20
The second comparison register 23 receives 1 and adds 1 to the comparison value 201 and outputs the result as the comparison value 203. The second comparison register 23 compares the comparison value 101 with the input image data 202 and outputs the first comparison result signal 2.
The first value comparison unit 24 that outputs the value 04 compares the comparison value 203 with the input image data 202 and outputs the second comparison result signal 20.
5 is a second comparing section that outputs the signal as 5.

Reference numeral 25 denotes a first pixel counter for counting the number of pixels when the input image data 202 is smaller in accordance with the first comparison result signal 204 and outputting a counted pixel number 206; When the input image data 202 is smaller in accordance with the comparison result signal 205, the second pixel counter that counts the number of pixels and outputs a count pixel number 207, 2
Reference numeral 7 denotes an ignored value register which sets the number of pixels to be ignored among the total number of pixels in one frame of the input image and outputs the number of ignored pixels as it is. Reference numeral 28 compares the number of counted pixels 206 with the number of ignored pixels 208. A third value comparing unit that outputs a comparison result signal 209, 29 compares the number of counted pixels 207 with the number of neglected pixels 208, and a fourth value comparing unit that outputs a fourth comparison result signal 210, and 30 denotes a third comparison result A comparison value increasing / decreasing unit that generates the increase / decrease signal 211 according to the signal 209 and the fourth comparison result signal 210.

Next, the operation of detecting the black level will be described with reference to FIG. First, a black level which is a predetermined constant luminance is set in the first comparison value register 21 and the number of pixels to be ignored is set in the ignore value register 27. In the second comparison value register 22, a black level obtained by adding one to the constant luminance set in the first comparison value register 21 is set. In this state, the input image data 202 is input to the first value comparison unit 23 and the second value comparison unit 24.

In the first value comparing section 23, the input image data 2
02 and the comparison value 201 are compared by the second comparison unit 24.
Then, the luminance of the input image data 202 is compared with the luminance of the comparison value 203. In the first value comparison unit 23, the input image data 202 has the first comparison result signal 204 as the comparison value 1.
When the value is equal to or less than 01, the logic value is “1”; otherwise, the logic value is “0”.
When the input image data 202 is equal to or smaller than the comparison value 203, the logic “1” is output. Otherwise, the logic “0” is output.

The first comparison result signal 204 is sent to the first pixel counter 25, and the first comparison result signal 204 becomes logic "1".
The number of pixels in the case of is counted. Also, the second comparison result signal 2
05 is sent to the second pixel counter 26 to count the number of pixels when the second comparison result signal 205 is logic “1”. This counting is performed for one frame of the input image. In this way, the number of pixels having a black level set to the comparison value 1 or less and the number of pixels having a black level set to the comparison value 2 or less in the image of one frame are counted.

The counted pixel numbers 206 and 207 are sent to the third value comparing section 28 and the fourth comparing section 29, respectively, where they are compared with the neglected pixel number 208 and the third comparison result signal 20 is output.
When the number of ignored pixels 208 exceeds the number of counted pixels 206, the logic “1” is set as 9, and the logic “0” is set otherwise.
When the ignored pixel number 208 exceeds the counted pixel number 207 as the fourth comparison result signal 210, the logic “1” is output. Otherwise, the logic “0” is output. Third comparison result signal 209
The fourth comparison result signal 210 is sent to the comparison value increasing / decreasing unit 30, where the following five determinations are made.

That is, whether the third comparison result signal 209 and the fourth comparison result signal 210 are both logic "0" or whether both the third comparison result signal 209 and the fourth comparison result signal 210 are logic "1". , The third comparison result signal 209 is logic “1” and the fourth comparison result signal 210 is logic “0”, or the third comparison result signal 209 is logic “0” and the fourth comparison result signal 210 is logic “1”. ”Or something else. By each determination, an increase / decrease signal 211 for increasing / decreasing the black level of the comparison value 1 is generated, and the set values of the first comparison value register 21 and the second comparison value register 22 are increased / decreased.

If the increase / decrease signal 211 is a signal indicating the above case, 4 is subtracted from the comparison value set in the comparison value register 21 and this value is output as a new comparison value 201, and the above-described operation is executed. . As a result, the increase / decrease signal 2
If 11 is a signal indicating the above case, 4 is further subtracted from the comparison value set in the comparison value register 21, and this value is output as a new comparison value 201, and the above-described operation is executed. This operation is continued until the increase / decrease signal 211 becomes the above-mentioned state. When this state change appears, the above operation is stopped, and the value of the comparison value 201 at this time is the desired black level.

If the increase / decrease signal 211 is a signal indicating the above case, 2 is added to the comparison value set in the comparison value register 21 and this value is output as a new comparison value 201, and the above-described operation is executed. . As a result, the increase / decrease signal 2
If 11 is a signal indicating the above case, 2 is added to the comparison value set in the comparison value register 21 and this value is output as a new comparison value 201, and the above-described operation is executed. This operation is continued until the increase / decrease signal 211 is in the above-mentioned state or in the above-mentioned state. When this state change appears, the above operation is stopped, and the value of the comparison value 201 at this time is the desired black level.

In the case where the increase / decrease signal 211 is as described above, the black level is determined at this point without increasing / decreasing the comparison value set in the comparison value register 21 and the comparison value set in the comparison value register 21 is This is the desired black level.

Next, the operation of white level detection will be described. Since the detection of the white level is essentially the same as the detection of the black level, it will be briefly described. First, a white level which is a predetermined constant luminance and the number of pixels to be ignored are set. In this state, the input image data is compared with the white level, and the input image data and the level obtained by adding 1 to the white level are compared with each other in the same manner as in the detection of the black level, and the result is obtained.

As a result of the comparison, only when the input image data is larger, the number of pixels is counted for one frame of the input image. In this way, the white level and the number of pixels equal to or higher than the level obtained by adding 1 to the white level in the image of one frame are counted. The two counted pixel numbers are compared with the neglected pixel number. If both of the two measured pixel numbers are larger than the neglected pixel number, 2 is added to the white level, and the above operation is repeated as a new white level. This operation is repeated until any one of the counted pixel numbers becomes smaller than the ignored pixel number, and the white level at the time when the repetition is stopped is the desired white level.

Conversely, when the two measured pixel numbers are both smaller than the ignored pixel number, the above operation is repeated by subtracting 4 from the white level to obtain a new white level. This operation is repeated until any one of the counted pixel numbers becomes larger than the ignored pixel number, and the white level at the time when the repetition is stopped is the desired white level.

[0032]

According to the present invention, as described above, the black level, the white level, the number of neglected pixels on the black side and the number of neglected pixels on the white side are set in advance, and the black level set in the input image data is set. Count the number of pixels below and the number of pixels above the set white level, compare the number of counted pixels with the number of ignored pixels,
The black level and white level are increased or decreased so that the number of pixels approaches each other, and the negligible black level and white level are obtained, so that an automatic contrast correction circuit can be realized with a simple configuration and cost reduction is achieved. Can be. In addition, since the configuration is simple, design errors are reduced and reliability can be improved.

[Brief description of the drawings]

FIG. 1 is a black level detection unit that calculates a black level that is equal to or less than a negligible constant luminance according to an embodiment of the present invention.

FIG. 2 shows a black level detection unit according to another embodiment of the present invention.

FIG. 3 is a block diagram of a main part of a conventional automatic contrast correction circuit.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Comparison value register 2 1st value comparison unit 3 Pixel counter 4 Neglect value register 5 2nd value comparison unit 6 Comparison value increase / decrease unit

Claims (2)

[Claims] 1. A method of converting luminance higher than a certain luminance on a high luminance side of input image data into a maximum luminance, and converting luminance below a certain luminance on a low luminance side into a minimum luminance, and linearly converting luminance between the luminances according to the luminance. In the automatic contrast correction circuit that expands from the lowest luminance to the highest luminance, the luminance on the low luminance side, the luminance on the high luminance side, the number of ignored pixels on the low luminance side, and the number of ignored pixels on the high luminance side are set in advance, and the input image data Count the number of pixels below the set low-luminance side and the number of pixels above the set high-luminance side, compare each counted pixel number with the negligible pixel number, and each pixel number becomes closest An automatic contrast correction circuit characterized in that the luminance on the low luminance side and the luminance on the high luminance side are increased / decreased, and the luminance on the low luminance side and the luminance on the high luminance side are respectively obtained when the increasing / decreasing direction is reversed. . 2. The method according to claim 1, wherein the luminance of the input image data, which is equal to or higher than a certain luminance on the high luminance side, is converted to the maximum luminance, and the luminance which is equal to or lower than the predetermined luminance on the low luminance side is converted to the minimum luminance. An automatic contrast correction circuit for expanding from the lowest luminance to the highest luminance, wherein the automatic contrast correction circuit has the following configuration for obtaining constant luminance on the low luminance side and constant luminance on the high luminance side. (B) Predetermined constant luminance on the high luminance side, constant luminance on the low luminance side, the number of ignored pixels on the high luminance side, and the number of ignored pixels on the low luminance side are set. Count the number of pixels of input image data equal to or higher than a certain luminance on the high luminance side and equal to or less than a certain luminance on the low luminance side for one frame of the input image. (B) comparing the number of counted pixels on the high-brightness side with the number of pixels on the low-brightness side described in (b) according to the comparison result described in (c). (E) increasing or decreasing the luminance of each of the high-luminance side and the low-luminance side by a predetermined number. (E) The number of pixels of the input image data equal to or higher than the luminance on the high-luminance side and equal to or lower than the luminance on the low-luminance side according to (d). Count for one frame. (F) Ignored image on the high luminance side described in (a) Compare the number and the number of pixels to be ignored on the low-brightness side with the number of counted pixels on the high-brightness side and the number of pixels on the low-brightness side described in (e), respectively, according to the comparison result described in (g) and (f). (D) increasing or decreasing the luminance of each of the high luminance side and the low luminance side described in (d). (H) repeating the counting and comparison of the number of pixels until the direction of increase or decrease is reversed. Calculate the luminance on the high luminance side and the luminance on the low luminance side respectively.