JP2006065688A - Image processor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processor for properly correcting image data, including the face of a person and performing correction in a short processing time in correcting the image data, even if a plurality of CPUs do not exist. <P>SOLUTION: An image processor 1 comprises a specified color designating means for designating a specified color; a specified color detecting means for detecting the designated specified color in the image data; an image pattern storage means for storing different image patterns; an image processing area setting means for matching the image pattern with the image data and setting an image processing area in the image data; a non-specified color detecting means for detecting a color other than the specified color, which is detected by the specified color detecting means within the image processing area which is set by the specified color detecting means and the image processing area setting means; and a correcting means for variously correcting the area of the color other than the specified color. The correcting means performs correction, by replacing the area of the color other than the specified color with the specified color in the neighborhood. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an image processing apparatus that processes received image data, and more particularly to an image processing apparatus that corrects a face portion of image data including a human face image.

  Conventionally, image data captured by a digital camera or the like is displayed on a personal computer and various corrections are performed. In particular, for image data including a human face, for example, by displaying a simulation image with makeup applied to the image data before actually performing it, it is possible to infer the result or remove spots and wrinkles from the actual image. Since it is possible to leave beautiful image data, the utility value thereof is high. Therefore, it is desired to realize a more accurate image correction method.

  By the way, in a video camera, contour enhancement processing is performed on an image pickup signal in order to compensate for characteristic deterioration of an image pickup device such as a CCD or to increase sharpness. As a result, rough skin, wrinkles, spots, etc. may be emphasized when shooting a person's face, etc., and it is possible to detect a skin-colored area and reduce the amount of contour enhancement in that area, or to reduce the frequency characteristics. Conventionally known.

  However, if the detail amount for contour enhancement is reduced in the specific color detection area, for example, a soft image that does not unnecessarily enhance rough skin, wrinkles, spots, etc. is obtained, but the detail amount at the detection area boundary is also reduced. However, if the amount of detail to be reduced is set larger than necessary, the details of the contour of the face that should not be reduced in order to maintain the image quality will be reduced and the contour will become blurred. Furthermore, since the amount of detail changes abruptly across the boundary of the specific color detection region, there is a problem that the image quality change near the boundary becomes abrupt and unnatural image data.

  A video signal processing apparatus described in Patent Document 1 (Japanese Patent Laid-Open No. 2000-152268) is a video signal processing apparatus for solving the above-described problems, and a specific color is designated for an input video signal. Specific color designating means, specific color detecting means for detecting a specific color designated by the specific color designating means for the input video signal, and high frequency component extracting means for extracting a high frequency component from the input video signal A noise removing unit that removes a signal of a noise slice level or less from the extracted high-frequency component video signal; and the noise slice level in the specific color detection region is variable based on an output of the specific color detection unit The noise slice level changes gently at the boundary between the control means for controlling and the specific color detection area and the area other than the detected specific color. It includes a smoothing means for processing, and an adding means for adding and outputting the output signal of said noise eliminating means on the input video signal.

  According to the above-described configuration, a high frequency component is extracted from the input video signal, a process of removing a signal having a noise slice level or less is performed on the extracted high frequency component video signal, and the signal is added to the input video signal. The As a result, the contour is emphasized while suppressing noise. Here, since the noise slice level can be variably controlled in the area where the specific color is detected by the control means, the noise slice level is set larger than the area other than the specific color in the area where the specific color is detected. The details of the detection area can be reduced and a smooth image can be obtained. Furthermore, since the noise slice level is processed so as to change gently at the boundary between the specific color detection area and the area other than the specific color, the outline of the specific color detection area is prevented from being blurred and has a clear outline. The image quality can be maintained, and the change in the amount of detail in the vicinity of the boundary of the specific color area can be moderated to prevent an unnatural change in image quality.

JP 2000-152268 A (Claim 2, paragraphs [0003], [0007], [0025])

  However, since the video signal processing apparatus described in Patent Document 1 performs image processing for a specific color and the entire area in a lump, the processing time is long. This is because the processing time can be shortened by dividing the relatively wide processing area into a plurality of relatively small areas and performing parallel processing using a plurality of CPUs. You have to prepare several.

  Therefore, in view of the above problems, the present inventors have studied various image processing apparatuses that can reduce the processing time without degrading the image quality. As a result, the processing area is not made the entire face as described above. The present inventors have found that if spot and wrinkle can be corrected in a spot manner, good image correction can be performed without performing parallel processing using a plurality of CPUs.

  That is, an object of the present invention is to provide an image processing apparatus capable of performing good correction of image data including a human face.

  Another object of the present invention is to provide an image processing apparatus capable of correcting image data in a short processing time without a plurality of CPUs.

In order to achieve the above object, the image processing apparatus according to claim 1, a specific color specifying unit that specifies a specific color based on image data including a specific image,
Specific color detection means for detecting a specific color designated by the specific color designation means in the image data;
Image pattern storage means for storing various image patterns;
Image processing area setting means for matching an image pattern stored in the image pattern storage means with the image data and setting an image processing area in the image data;
Non-specific color detection means for detecting a color other than the specific color detected by the specific color detection means within the image processing area set by the specific color detection means and the image processing area setting means;
A correction unit that performs various corrections on a color region other than the specific color detected by the non-specific color detection unit,
The correction means corrects the color area by replacing a specific color area detected by the non-specific color detection means with a specific color in the vicinity thereof.

  The invention according to claim 2 relates to the image processing apparatus according to claim 1, wherein the specific image is an image of a human face, and the specific color is a skin tone. To do.

  According to a third aspect of the present invention, there is provided the image processing apparatus according to the first aspect, wherein the image pattern storage means stores a human face pattern.

  The invention according to claim 4 relates to the image processing apparatus according to claim 1, wherein the non-specific color detection means uses a skin tone region from image data including a human face by the specific color detection means. And a color other than a specific color is detected by matching a human face pattern in the image pattern storage unit with the image data by the image processing area setting means.

  According to the first aspect of the present invention, the processing area to be subjected to image processing is detected by the specific color detection means and the image processing area setting means, and the specific color designated by the specific color designation means is detected in the detected area. By detecting different color areas with the non-specific color detection means and correcting the different color areas, it is possible to delete color elements other than the specific colors in the processing area and the set area. .

  According to the second and third aspects of the present invention, image data including a person's face is matched with a person's face pattern stored in the image pattern storage unit to determine a processing area, and further, a non-specific color detecting unit performs skinning. If a color other than a tone is detected, it is possible to detect a person's spots and wrinkles in the image data, and by correcting this, it is possible to obtain image data including a person's face from which the spots and wrinkles have been removed. become.

According to the invention of claim 4, the face stored in the image pattern storage means is obtained by matching the human face pattern by the image processing area setting means after detecting the skin tone area by the specific color detecting means. The portion that should be a skin tone in the pattern of the pattern and the area of the skin tone detected by the specific color detecting means can be compared. If it is present, it can be immediately determined as a spot or a wrinkle, so that a spot or the like can be detected with high accuracy by a simple method.
Further, since the face pattern of the person is matched with the image data by the image processing area setting means, it is possible to prevent correction of parts that should not be corrected (for example, eyebrows, facial contours, etc.). become.

  Hereinafter, the best embodiment of the present invention will be described with reference to the drawings. However, the embodiments described below exemplify image processing apparatuses for embodying the technical idea of the present invention, and are not intended to specify the present invention, and are not intended to limit the scope of the claims. Other embodiments included in are equally applicable.

  FIG. 1 is a block diagram showing a configuration of an image processing apparatus of the present invention, FIG. 2 is a flowchart showing steps of color correction processing of image data supplied to the image processing apparatus of the present invention, and FIG. 3 is an unspecified color It is a figure which expands and shows this area | region.

  The image processing apparatus 1 of the present invention includes a scanner unit 10, a FAX unit 15, an engine unit 20, a user operation unit 25, an HDD (Hard Disk Drive) 30, a printer controller 35, and a control unit 40. Has an image reading CCD 11 for scanning an image and a DF (Direction Finder) 12 for detecting the reading direction of the image. The FAX unit 15 has a modem 16 therein, and an NCU (Network Control Unit). ) Is connected to the public line 50 via 17 and transmits and receives image data.

  The engine unit 20 is an output engine for an image supplied to the image processing apparatus 1 and includes a laser scanning unit 21 that scans an image with a laser. The user operation unit 25 includes a key input unit 26, a liquid crystal display unit 27, and an LED display unit 28, and can change various settings of the image processing apparatus 1. The HDD 30 is a non-volatile memory that stores settings of the image processing apparatus 1. The printer controller 35 is connected to a network 51 such as a LAN (Local Area Network) and receives data transmitted from a PC in the network 51. Various data are transmitted to each of the PCs 52, and a management PC 36 for managing the image processing apparatus 1 is connected as necessary.

  The control unit 40 is a part that controls various data supplied to the image processing apparatus 1, and includes a job management unit 41 that manages a plurality of supplied jobs, and an image that performs various image processes on the supplied image data. And a processing unit 42. The control unit 40 is connected to the scanner unit 10, FAX unit 15, engine unit 20, user operation unit 25, HDD 30, and printer controller 35 via an interface (I / F) 43. The image data processed by each unit such as the above is subjected to various controls by the control unit 40.

  Next, the image correction process by the image processing apparatus of the present invention will be described below with reference to FIG. However, the image processing performed here describes correction for eliminating human spots or wrinkles in image data including a human face, but is not limited to correction of human spots or wrinkles of the present invention. It will be apparent to those skilled in the art that other various corrections can be applied.

  First, as shown in FIG. 2, when the image processing apparatus 1 receives image data in step S1, whether or not to perform color correction of the image data in step S2 after performing predetermined image processing such as bitmap development. When the color correction is not performed, the process proceeds to a normal output process. When the color correction is performed, the process proceeds to step S3. At this time, whether or not to perform color correction is set by an operation by the user operation unit 25, for example.

  If it is determined in step S2 that color correction is to be performed, a specific color (skin tone in this embodiment) that serves as a reference for color correction is determined by the user operation unit 25 or the PC 52 connected via the network 51 in step S3. Set. Incidentally, since the color to be used as a reference differs for each subject in the setting of the skin tone, it is preferable to perform the subsequent processing based on the skin tone extracted from the image data. Therefore, a better correction process can be performed if the color width is set in the same manner. In step S4, a portion drawn with a skin tone in the image data is detected. In step S5, an image pattern of image data (in this embodiment, a human face pattern) stored in advance in the HDD 30 is called, and the most appropriate face pattern is selected from the face image information in the image data. , Matching with image data.

  Step S6 is a step of detecting a non-specific color region using the matching result of the specific color region and the face pattern in the image data detected in steps S4 and S5. In this detection method, as a result of matching a face pattern, a portion that is determined to be a skin tone region and that does not correspond to the specific color region detected in step S4 is determined as a non-specific color region. It is.

  When the non-specific color region is detected in step S6, the non-specific color part is corrected in step S7. In this correction method, as shown in FIG. 3, the pixel D ′ determined to be a non-specific color region is replaced with the surrounding pixel D (upper and lower pixels adjacent to the pixel D ′ in FIG. 3). As described above, if the pixel is replaced with a neighboring pixel, smooth correction can be performed without the skin tone changing extremely.

  In order to obtain a smoother image, high-frequency components may be removed using a low-pass filter after performing the above-described correction. By performing filtering in this way, the frequency change can be made gentle, and as a result, smooth and natural correction can be performed.

  As described above, the image processing apparatus of the present invention can detect a non-specific color region to be corrected by detecting a specific color region in the supplied image data and further matching the image pattern. Therefore, it is not necessary to correct the entire image data as in the prior art, and spot-like correction is possible, so that the image can be corrected in a short time without using a plurality of CPUs.

FIG. 1 is a block diagram showing a configuration of an image processing apparatus according to the present invention. FIG. 2 is a flowchart showing the steps of color correction processing of image data supplied to the image processing apparatus of the present invention; FIG. 3 is an enlarged view showing an unspecified color region.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Image processing apparatus 10 Scanner part 15 FAX part 20 Engine part 25 User operation part 30 HDD
35 Printer Controller 40 Control Unit 42 Image Processing Unit

Claims (4)

Specific color designating means for designating a specific color based on image data including a specific image;
Specific color detection means for detecting a specific color designated by the specific color designation means in the image data;
Image pattern storage means for storing various image patterns;
Image processing area setting means for matching an image pattern stored in the image pattern storage means with the image data and setting an image processing area in the image data;
Non-specific color detection means for detecting a color other than the specific color detected by the specific color detection means within the image processing area set by the specific color detection means and the image processing area setting means;
A correction unit that performs various corrections on a color region other than the specific color detected by the non-specific color detection unit,
The image processing apparatus according to claim 1, wherein the correction unit corrects the color region other than the specific color detected by the non-specific color detection unit by replacing the region with a specific color in the vicinity thereof.   The image processing apparatus according to claim 1, wherein the specific image is an image of a human face, and the specific color is a skin tone.   The image processing apparatus according to claim 1, wherein a pattern of a human face is stored in the image pattern storage unit.   The non-specific color detection unit detects a skin tone area from the image data including a person's face by the specific color detection unit, and further, a human face pattern in the image pattern storage unit by the image processing area setting unit. The image processing apparatus according to claim 1, wherein a color other than a specific color is detected by matching the image data with the image data.

Images