JP4061030B2 - Outline enhancement signal correction apparatus and method - Google Patents

Description

[0001]
【Technical field】
The present invention relates to an apparatus and method for correcting an edge enhancement signal.
[0002]
BACKGROUND OF THE INVENTION
When the image is blurred, the edge enhancement process is generally performed. Due to the contour enhancement, the blurred image becomes clear.
[0003]
However, if the outline is simply emphasized, the image of the part where the outline was originally clear is also uniformly emphasized, which may result in an unnatural image.
[0004]
DISCLOSURE OF THE INVENTION
It is an object of the present invention to obtain a natural image even if the contour is emphasized.
[0005]
The contour emphasis signal correcting apparatus according to the present invention includes an edge emphasis signal generating means for generating an edge emphasis signal for emphasizing the contour of an image represented by a given video signal, and an edge emphasis signal generated by the edge emphasis signal generation means. The contour emphasizing signal when the level of the contour emphasis signal is determined to be greater than or equal to a predetermined threshold by the determining means for determining whether the level is equal to or higher than a predetermined threshold. Signal suppression means for suppressing the level of the contour enhancement signal so that the suppression amount increases as the level of the signal increases, and the suppression amount decreases as the level of the contour enhancement signal decreases. .
[0006]
The present invention also provides a method suitable for the above apparatus. That is, this method generates a contour emphasizing signal that emphasizes the contour of an image represented by a given video signal, determines whether the level of the generated contour emphasizing signal is equal to or higher than a predetermined threshold, and In response to the determination that the level of the enhancement signal is greater than or equal to a predetermined threshold value, the suppression amount increases as the level of the contour enhancement signal decreases, so that the suppression amount increases as the level of the contour enhancement signal increases. The level of the contour emphasis signal is suppressed so that becomes smaller.
[0007]
According to the present invention, a contour emphasis signal that enhances the contour of an image is generated. It is determined whether the level of the generated edge enhancement signal is equal to or higher than a predetermined threshold value. If it is determined that the threshold value is equal to or greater than a predetermined threshold value, the level of the contour enhancement signal is suppressed so that the suppression amount increases as the level of the contour enhancement signal increases, and the suppression amount decreases as the level of the contour enhancement signal decreases. Is done. When the generated contour emphasis signal is equal to or greater than a predetermined threshold value, the suppressed contour emphasis signal is added to the video signal. When the generated contour emphasis signal is equal to or lower than a predetermined threshold, the generated contour emphasis signal is added to a given video signal without being suppressed.
[0008]
When the level of the generated contour emphasis signal is equal to or higher than a predetermined threshold value, the suppressed contour emphasis signal is added to the video signal, so that there is no possibility of enhancing the contour more than necessary. In particular, since the level of suppression is such that the suppression amount increases as the level of the contour enhancement signal increases, and the suppression amount decreases as the level of the contour enhancement signal decreases, the waveform of the contour enhancement signal after suppression is reduced. Is similar to the waveform of the contour enhancement signal before suppression. Since a contour enhancement signal after suppression similar to the waveform of the contour enhancement signal before suppression can be added to the video signal, natural contour enhancement can be realized.
[0009]
For example, when the level difference between a video signal level representing a pixel constituting the contour and a video signal level representing a pixel adjacent to the pixel constituting the contour is equal to or greater than a certain level, the determination means It is determined that the level is equal to or higher than a predetermined threshold value.
[0010]
[Explanation of Examples]
FIG. 1 shows the contour enhancement process.
[0011]
The contour enhancement process is performed on the luminance signal (luminance data).
[0012]
When a luminance signal of an image to be contour-enhanced is input (input luminance signal), a portion where the luminance signal level is changed is detected as an edge (contour portion). A contour emphasis signal is generated based on the detected luminance signal of the edge portion.
[0013]
The generated contour emphasis signal is added to the input luminance signal. An edge-enhanced luminance signal is generated and output (output luminance signal). Due to the contour enhancement, the blurred image becomes a sharp image.
[0014]
2A and 2B show how the edge emphasis signal is corrected according to the embodiment of the present invention.
[0015]
In the center of FIG. 2A, the contour enhancement signal (original contour enhancement signal) generated as described above is shown. When it is determined that the original contour enhancement signal is equal to or greater than a predetermined threshold, if the original contour enhancement signal is simply added to the input luminance signal, the contour enhancement signal (or the contour enhancement signal does not fit within the dynamic range of the device). (Luminance signal) may be clipped (see the right side of FIG. 2A). Due to this clip, the contour of the image represented by the luminance signal after contour enhancement becomes thick and unnatural. For this reason, in this embodiment, when the original contour emphasis signal is determined to be greater than or equal to a predetermined threshold value, the level of the original contour emphasis signal is suppressed while the signal waveform of the original contour emphasis signal is substantially maintained ( See the left side of FIG.
[0016]
Since the edge-enhanced luminance signal falls within the dynamic range, the luminance signal after edge enhancement is prevented from being clipped in advance. In addition, since the signal waveform of the luminance signal after the contour enhancement is suppressed to be similar to the signal waveform of the original contour enhancement signal, the image represented by the luminance signal after the contour enhancement does not have to be unnatural.
[0017]
Also, as shown in the center of FIG. 2B, when the original contour emphasizing signal includes signal portions above the threshold and signal portions below the threshold alternately, If the signal is cut, a discontinuous point is generated (see the right side of FIG. 2B). In this embodiment, as described above, the level of the original contour emphasizing signal is suppressed while substantially maintaining the signal waveform of the original contour emphasizing signal (see the left side of FIG. 2B). It can be prevented beforehand.
[0018]
FIG. 3 shows an embodiment of the present invention and is a block diagram showing an electrical configuration of the contour emphasizing apparatus. FIG. 4 is a flowchart showing the processing procedure of the contour enhancement processing.
[0019]
RGB image data is given, and the RGB / YC conversion circuit 1 generates luminance data Y and color difference data C. The generated luminance data Y is input to the adder circuit 12 and the contour enhancement data generation circuit 2. The generated color difference data is input to the hue correction circuit 12.
[0020]
When the luminance data is input to the contour emphasis data generation circuit 2, original contour emphasis data (original contour emphasis signal) is generated as described above (step 21). The generated original outline emphasis data is given to the line memory 3 and temporarily stored (step 22). The line memory 3 can temporarily store original outline emphasis data for three lines.
[0021]
The original outline emphasis data for three lines stored in the line memory 3 is read and applied to the difference circuit 4. The difference circuit 4 is a circuit that calculates a difference between the level of a specific pixel and the levels of eight pixels adjacent to the specific pixel. In the difference circuit 4, the difference between the specific pixel and the eight pixels adjacent to the specific pixel is calculated (step 23).
[0022]
FIG. 5 shows an image portion of 3 × 3 pixels represented by the contour emphasis data stored in the line memory 3.
[0023]
The center pixel is a specific pixel (Eg5) of interest. Eight neighboring pixels Eg1 to Eg4 and Eg6 to Eg9 are arranged around the central pixel. In the difference circuit 4, the difference between the specific pixel Eg5 and the neighboring pixels Eg1 to Eg4 and Eg6 to Eg9 is calculated.
[0024]
The eight types of difference data calculated in the difference circuit 4 are given to the absolute value conversion circuit 5 and converted into absolute values (step 24). The eight kinds of absolute difference data are averaged in the average value calculation circuit 6 (step 25). Assuming that the data (levels) from the pixels Eg1 to Eg9 is Eg1 to Eg9, the averaging circuit 6 outputs the average data EDAV according to Equation 1.
[0025]
EDAV = {(| Eg5-Eg1 |) + (| Eg5-Eg2 |) + (| Eg5-Eg3 |) + (| Eg5-Eg4 |) + (| Eg5-Eg6 |) + (| Eg5-Eg7 |) + (| Eg5-Eg8 |) + (| Eg5-Eg9 |)} / 8 Equation 1
[0026]
The averaged data EDAV output from the averaging circuit 6 is given to the division circuit 7.
[0027]
The contour emphasis data Eg0 is read from the line memory 3 and supplied to the first multiplication circuit 8 as well. The first multiplication circuit 8 is also given data indicating a predetermined threshold value EDTH. In the first multiplication circuit 8, the edge emphasis data Eg0 is multiplied by a threshold value EDTH (step 26). Output data from the first multiplication circuit 8 is given to the division circuit 7.
[0028]
Further, data representing the threshold value EDTH and averaged data representing the average value EDAV are input to the comparison circuit 9. In the comparison circuit 9, the threshold value EDTH and the average value EDAV are compared (step 28), and a signal indicating the comparison result is given to the selection circuit 10 described later.
[0029]
In the division circuit 7, the average value EDAV is divided by Eg0 × EDTH output from the first multiplication circuit (step 27) (Egp1 = (Eg0 × EDTH) / EDAV). By this division processing, the contour enhancement data is suppressed (contour enhancement data Egp1). The suppressed edge emphasis data Egp1 is input to the selection circuit 10. The original outline emphasis data Eg0 output from the line memory 3 is also input to the selection circuit 10.
[0030]
If the comparison circuit 9 determines that the average value EDAV is smaller than the threshold value EDTH, it can be considered that the original contour emphasis data Eg0 itself is within the dynamic range even if it is added to the luminance data. The original outline emphasis data Eg0 is output from the selection circuit 10. If the comparison circuit 9 determines that the average value EDAV is larger than the threshold value EDTH, the original outline emphasis data Eg0 itself (that is, Eg5) may be added to the luminance data and may not be within the dynamic range. . The suppressed original contour emphasis data Egp1 is output from the selection circuit 10 (step 29).
[0031]
The contour emphasis data (Eg0 or Egp1) output from the selection circuit 10 is input to the second multiplication circuit 11. Predetermined gain up data is given to the second multiplication circuit 11. In the second multiplication circuit 11, the edge enhancement data is amplified (step 30).
[0032]
The contour emphasis data output from the second multiplication circuit 11 is input to the addition circuit 12 and added to the luminance data Y output from the RGB / YC conversion circuit 1 (step 31). The luminance data output from the adder circuit 12 is input to the YC / RGB conversion circuit 13. The color difference data whose hue has been corrected by the hue correction circuit 14 is also input to the YC / RGB conversion circuit 13.
[0033]
The YC / RGB conversion circuit 13 generates and outputs RGB image data from the luminance data Y and the color difference data C.
[0034]
In the embodiment described above, each circuit is constituted by hardware, but it goes without saying that it may be constituted by software.
[0035]
In the above, the suppression amount of the contour enhancement signal may be decreased as the correlation between the own pixel and the neighboring pixels adjacent to the own pixel is higher, and the suppression amount of the contour enhancement signal may be increased as the correlation is lower.
[Brief description of the drawings]
FIG. 1 shows contour enhancement processing.
FIGS. 2A and 2B show a corrected contour emphasizing signal, an uncorrected original contour emphasizing signal, and a contour emphasizing signal corrected by a conventional process according to this embodiment.
FIG. 3 is a block diagram showing an electrical configuration of a contour enhancement signal correction circuit.
FIG. 4 is a flowchart showing contour enhancement processing.
FIG. 5 shows pixels that form part of an image.
[Explanation of symbols]
2 Outline enhancement data generation circuit 3 Line memory 4 Difference circuit 5 Absolute value conversion circuit 6 Average value calculation circuit 7 Division circuit 8, 11 Multiplication circuit 9 Comparison circuit
10 Selection circuit
12 Adder circuit

Claims (3)

Contour enhancement signal generating means for generating a contour enhancement signal for enhancing the contour of an image represented by a given video signal;
Determining means for determining whether the level of the contour emphasis signal generated by the contour emphasis signal generating means is greater than or equal to a predetermined threshold; and the determination means determines that the level of the contour emphasis signal is greater than or equal to a predetermined threshold. When it is determined that there is an image, the average value of the level difference between the video signal level representing the pixels constituting the contour and the video signal level representing pixels adjacent to the pixels constituting the contour is used. Signal suppression means for dividing the contour enhancement signal to suppress the level of the contour enhancement signal;
A contour emphasizing signal correction device. The contour for emphasizing the contour when the level difference between the video signal level representing a pixel constituting the contour and a video signal level representing a pixel adjacent to the pixel constituting the contour is greater than or equal to a certain level. It is determined that the level of the enhancement signal is equal to or higher than a predetermined threshold value.
The outline emphasis signal correction device according to claim 1 . Generating a contour enhancement signal that enhances the contour of the image represented by the given video signal;
Determine whether the level of the generated edge enhancement signal is above a predetermined threshold,
In response to determining that the level of the contour enhancement signal is equal to or higher than a predetermined threshold value, the video signal level representing the pixels constituting the contour and the pixels adjacent to the pixels constituting the contour are represented. Dividing the contour emphasizing signal using an average value of the respective level differences from the video signal level to suppress the level of the contour emphasizing signal;
Outline enhancement signal correction method.

Images