JP2008271259A - Image processor, image processing method, image processing program and recording medium which records the program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor which reduces amounts of required resources and prevents that an edge area can not be detected and algorithm fails. <P>SOLUTION: The image processor 1 concerning the present invention is provided with a differential code detection part 6 which detects a section in which there are three pixels or more from a pixel immediately before a differential code of luminance values between the adjacent pixels changes to a pixel immediately before the differential code changes next in image data of an original image as the edge area, and a whole differential value detection part 7 which sets the edge area as an area for correction by a correction means when a whole differential value which is a differential value between the maximum value and the minimum value of the luminance values of pixels constituting the edge area is more than a predetermined threshold. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image processing apparatus, an image processing method, an image processing program, and a recording medium storing the program using a scaling process for enlarging or reducing an image.

  In recent years, high-resolution liquid crystal monitors and liquid crystal televisions that can display high-definition and high-definition images such as HD (high definition) images have rapidly spread, and display devices have been increased in resolution. On the other hand, there are still many low resolution contents such as SD (standard definition) video, and low resolution contents are often displayed on a high resolution device.

  However, in order to display low-resolution content by a high-resolution device, it is necessary to enlarge and display the low-resolution content, and the sharpness of the image, particularly the sharpness of the contour portion of the enlarged image, is significantly reduced. Therefore, when low-resolution content is displayed on a high-resolution device, contour enhancement processing is performed in order to improve the reduction in sharpness of the enlarged image.

  Non-Patent Document 1 discloses a contour enhancement processing technique using a high-pass filter. Specifically, the technique described above is a technique for generating a signal obtained by increasing a contour portion signal by generating a signal obtained by extracting a contour portion using a high-pass filter and adding the signal to the original signal. As a result, the density change in the contour portion can be emphasized, so that the sharpness of the processed image appears to be improved as compared with the original image.

  However, in the edge enhancement processing technique described in Patent Document 1, as shown in FIG. 12, undershoot and overshoot that are not included in the original image are formed, and there is a portion where the image after the edge enhancement process looks unnatural. Arise. FIG. 12 is a diagram showing a luminance graph in image data before correction and a luminance graph in image data after correction in the conventional contour correction processing.

  In addition, since the entire image is processed uniformly, when it is processed according to the originally steep outline part such as subtitles or telop part, the processing degree of the background part becomes minute, the sharpness is hardly improved, and the landscape In the case of emphasizing a sluggish contour portion, the originally steep contour portion is further sharpened, resulting in a problem of a glaring image. In addition to this, there is also a problem that the image quality is worsened by emphasizing the noise part.

  Non-Patent Documents 2 and 3 include an image processing technique for correcting an image by detecting a contour portion and selecting a processing region, instead of uniformly performing a contour enhancement process in order to solve the above problem. It is disclosed. The above technique will be specifically described below. First, for each contour width (contour) to be detected, a contour is extracted by comparing an easy gradation transition pattern and an input pixel in advance. Then, the detected contour is corrected by scaling the periphery of the correction target pixel in the extracted contour with a predetermined magnification control pattern.

  As a result, it is possible to realize a process in which the caption correction process is not performed on a noise portion such as a minute amplitude while the subtitle or the telop portion is sharpened by the contour correction (hereinafter also simply referred to as correction) processing.

In addition, since processing is performed using a scaling technique instead of using a processing method for adding undershoot and overshoot as in the technique described in Non-Patent Document 1, the problem of causing a shoot in the processed image is solved. it can.
Supervised by Nobuyuki Yagi “Digital Video Processing” Ohmsha Video Information Media Society 2000 Akihiro Nagase et al. “Contour Correction Method Applying Scaling Technology” 2005 Annual Conference of the Institute of Image Information and Television Engineers Akihiro Nagase et al., "Contour Detection Method for Contour Correction Applying Scaling Technology" 2005 Annual Conference of the Institute of Image Information and Television Engineers

  However, in the techniques described in Non-Patent Documents 2 and 3, for detecting the contour width, it is necessary to compare the gradation transition pattern prepared in advance with the input pixel, so the gradation transition according to the contour width is required. It is necessary to prepare many patterns. Therefore, when hardware is used for the above technique, or when the image processing apparatus is controlled by software, a very large amount of resources are required. In addition, if the contour width does not fit in the gradation transition pattern prepared in advance, the algorithm may fail.

  Furthermore, in the above-described technique, a region for detecting a low frequency component or a high frequency component with respect to a frequency component is provided to determine a region where an erroneous detection of a contour portion occurs. Thus, when it is determined that the detection is erroneous, the above technique adjusts the scaling factor so as to suppress the contour correction. However, in the above technique, no consideration is given to the control of the correction magnification near the threshold. Therefore, in the threshold value control in the above technique, in the contour portion near the threshold value, a corrected portion and a non-corrected portion are generated even though the contour portion is the same, and the image quality is deteriorated in the corrected image. There is a fear.

  The present invention has been made in view of the above problems, and its main purpose is to detect an edge region in a correction target image (original image) without preparing a gradation transition pattern corresponding to the contour width in advance, and is necessary. An image processing apparatus that reduces the amount of resources and prevents an algorithm from failing because an edge region cannot be detected.

  In order to solve the above-described problem, an image processing apparatus according to the present invention is an image processing apparatus including a correction unit that corrects image data of an original image by a scaling process using a predetermined correction ratio. An edge area detecting means for detecting, as an edge area, a section having three or more pixels from the pixel immediately before the sign of the difference in luminance value between adjacent pixels in the data to the pixel immediately before the next change; The edge region is set as a correction target region by the correction unit when the overall difference value, which is the difference value between the maximum value and the minimum value of the luminance values of the pixels constituting the edge region, is equal to or greater than a predetermined threshold value. And a correction target area setting means.

  According to the above configuration, the image processing apparatus detects, as an edge region, a section in which there are three or more pixels until the next change from the pixel immediately before the sign of the difference in luminance value between adjacent pixels changes. Then, a difference value between the maximum value and the minimum value of the luminance values of the pixels constituting the detected edge region, that is, an overall difference value is calculated. If the overall difference value is equal to or greater than the threshold value, the image processing apparatus determines that the detected edge region is a correction target.

  As a result, the image processing apparatus can detect an edge region in the image data of the original image without preparing a transition pattern having a gradation corresponding to the contour width in advance. Therefore, it is possible to reduce the amount of resources required in the image processing apparatus and to prevent the algorithm from failing because the edge region cannot be detected.

  The image processing apparatus according to the present invention preferably further includes a correction ratio setting means for setting a value based on the overall difference value as a correction ratio for the correction target area.

  According to the above configuration, the calculated overall difference value can be used not only as a parameter for determining whether or not it is a correction target region, but also as a parameter for setting a correction ratio. That is, the correction ratio can be set as a value corresponding to the calculated overall difference value. As a result, the correction ratio can be set so as not to cause deterioration of the image quality even in the correction target region near the threshold value.

  Further, the image processing apparatus according to the present invention further includes partial difference maximum value calculation means for calculating a maximum value among partial difference values that are difference values of luminance values between adjacent pixels in the image data. The correction target area setting means sets the edge area as the correction target area when the maximum value of the partial difference values is equal to or greater than a predetermined threshold, and the correction ratio setting means When the maximum difference value is equal to or greater than a predetermined threshold value, it is preferable to set a value based on the maximum value of the partial difference value as the correction ratio with respect to the correction target region.

  According to the above configuration, there is an effect that an edge region that does not have a steep section in the edge region and does not need to be corrected can be removed from the edge region to be corrected.

  Further, by detecting the steepest portion in the detected edge region and correcting the detected portion as the center, the detected edge region can be corrected to give a sharper image.

  The image processing apparatus according to the present invention further includes a partial difference maximum value calculating unit that calculates a maximum value among partial difference values that are difference values of luminance values between adjacent pixels in the image data; Position detection means for detecting the position of the pixel constituting the correction target area in the correction target area based on the maximum difference value, and the correction ratio setting means When the maximum value is equal to or greater than a predetermined threshold, it is preferable to set the value based on the position as the correction ratio used for correcting the pixels constituting the correction target region.

  According to said structure, since a different correction | amendment ratio can be set for every pixel, there exists an effect which can correct | amend the detected edge area | region as a still smoother edge.

  The image processing apparatus according to the present invention further includes a partial difference maximum value calculating unit that calculates a maximum value among partial difference values that are difference values of luminance values between adjacent pixels in the image data; Gradation detection value calculating means for calculating, for each partial difference value, a gradation detection value that is a difference value between the maximum difference value and the partial difference value, and the correction ratio setting means includes: When the maximum value of the partial difference value is equal to or greater than a predetermined threshold value, the gradation detection value corresponding to the partial difference value is set for each pixel that is the calculation target of the partial difference value in the correction target region. It is preferable to set the value based on the correction ratio used for correcting the pixel.

  According to the above configuration, since a correction ratio in consideration of gradation can be set in a portion where gradation is formed, there is an effect that deterioration in image quality due to collapse of gradation can be suppressed.

  Further, according to the above configuration, the correction ratio can be set by a combination of values based on the overall difference value, the maximum value of the partial difference value, the position in the edge region, and the gradation detection value. As a result, the correction ratio can be set so as to be most suitable for correcting the correction target area.

  In the image processing apparatus according to the present invention, it is preferable that the correction unit corrects pixels other than the pixels at both ends in the correction target region.

  According to said structure, it can prevent that the pixel of the edge overlap part which may be contained also in an adjacent edge area | region is correct | amended based on any one edge area | region. As a result, there is an effect that it is possible to prevent deterioration of image quality in the corrected image.

  An image processing method according to the present invention is an image processing method including a correction step of correcting image data of an original image by scaling processing using a predetermined correction ratio, and a luminance value between adjacent pixels in the image data. An edge area detecting step for detecting, as an edge area, a section in which there are three or more pixels from the pixel immediately before the difference sign changes to the pixel immediately before the next change, and the luminance value of the pixels constituting the edge area A correction target region setting step of setting the edge region as a correction target region in the correction step when an overall difference value, which is a difference value between the maximum value and the minimum value of the first and second values, is equal to or greater than a predetermined threshold value. It is characterized by that.

  According to said structure, there exists an effect similar to the image processing apparatus which concerns on this invention.

  The image processing apparatus may be realized by a computer. In this case, an image processing program for realizing the image processing apparatus in the computer by operating the computer as each of the above means and a computer-readable recording medium on which the image processing program is recorded also fall within the scope of the present invention.

  As described above, the image processing apparatus according to the present invention corrects the edge region based on the edge region detection unit that detects the edge region from the image data in the original image and the overall difference value calculated from the detected edge region. And a correction target edge region detecting means for determining whether or not the target is set and setting a correction ratio. Thus, it is possible to reduce the amount of necessary resources in the image processing apparatus and to prevent the algorithm from failing because the edge region cannot be detected.

Embodiment 1
An image processing apparatus according to an embodiment of the present invention will be described below with reference to FIGS.

(Configuration of the image processing apparatus 1)
First, an image processing apparatus 1 according to the present invention will be described below with reference to FIG. FIG. 1 is a block diagram showing a main configuration of an image processing apparatus 1 according to an embodiment of the present invention. As shown in FIG. 1, the image processing apparatus 1 includes an edge detection unit 2, a scaling processing unit 3 (correction means), a memory controller 4, a line memory 5, a correction rate calculation unit 9, a pixel position detection unit 10, and a correction direction setting. A section 11 and a gradation detection section 12 are provided.

(Configuration of edge detection unit 2)
The edge detection unit 2 is for detecting an edge region from the image signal of the correction target image input to the image processing apparatus 1. As shown in FIG. 1, the edge detection unit 2 includes a differential code detection unit 6 (edge region detection unit), an overall difference value detection unit 7 (correction target region detection unit), and a partial difference maximum value detection unit 8. . Details of each member will be described later.

(Configuration of scaling processing unit 3)
The scaling processing unit 3 (correction unit) calculates and calculates the correction factor calculation unit 9 (correction ratio setting unit), the pixel position detection unit 10 and the correction direction setting unit 11 described later on the edge region detected by the edge detection unit 2. The image to be corrected is subjected to scaling processing based on the set correction information. As shown in FIG. 1, the scaling processing unit 3 includes a target pixel correction unit 13 and a filter calculation unit 14. Details of each member will be described later.

(Memory controller 4 and line memory 5)
The memory controller 4 receives the position information of the pixels in the edge region in the correction target image to be corrected by the scaling processing unit 3 from the scaling processing unit 3 and transmits it to the line memory 5. The line memory 5 is a memory that records an image signal of the correction target image, and supplies pixel information necessary for scaling to the scaling processing unit 3 based on information transmitted from the memory controller 4.

(Correction rate calculation unit 9, pixel position detection unit 10, correction direction setting unit 11 and gradation detection unit 12)
The correction rate calculation unit 9 is for calculating a correction rate (correction rate) based on the correction rate of each pixel in the edge region detected by the edge detection unit 2. The pixel position detection unit 10 detects a pixel position in the edge portion detected by the edge detection unit 2. The correction direction setting unit 11 sets the correction direction of the pixel whose position has been clarified by the pixel position detection unit 10. The gradation detection unit 12 determines whether or not gradation is formed in the edge region detected by the edge detection unit 2. Details of each member will be described later.

  Here, the “correction rate” in the present embodiment means a correction rate used during the scaling process. In the present embodiment, the “correction ratio” means a parameter for calculating the correction ratio.

(Outline of image processing process)
Next, an outline of an image processing process in the image processing apparatus 1 will be described below with reference to FIG. FIG. 2 is a flowchart showing an outline of an image processing process in the image processing apparatus 1.

(Image signal input)
First, the image processing apparatus 1 acquires an image signal (image data) of a correction target image (original image) (step S50). The correction target image is not particularly limited, and may be, for example, a still image input from a computer or the like, or a moving image input from a television or video.

(Detection of edge area)
The edge detection unit 2 detects an edge region from the input image signal (step S51). Specifically, first, the differential code detection unit 6 of the edge detection unit 2 detects the luminance value in the image signal acquired in step S50, and based on the detected luminance value, the luminance value of the adjacent pixel is detected. The code when the difference is taken is calculated as the difference code. Then, an interval in which there are three or more pixels from the pixel immediately before the pixel whose sign has changed (changed from + to − or from − to +) to the pixel immediately before the pixel whose sign changes next is “ Edge region "is detected. Also, the previous pixel where the difference code changes is set as an edge overlapping portion and is not corrected. The reason why the pixel set as the edge overlapping portion is not corrected will be described below with a specific example.

(Detection of edge area to be corrected)
Next, the edge detection unit 2 determines whether or not the detected edge region is to be corrected (steps S52 and S53). Specifically, the overall difference value detection unit 8 of the edge detection unit 2 calculates an overall difference value that is a difference value between the maximum value and the minimum value of the luminance values in the edge region (step S52). FIG. 3 schematically shows an example of the entire difference value in the edge region.

  In step S53, it is determined whether the calculated overall difference value is greater than a threshold value. If the overall difference value is smaller than the threshold value (No in step S53), it is determined that the detected edge region is a noise component such as a minute amplitude, and no contour correction is performed in order to avoid contour enhancement such as a noise component. (Step S55). If the overall difference value is larger than the threshold value (Yes in step S53), it is determined that the detected edge region is a correction target. That is, the entire difference value detection unit 7 detects “a correction target region” that is an edge region that is a correction target by the scaling processing unit 3 from the edge region detected in the differential code detection 6.

Note that the threshold value in step S53 may be set as appropriate by an operator who operates the image processing, or may be a numerical value set in advance. Further, it may be calculated for each edge region each time.
(Detection of edge area that needs correction)
If the overall difference value detection unit 7 determines that the detected edge region is a correction target region (Yes in step S53), the edge detection unit 2 determines whether to correct the detected correction target region (step S53). S54 and step S56). Specifically, the partial difference maximum value detection unit 8 of the edge detection unit 2 calculates the partial difference maximum value (step S54). The partial difference maximum value is set by calculating a partial difference value that is a difference in luminance value between adjacent pixels and detecting the maximum value of the calculated partial difference values. FIG. 4 schematically shows an example of the partial difference value and the partial difference maximum value in the edge region.

  In step S56, it is determined whether or not the calculated partial difference maximum value is larger than a threshold value. That is, the partial difference maximum value detection unit 8 determines that the correction target region is an edge region that needs to be corrected by the scaling processing unit 3 when the detected correction target region has a steep section in the region. To do. That is, the partial difference maximum value detection unit 8 narrows down the edge regions that need to be corrected among the correction target regions detected by the overall difference value detection unit 7.

  When the partial difference maximum value is smaller than the threshold (No in step S56), it is determined that it is not necessary to correct the detected correction target region, and no contour correction is performed. (Step S55). When the partial difference maximum value is larger than the threshold value (Yes in step S56), it is determined that the detected correction target area is an edge area that needs to be corrected.

  Note that the threshold value in step S56 may be set as appropriate by an operator who operates the image processing, or may be a numerical value set in advance. Further, it may be calculated for each edge region each time.

(Setting correction direction and calculating correction factor j)
Next, the correction factor calculation unit 9, the pixel position detection unit 10, and the correction direction setting unit 11 set correction information for correcting an area determined to be corrected (step S57). The correction information is a correction rate, a correction direction, and coordinates of a correction target pixel.

  Specifically, the correction rate calculation unit 9 calculates a correction rate j for correcting the edge region, and the pixel position detection unit 10 and the correction direction setting unit 11 determine the correction direction of each pixel constituting the edge region. Set. A more detailed method for calculating the correction rate j and the correction direction setting method will be described later.

(Execution of contour correction)
Finally, the scaling processing unit 3 executes edge region scaling processing based on the correction information calculated and set in step S57 (step S58).

  Specifically, the target pixel correction unit 13 in the scaling processing unit 3 acquires the correction information calculated and set in step S57 together with the position of the correction target pixel detected by the pixel position detection unit 10. When the scaling processing unit 3 acquires the correction rate j, the correction direction, and the position of the correction target pixel to the target pixel correction unit 13, the scaling processing unit 3 sends the pixel position to be read out of the correction target image to the memory controller 4, The memory controller 4 reads pixel information necessary for scaling from the line memory 5. The filter calculation unit 14 performs a filter calculation on the pixel information read from the line memory 5 based on the correction factor j and the correction direction parameter of the target pixel correction unit 13.

  Note that when enlarging an image, it is most preferable to perform the same processing as that at the same magnification after conversion to the resolution of the output destination device. However, by correcting the interpolation pixel with respect to the target pixel based on the positional relationship in the edge region of the target pixel to be corrected before the image is enlarged, the image can be enlarged and corrected at the same time.

  Through the above processing, the image processing apparatus 1 outputs the corrected image signal. In addition, the conventionally well-known method can be used for the scaling process in the scaling process part 4. FIG.

(Advantages of this embodiment)
As described above, the image processing apparatus 1 according to the present invention can detect the edge region in the correction target image without comparing the gradation transition pattern prepared in advance with the pixel of the image signal in the correction target image.

  This eliminates the need to prepare a gradation transition pattern in advance, thereby reducing the amount of necessary resources and calculating the correction ratio based on the luminance value of the pixels constituting the detected edge region. This does not apply to the key transition pattern, and the algorithm can be prevented from failing.

  Further, in the image processing apparatus 1 according to the present invention, in order to detect the edge region, the difference value between the maximum value and the minimum value of the luminance value of the pixel in the edge region detected by the difference code detection unit 6, that is, the total difference maximum value And the calculated overall difference maximum value is compared with the threshold (steps S52 and 53).

  Thereby, a minute amplitude such as a noise component can be removed from the edge region detected by the differential code detector 6. Accordingly, it is possible to prevent the corrected image from being deteriorated by enhancing the outline of a noise component or the like.

  Further, in the image processing apparatus 1 according to the present invention, the maximum value of the partial difference values that are differences in luminance values between adjacent pixels, that is, the partial difference maximum value is compared with a threshold value (steps S54 and S56).

  As a result, the edge region that does not need to be corrected can be removed from the correction target region. Further, by detecting a steep portion in the edge region and correcting the detected portion as the center, the detected edge region can be corrected so as to have a sharper image.

(Specific example of detection of edge region to be corrected)
Next, a specific example of detection of an edge region to be corrected in the image processing apparatus 1 described above will be described below with reference to FIG. FIG. 5 is a diagram illustrating a luminance graph 90 in the input image signal input to the image processing device 1, data calculated from the image signal, and a luminance graph 100 in the output image signal output from the image processing device 1. .

  In the input image signal in FIG. 5, for example, from the pixel 90 a immediately before the pixel 90 b whose sign has changed (changed from + to − or from − to +), the pixel 90 l immediately before the pixel 90 l whose sign changes next. Up to the pixel 90h is detected as the edge region 91. In FIG. 5, the symbol is indicated as “+” when the luminance value increases from the previous pixel, and as “−” when the luminance value decreases.

  The overall difference value in the detected edge region 91 is “227”, which is the difference value between the pixel 90a (luminance value: 270) and the pixel 90h (luminance value: 3). That is, when the threshold value is larger than “227”, the overall difference value detection unit 7 determines that the detected edge region 91 is a correction target region if it is a correction target.

  The partial difference values in the edge region 91 are as shown in FIG. 5, and “68”, which is the partial difference value between the pixel 90c and the pixel 90d, is the partial difference maximum value. That is, when the threshold value is larger than “68”, the partial difference maximum value detection unit 8 determines that the edge region 91 is an edge region that needs to be corrected.

  The pixels 90 a and 90 h are set as edge overlapping portions when the edge region 91 is detected by the differential code detection unit 6 and are not corrected. This is because the pixel 90a is also included in the edge region 92 formed by the pixels from the pixel 90i to the pixel 90a. When the pixel 90a is corrected, the correction is biased to either the edge region 91 or the edge region 92. This is because the image quality may be deteriorated as compared with that before correction. The pixel 90h is also a pixel included in the edge region 93 formed by the pixels from the pixel 90h to the pixel 90m, and is the same as the case of the pixel 90a. Therefore, in order to secure at least one point that can be corrected, the difference code detection unit 6 has a range from the pixel immediately before the pixel whose code has changed to the pixel immediately before the pixel whose code changes next. A section that is greater than or equal to a pixel is detected as an edge region.

(Correction direction setting method and correction factor j calculation method)
Next, a more detailed calculation method of the correction rate j in step S57 and a more detailed setting method of the correction direction will be described below with reference to FIG. FIG. 6 is a flowchart showing a more detailed calculation method of the correction rate j and a more detailed setting method of the correction direction in the present embodiment.

  First, the total difference value claim bottom 7 and the partial difference maximum value 8 are set with a correction ratio f and a correction ratio g as parameters for calculating the correction ratio j (steps S57a and S57b). A detailed method for setting the correction ratios f and g will be described later.

  Next, the pixel position detection unit 10 detects the positions of the pixels constituting the edge region (step S57c). Then, a correction ratio h is set based on the detected pixel position information (step S57d). The correction ratio h is set for each pixel constituting the correction target area. A detailed method for detecting the pixel position and setting the correction ratio h will be described later.

  Note that the position of the pixels constituting the edge region detected by the pixel position detection unit 10, that is, the pixel coordinates, is not limited to the setting of the correction ratio h, but is set in the correction direction and the line memory during the scaling process described below. 5 is also used for reading an image signal from 5.

  Thereafter, the gradation detecting unit 12 determines whether or not each section of the edge region to be corrected forms a gradation (step S57e). When gradation is formed in the edge region, as shown in FIG. 7, the difference value between the partial difference maximum value and the partial difference value is very small. In such a case, when the section having the maximum partial difference is preferentially corrected, as shown in FIG. 8, the formed gradation is destroyed, and the image quality is deteriorated in the corrected image. FIG. 7 schematically shows an example of gradation in the edge region. FIG. 8 shows a luminance graph when the image is corrected while ignoring the formation of gradations. The solid line is the luminance graph before correction, and the broken line is the luminance graph after correction.

  Therefore, the gradation detection unit 12 calculates, for each partial difference value, a gradation detection value that is a difference value between the partial difference maximum value and the partial difference value in order to prevent the gradation from being collapsed. Based on the calculated gradation detection value, A correction ratio i is set (step S57f). Since the gradation detection value is calculated for each partial difference value, the correction ratio i is set for each pixel constituting the correction target area. A detailed method for setting the correction ratio i will be described later.

  When the correction ratios f to i are set, the correction ratio calculation unit 9 calculates the correction ratio j by combining the set correction ratios f to i (step S57g). The calculation method of the correction rate j is not particularly limited as long as it is calculated based on the correction ratios f to i. As a method of calculating the correction rate j, for example, a method of calculating a product of correction ratios f to i (that is, f × g × h × i) can be given as an example.

  When the correction factor j is calculated by the correction factor calculator 9, the correction direction setting unit 11 sets the correction direction of the pixels constituting the edge region based on the pixel position detected by the pixel position detector 10 ( Step S57h). Specifically, a directionality flag is set for each pixel so that the correction direction changes across the partial difference maximum value calculated by the partial difference maximum value detection unit 8. The direction flag is set so that the corrected edge region is steep. That is, the correction direction setting unit 11 sets the directionality flag so as to approach the pixels of the edge overlapping portion located in the vicinity of each pixel. The direction flag is set for each pixel to be corrected.

  In this embodiment, the correction ratios are set in the order of the correction ratio f, the correction ratio g, the correction ratio h, and the correction ratio i. However, the setting order of the correction ratios f to i is limited to this. It is not a thing. For example, after setting the correction ratios f and g, the correction ratio i and the correction ratio h may be set in this order, or the gradation and the pixel position are detected before the correction ratios f and g are set, and the correction ratio h And i may be set.

  In this embodiment, the correction direction is set after calculating the correction rate j. However, the present invention is not limited to this, and the correction rate j may be calculated after setting the correction direction.

(Detailed setting method of correction ratio f)
The method for setting the correction ratio f is not particularly limited. For example, the threshold value (th_ads) set in step S53 is set as the correction rate 0%, and the overall difference value (th_ade) is set as the correction rate is set to 100%, and the correction rate is set according to the calculated overall difference maximum value. The method of setting f can be mentioned. For example, as shown in FIG. 9, a proportional relationship is established between the overall difference value and the correction ratio f, and the correction ratio f may be set based on the proportional relationship. FIG. 9 is a diagram illustrating an example of the relationship between the overall difference value and the correction ratio f.

  Of course, the correction ratio f is not limited to the setting based on the proportional relationship between th_ads and th_ade, and may be set so as to increase exponentially or may be set based on other functions. Good.

  This makes it possible to set a small correction ratio when the total difference value is small, and a large correction ratio when the total difference value is large. The ratio can be set and deterioration of the image after correction can be prevented.

  Note that th_ade may be a fixed value set in advance or may be set as appropriate by the operator. Further, it may be appropriately calculated for each edge region based on the calculated overall difference value.

  In the specific example of FIG. 8, the overall difference value detection unit 7 sets the correction ratio f using “277” which is the overall difference value.

(Detailed setting method of correction ratio g)
The method for setting the correction ratio g is not particularly limited as is the case with the method for setting the correction ratio f. For example, the threshold value (th_mds) set in step S56 (see FIG. 2) is set as the correction ratio 0%, the partial difference maximum value (th_mde) is set to set the correction ratio as 100%, and the calculated partial difference maximum A method of setting the correction ratio g according to the value can be mentioned. For example, as shown in FIG. 10, a proportional relationship is established between the partial difference maximum value and the correction rate g, and the correction rate g may be set based on the proportional relationship. FIG. 10 is a diagram illustrating an example of the relationship between the partial difference maximum value and the correction ratio g.

  Of course, the correction ratio g is not limited to the setting based on the proportional relationship between th_mds and th_mde, and may be set so as to increase exponentially or may be set based on other functions. Good.

  This makes it possible to set a small correction ratio when the partial difference maximum value is small, and a large correction ratio when the partial difference maximum value is large. A correct correction ratio can be set, and deterioration of the image after correction can be prevented.

  Note that th_mde may be a fixed value set in advance or may be set as appropriate by the operator. Further, it may be appropriately calculated for each edge region based on the calculated partial difference maximum value.

  In the specific example of FIG. 8, the partial difference maximum value detection unit 8 sets the correction ratio g using the partial difference maximum value = 68.

(Detailed setting method of correction ratio h)
The method for setting the correction ratio h is not particularly limited as is the case with the method for setting the correction ratios f and g. For example, a setting method in which the pixel having the maximum partial difference value calculated by the partial difference maximum value detection unit 8 is set as the center, and the correction ratio is gradually reduced as the distance from the set pixel can be gradually increased.

  A specific example of the method of setting the correction ratio h will be described below with reference to FIG. As described above, since the partial difference maximum value in the edge region 91 is the difference value between the luminance value of the pixel 90c and the luminance value of the pixel 90d, the pixel position detection unit 10 places the pixel 90c and the pixel 90d in the edge region 91. Set as the center of. In FIG. 5, the pixel 90 c and the pixel 90 d are set to the in-edge position “1”. Subsequently, the pixel 90b and the pixel 90e, which are pixels adjacent to the pixel 90c and the pixel 90d, are set to the in-edge position “2”. Thereafter, similarly, the pixel 90a and the pixel 90f are set to an in-edge position “3”, the pixel 90g is set to an in-edge position “4”, and the pixel 90h is set to an in-edge position “5”.

  When the edge position detection unit 10 sets the in-edge positions of the pixels 90 a to 90 h constituting the edge region 91, the edge position detection unit 10 sets the correction ratio h for each pixel based on the set in-edge positions. For example, the correction ratio h of the pixel 90c and the pixel 90d in which the position in the edge is set to “1” is 100%, the correction ratio h of the pixel 90b and the pixel 90e in which the position in the edge is set to “2” is 80%, The correction ratio h of the pixel 90f whose position is set to “3” is set to 60%, and the correction ratio h of the pixel 90g whose position in the edge is set to “4” is set to 40%. Note that the correction ratio h is not set because the pixels 90a and 90h are set as edge overlapping portions.

  In this way, a different correction ratio is set for each pixel constituting the edge area, and the correction ratio h is decreased as it approaches the pixels at both ends in the edge area, that is, the pixels set as the edge overlap portion. The edge area in the subsequent image can be corrected more smoothly.

  The correction ratio h at the in-edge position may be set in advance or may be set as appropriate by the operator. Further, for example, the correction is performed so that the correction ratio h for the position “1” within the edge is 100%, the correction ratio “2” is 60%, and the correction ratio “3” is 50%. The proportional relationship may not be established with the ratio h.

(Detailed setting method of correction ratio i)
The method for setting the correction ratio i is not particularly limited, as is the method for setting the correction ratios f, g, and h. For example, when the gradation detection value is 0, the correction ratio is set to 0%, a threshold value (th_dmax) is set to set the correction ratio to 100%, and the correction ratio i is set according to the calculated gradation detection value. A method can be mentioned. Specifically, as shown in FIG. 11, a proportional relationship is established between the gradation detection value and the correction ratio i, and the correction ratio i may be set based on the proportional expression. FIG. 11 is a diagram illustrating an example of a relationship between the partial difference maximum value and the correction ratio i.

  Of course, the correction ratio i is not limited to the setting based on the proportional relationship between the gradation detection value 0 and th_dmax, and may be set so as to increase exponentially or based on other functions. May be.

  As a result, a correction ratio corresponding to the gradation can be set in the portion where the gradation is formed. Therefore, the deterioration of the image quality in the image after the correction can be prevented due to the collapse of the gradation.

  Note that th_dmax may be a fixed value set in advance, or may be set as appropriate by the operator. Further, it may be appropriately calculated for each edge region based on the calculated gradation detection value.

(Specific example of correction direction setting)
A specific example of setting the correction direction will be described below with reference to FIG.

  As described above, the partial difference maximum value in the edge region 91 is a difference value between the luminance value of the pixel 90c and the luminance value of the pixel 90d. Therefore, in the edge region 91, the directionality flags of the pixels 90b and 90c are upward (“−1” in FIG. 8), and the directionality flags of the pixels 90d, 90e, 90f, and 90g are downward ( In FIG. 8, it is set to “1”). Note that the correction direction is not set for the pixels 90a and 90h, which are edge overlapping portions, because they are not corrected.

(Modification of calculation method of correction factor j)
The correction rate j calculated by the correction rate calculation unit 9 can be appropriately calculated according to the combination of the correction rates f to i. That is, in the example described above, the correction rate j is calculated using all the correction rates f to i, but is not limited to this. As a result, the correction factor j can be calculated as the most suitable value for correcting the detected edge region.

  For example, the correction rate j may be calculated using the correction rate g, the correction rate h, and the correction rate i, which are the remaining three correction rates excluding the correction rate f.

  As a result, the correction factor j based on the size of the partial difference maximum value can be calculated regardless of the size of the overall difference value.

  In addition, when the correction rate j is calculated using the correction rate f and the correction rate g, which are the remaining two correction rates excluding the correction rate h and the correction rate i, the pixels constituting the edge region Can be corrected using the same correction factor.

  In this case, although the smoothness of the image after the correction is inferior to the correction using all the correction ratios f to i, the correction can be performed without calculating the correction rate for each pixel, so that the time required for the correction can be shortened.

  Of course, the combinations of the correction ratios f to i for calculating the correction ratio j are not limited to the above example, and the combinations of the correction ratios f to i are possible in all 12 ways. Further, any one of the correction ratios f to i may be set as the correction ratio j without combining the correction ratios f to j.

  In general, the section having the maximum partial difference value is near the center of the edge, and the partial difference value decreases as it approaches the both ends of the edge region. Therefore, even if the correction rate for each pixel is equal, it is sufficiently smooth. Can be corrected.

  Further, the correction ratios f to i that are parameters for calculating the correction ratio j may be automatically set according to the characteristics of the edge region, or may be appropriately selected by the operator.

(Additional notes)
Finally, each block of the image processing apparatus 1 may be configured by hardware logic, or may be realized by software using a CPU as follows.

  That is, the image processing apparatus 1 includes a CPU (central processing unit) that executes instructions of a control program that realizes each function, a ROM (read only memory) that stores the program, and a RAM (random access memory) that expands the program. And a storage device (recording medium) such as a memory for storing the program and various data. An object of the present invention is a recording medium on which a program code (execution format program, intermediate code program, source program) of a control program of the image processing apparatus 1 which is software that realizes the above-described functions is recorded so as to be readable by a computer. This can also be achieved by supplying the image processing apparatus 1 and reading and executing the program code recorded on the recording medium by the computer (or CPU or MPU).

  Examples of the recording medium include tapes such as magnetic tapes and cassette tapes, magnetic disks such as floppy (registered trademark) disks / hard disks, and optical disks such as CD-ROM / MO / MD / DVD / CD-R. Card systems such as IC cards, IC cards (including memory cards) / optical cards, or semiconductor memory systems such as mask ROM / EPROM / EEPROM / flash ROM.

  Further, the image processing apparatus 1 may be configured to be connectable to a communication network, and the program code may be supplied via the communication network. The communication network is not particularly limited. For example, the Internet, intranet, extranet, LAN, ISDN, VAN, CATV communication network, virtual private network, telephone line network, mobile communication network, satellite communication. A net or the like is available. Further, the transmission medium constituting the communication network is not particularly limited. For example, even in the case of wired such as IEEE 1394, USB, power line carrier, cable TV line, telephone line, ADSL line, etc., infrared rays such as IrDA and remote control, Bluetooth ( (Registered trademark), 802.11 wireless, HDR, mobile phone network, satellite line, terrestrial digital network, and the like can also be used. The present invention can also be realized in the form of a computer data signal embedded in a carrier wave in which the program code is embodied by electronic transmission.

  Note that the present invention is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims, and can be obtained by appropriately combining technical means disclosed in different embodiments. Embodiments are also included in the technical scope of the present invention.

  By using the image processing apparatus according to the present invention, even when an image is enlarged and displayed, it can be displayed as an image whose sharpness is not lowered. Specifically, when an image recorded at the SD resolution is displayed on a television having an HD resolution, the image can be displayed as a sharp image in which deterioration in image quality is suppressed.

It is a block diagram which shows the principal part structure of the image processing apparatus which concerns on this invention. 3 is a flowchart showing an outline of a processing process of the image processing apparatus according to the present invention. It is a figure which shows typically the whole difference value in the edge area | region detected in the edge detection part. It is a figure which shows typically the partial difference value and partial difference maximum value in the edge area | region detected in the edge detection part. It is a figure which shows the luminance graph in the input image signal input into the image processing apparatus which concerns on this invention, the data calculated from the said image signal, and the luminance graph in the output image signal output from the image processing apparatus. It is a flowchart which shows the detailed setting process of the correction factor and correction direction of the image processing apparatus which concerns on this invention. It is a figure which shows typically an example of the gradation in the edge area | region detected in the edge detection part. It is a figure which shows the brightness | luminance graph at the time of correcting an image ignoring formation of the gradation in an edge area | region, a continuous line is a brightness | luminance graph before correction | amendment, and a broken line is a brightness | luminance graph after correction | amendment. It is a figure which shows an example of the relationship between the correction | amendment ratio f and a whole difference value. It is a figure which shows an example of the relationship between the correction | amendment ratio g and a partial difference maximum value. It is a figure which shows an example of the relationship between the correction ratio i and a gradation detection value. It is a figure which shows the conventional outline correction process, A solid line is a luminance graph in the image data before correction | amendment, and a broken line is a luminance graph in the image data after correction | amendment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 2 Edge detection part 3 Scaling process part (correction means)
4 memory controller 5 line memory 6 differential code detector (edge area detector)
7 Overall difference value detection unit (correction target area detection means)
8 Partial difference maximum value detection unit 9 Correction rate calculation unit (correction rate setting means)
DESCRIPTION OF SYMBOLS 10 Pixel position detection part 11 Correction | amendment direction setting part 12 Gradation detection part (gradation detection value calculation means)
13 Target pixel correction unit 14 Filter calculation unit

Claims (9)

An image processing apparatus comprising correction means for correcting image data of an original image by scaling processing using a predetermined correction ratio,
Edge region detecting means for detecting, as an edge region, a section in which there are three or more pixels from the pixel immediately before the sign of the luminance value difference between adjacent pixels in the image data to the pixel immediately before the next change ,
The edge area is set as a correction target area by the correction means when the overall difference value, which is the difference value between the maximum value and the minimum value of the luminance values of the pixels constituting the edge area, is equal to or greater than a predetermined threshold value. Correction target area setting means to perform,
An image processing apparatus comprising:   The image processing apparatus according to claim 1, further comprising a correction ratio setting unit that sets a value based on the overall difference value as a correction ratio for the correction target area. The image data further includes a partial difference maximum value calculating means for calculating a maximum value among partial difference values that are difference values of luminance values between adjacent pixels.
The correction target area setting means sets the edge area as the correction target area when the maximum value of the partial difference values is equal to or greater than a predetermined threshold value,
The correction ratio setting means sets a value based on the maximum value of the partial difference value as the correction ratio for the correction target area when the maximum value of the partial difference value is equal to or greater than a predetermined threshold. The image processing apparatus according to claim 2. A partial difference maximum value calculating means for calculating a maximum value among partial difference values that are difference values of luminance values between adjacent pixels in the image data;
Position detection means for detecting the position of the pixel constituting the correction target area in the correction target area based on the maximum value of the partial difference value;
The correction ratio setting means uses, for each pixel, a value based on the position for correcting the pixels constituting the correction target area when the maximum value of the partial difference values is equal to or greater than a predetermined threshold. 4. The image processing apparatus according to claim 2, wherein the correction ratio is set as the correction ratio. A partial difference maximum value calculating means for calculating a maximum value among partial difference values that are difference values of luminance values between adjacent pixels in the image data;
Gradation detection value calculating means for calculating, for each partial difference value, a gradation detection value that is a difference value between the maximum value of the partial difference value and the partial difference value;
When the maximum value of the partial difference value is equal to or greater than a predetermined threshold, the correction ratio setting unit performs the partial difference value for each pixel that is the calculation target of the partial difference value in the correction target region. 5. The image processing apparatus according to claim 2, wherein a value based on the gradation detection value corresponding to is set as the correction ratio used for correcting the pixel. 6.   The image processing apparatus according to claim 1, wherein the correction unit corrects pixels other than pixels at both ends in the correction target region. An image processing method including a correction step of correcting image data of an original image by a scaling process using a predetermined correction ratio,
An edge region detection step of detecting, as an edge region, a section in which there are three or more pixels from the pixel immediately before the change in the sign of the luminance value difference between adjacent pixels in the image data to the pixel immediately before the next change ,
The edge region is set as a correction target region in the correction step when an overall difference value that is a difference value between the maximum value and the minimum value of the luminance values of the pixels constituting the edge region is equal to or greater than a predetermined threshold value. Correction target area setting step to be performed,
An image processing method comprising:   A program for operating the image processing apparatus according to any one of claims 1 to 6, wherein the program causes a computer to function as each of the above means.   A computer-readable recording medium in which the program according to claim 8 is recorded.

Images