JP2003123061A - Image processing method for digital image data - Google Patents

Abstract

(57) [Problem] To provide an image processing method of digital image data that can easily and smoothly perform repair work and can greatly improve image quality after repairing digital image data. . A repair object (31) provided by image processing software is used to select a repair object (31) when repairing digital image data (10). Read the color gradient value of each pixel. Also,
This is an image processing method of arranging the read picture elements in the order of color gradient values, selecting a color gradient value of an intermediate number, and replacing the color gradient value of the picture element corresponding to the position to be repaired.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital image data image processing method for repairing a pixel of digital image data.

[0002]

2. Description of the Related Art In recent years, electronic science and technology have advanced day by day. Computers and their peripheral products are constantly improving their performance, and various softwares are constantly being developed with new products, and their prices are lower than before, so they are popular in our daily work and life. Because of the above trends, many visual information product vendors are researching and developing new digital electronic devices in consideration of the large emerging computer consumers. Taking a digital camera as an example, a large manufacturer attracts the consumer with its products and occupies the market share, so when the consumer purchases the digital camera, he / she provides various image processing software. The user reads the digital image data captured by the digital camera with the software, links it with a display device such as a computer or a television, and views or edits the image captured by the computer or the television. Therefore, the cost of photo phenomenon is saved, the images can be laid out or processed in various ways, and the waste of photo save space and correlation processing resources can be greatly reduced.

Currently, various existing image processing softwares usually provide a function of editing and processing digital image data. However, in order to handle defects of the digital image data by the user, the digital image data of the autopsy is processed. It is necessary to repair the pixels and remove the spots. However, in the conventional image processing software, when the digital image data is repaired to remove the spots, the user clicks the eraser object (OBJECT) from the tool box provided by the software to select the defective image pixel of the digital image data. Work to erase. next,
Select a brush object for painting from the tool box and repair the erased pixel position.

[0004]

However, the above-described operation process is not only very complicated, but also requires a great deal of skill when repairing it, so that ordinary users cannot easily bear the work. . Traditionally, the service item of photographers has always been a service of repairing photographs. Since such a service is a very precise and difficult technology, a specialist who has been involved in the service has a wealth of repair experience when repairing defects in the film or digital image data, as well as photography technology, color It is also necessary to recognize. After repairing or removing fine defects in the film or digital image data, a natural color tone that matches the color of the surrounding normal background appears at the repaired or removed spots. For example, as shown in FIG. 1, the image data 10 of a human photograph will be described as an example. If you want to remove the hair 21 over the forehead of the human photograph 20 and keep the natural skin color of the forehead, unless you are a professional repairer with extensive experience, you can apply, wipe, blur, imitate such image processing software. Not even with the best editing tools. In other words, it is impossible to give the forehead a natural skin color after removing the hair. Forcing such an objective to be achieved takes a long time and enormous energy. Therefore, the editing tools provided by all conventional image processing software are somewhat effective for repairing defects in digital image data and removing spots, but cannot repair or remove spots like a specialist. Therefore, even if the user wants to repair the defects in the digital image data, it is difficult to easily try those editing tools. An object of the present invention is to provide a digital image data image processing method that can perform repair work significantly easily and smoothly and that can significantly improve the image quality after repairing digital image data.

[0005]

In order to solve the above problems, the present invention provides a digital image data image processing method for repairing a pixel of digital image data. The method uses the repair object provided by the image processing software. Then, when the digital image data is repaired, the color gradient value of the pixel selected in the repair object, the position to be repaired around the position to be repaired, and the color gradient value of each pixel in the adjacent fixed range is read. Further, after arranging the read pixels in the order of the color gradient values, the color gradient value of the intermediate number is selected, and the color gradient values of the pixels corresponding to the position to be repaired are replaced.

When the pixels of the digital image data are repaired to remove the spots, the object to be repaired and the spots to be removed can be identified based on the background color in a certain range around the position to be repaired, and the background of the background can be identified. Since the color does not change and does not affect, the repair work can be performed significantly easily and smoothly. Moreover, the image quality after repairing the digital image data can be greatly improved.

Also, the image processing software provides at least one or more repair objects, and when each repair object repairs digital image data, at least one or more selected positions to repair are selected according to different repair objects. The color gradient value of each pixel in the position to be repaired, and in a certain range adjacent to it, is read with the pixel in the center as the center. Further, after arranging the read pixels in the order of the color gradient values, the color gradient value of the intermediate number is selected, and the color gradient values of the pixels corresponding to the position to be repaired are replaced. Therefore, the user selects a different repair object, and performs a repair work such that the position to be repaired is finely or blurred.

[0008]

DETAILED DESCRIPTION OF THE INVENTION The present invention utilizes repair objects provided by image processing software when repairing any digital image data on a computer monitor. On the other hand, when repairing digital image data,
Focusing on the position to be repaired by selecting the repair object, the color gradient value of the position to be repaired and the color gradient value of each pixel in a certain range adjacent thereto is read. Further, after arranging the read pixels in the order of the color gradient values, the color gradient value of the intermediate number is selected, and the color gradient values of the pixels corresponding to the position to be repaired are replaced. In that way
When the pixels of the digital image data are repaired to remove spots, it is possible to identify the object to be repaired and the spots to be removed based on the background color in a certain range around the position to be repaired, and the background color of the background changes. Since there is no impact, repair work is much easier and smoother.

Embodiments of the present invention will be described below with reference to the drawings. A first embodiment of the present invention will be described. Please refer to FIG. After the image processing software of this embodiment is installed in the computer, the user operates the software to read the digital image data 10 and display it on the monitor of the computer. Then, the user uses the repair object 31 provided by the tool box 30 of the image processing software. Manipulate and control the repair object 31 via a mouse or other input device (eg, a sensitive pen on a digital construction board) to repair a digital image data sight 20 (eg, a human photo).

Referring to FIG. In the present embodiment, the user repairs the hair on the forehead 21 of the human photograph 20 by removing the hair. The cursor 40 of the repair object 31 is operated and controlled to move it to the hair position above the forehead 21 of the human photograph 20 to be removed. Also, click the click button on the input device to select the part you want to remove. At that time
Based on the following formula, the image processing software reads the color gradient value of each pixel in the adjacent square area whose side length is nSquare to be repaired based on the following formula based on the selected region.

NSquare = nRadius × 2 + 1 ··· (1) nRadius: Radius range value of a pixel centered on the selected hair portion, nSquare: Selected hair In the present embodiment, the image processing software reads nSquare and then reads all the pixels included in the square range having the side length of nSquare. The color gradient values of those pixels are arranged in the order of large and small. The color of the spots or lines that you want to repair in the digital image data has a clear difference from the surrounding background color, and the number of pixels should be smaller than the surrounding background pixels, so those pixels are When arranged in order, the color gradient value of the pixel of the spot or line to be repaired is always the number value at the front or the end. Since there are many other background pixels included in the square range whose side length is nSquare, we know that the number should be close to the middle position. In short, by examining the distribution of the sequence numbers of those pixels, the difference between the background color of the object to be repaired and the spots to be removed can be easily identified in a certain range around the position to be repaired.

This embodiment utilizes such a characteristic, and after arranging the color gradient values of those pixels in order, calculates the intermediate order number by the following formula. Intermediate sequence number = (nSquare x nSquare-1) / 2 (2) Moreover, the color gradient value corresponding to the intermediate sequence number is selected, and as the repair value, the pixel at the position to be repaired is selected. Replace the corresponding original color gradient value. Then, in response to the user moving the cursor 40 of the repair object 31, while moving to the region of the forehead 21 of the human photograph 20 to be removed one by one, by pressing the click button of the input device as shown in FIG. When the hairs to be removed are selected, all the hairs on the forehead 21 of the human photograph 20 are removed one by one. In addition, as shown in FIG. 4, the background color of the forehead 21 of the human photograph 20 is automatically complemented, and the base color does not change and there is no effect. Therefore, the user can perform the work of repairing the forehead significantly easily. It can be done smoothly.

In this embodiment, the image processing software uses the hair selected on the forehead 21 of the human photograph 20 as the center and sets the five pixels around the position to be repaired as the radius (that is, nRadius = 5). As shown in FIG. 5, according to the formula (1) described above, the image processing software has nSquare as the position to be repaired and the side length around the selected hair part.
= NRadius x 2 + 1 = 11 read the color gradient values of a total of 121 pixels in the adjacent area of the square, and arrange them in order according to the magnitude of the color gradient values. At that time, 11 spots A2, A3, A4 ... A11, A12 to be compensated
Color gradient values of 220, 221, ... 22 respectively
5 and 225, and the color gradient values of 110 pixels other than the surrounding background are 35. Therefore, after arranging the order of the color gradient values of those pixels, 11 spots A2 and A3 to be repaired are arranged. , A4 ... A11, A12, the sequence number values of the color gradient values are 1, 2, 3, ... 11, 1 respectively.
It becomes 2. The sequence number values of the other 110 background pixels included in the adjoining range of the square whose side length is nSquare are large, and therefore the sequence numbers are distributed near the intermediate position. Therefore, the background color gradient value 35 corresponding to the intermediate order number = (nSquare × nSquare−1) / 2 is used as a repair value to replace the original color gradient value 220 of the pixel A2 to be repaired. By repeating the above-mentioned operation, all the hairs on the forehead 21 of the human photograph 20 are removed one by one, and the background color is automatically complemented, so that the repair work can be smoothly performed.

Referring to FIG. In a second embodiment of the present invention, the image processing software toolbox 30 provides at least one or more repair objects 311. The user selects, manipulates, and controls each repair object 31 via a mouse or other input device (e.g., a sensitive pen on a digital drawing board) to view a scene 20 (e.g., a human photo) of digital image data. Repair. According to the different repair object 31 used, the image processing software determines the color gradient value of each pixel in a certain range adjacent to the position to be repaired, centering on at least one or more pixels at the selected position to repair. read. Further, after arranging the read pixels in the order of color gradient values, the color gradient value of the intermediate number is selected, and the color gradient values of at least one or more pixels corresponding to the position to be repaired are replaced. Then, the user can select a different repair object 31 and perform a fine or vague repair work on the position to be repaired.

[Brief description of drawings]

FIG. 1 is a schematic diagram showing a state in which digital image data is displayed on a monitor of a computer by image processing software in an image processing method of digital image data according to a first embodiment of the present invention.

FIG. 2 is a schematic diagram showing a screen for repairing digital image data according to the first embodiment of the present invention.

FIG. 3 is a schematic diagram showing a screen for repairing digital image data according to the first embodiment of the present invention.

FIG. 4 is a schematic diagram showing a screen for repairing digital image data according to the first embodiment of the present invention.

FIG. 5 is a schematic diagram showing a pixel distribution of a portion to be repaired in the image processing method of digital image data according to the first embodiment of the present invention.

FIG. 6 is a schematic diagram showing a screen of a repair object according to a second embodiment of the present invention.

[Explanation of symbols]

10 Digital image data 30 tool boxes 31 Repair Object

Claims (7)

[Claims] 1. A step of repairing digital image data by using a repair object provided by image processing software on a monitor of a computer, and clicking to select a position of the digital image data to be repaired, and the position to be repaired. Centering on, the position to be repaired, and the step of reading the color gradient value of each pixel in a predetermined range adjacent to the position to be repaired, the step of arranging each of the pixels in the order of the color gradient value, And a step of selecting a color gradient value of an intermediate number of each of the arranged pixels and replacing the color gradient value of the drawing element corresponding to the position to be repaired, the image processing method of digital image data. 2. The repair object is provided in a tool box provided by the image processing software, and when the repair object is operated and controlled by an input device connected to the computer, a scene of the digital image data is displayed. The image processing method for digital image data according to claim 1, wherein repairing is performed. 3. The repair object is moved to the position to be repaired, a click button of the input device is pressed to select a part to be removed, and at the time of selection, the image processing software focuses on the selected part as follows. Formula (1), nSquare = nRadius × 2 + 1 (1) nRadius: Radius range value of a pixel centered on the selected region, nSquare: Based on the side length of a square centered on the selected site, the position to be repaired and the side length are nS
3. The image processing method for digital image data according to claim 2, wherein the color gradient value of each pixel in the adjacent square area that is a quare is read. 4. The image processing software, after reading the color gradient values of all the pixels included in the square range of the side length nSquare, arranges them in order according to the magnitude of the color gradient values. The image processing method for digital image data according to claim 3. 5. The image processing software, after arranging the color gradient values of the pixels in order, calculates an intermediate sequence number by the following formula (2), and the intermediate sequence number = (nSquare × nSquare -1) / 2 ... (2) Select a color gradient value corresponding to the intermediate sequence number as a repair value, and replace the original color gradient value corresponding to the pixel at the position to be repaired. 5. The method according to claim 4, wherein
An image processing method of the described digital image data. 6. The tool box provides at least one of the repair objects, and when repairing a scene of the digital image data with the repair object, at least one of the desired repair positions is selected according to the different repair object. The position to be repaired centering on one pixel, and the color gradient value of each pixel in a predetermined range adjacent to the position to be repaired is read, and after the respective pixels are arranged in the order of the color gradient value, 3. The digital image data according to claim 2, wherein a color gradient value of an intermediate number of each of the arranged pixels is selected and the color gradient value of at least one pixel corresponding to the position to be repaired is replaced. Image processing method. 7. When repairing a scene of the digital image data with the repair object, the position to be repaired and the respective corresponding pixels in different corresponding ranges adjacent to the position to be repaired are different according to the different repair object. The image processing method for digital image data according to claim 6, wherein a color gradient value is read.