WO2015025999A1 - Method for changing colors of video all at once and recording medium thereof - Google Patents

Abstract

A method for changing the colors of a video all at once and a recording medium thereof are disclosed. A method for changing the colors of a video all at once, according to one embodiment of the present invention, comprises: a first step of segmenting a video into a plurality of shots; a second step of selecting one of the shots and designating at least one reference image from among images constituting the shot; a third step of inputting a target image including the same image as the reference image and having color information different from that of the reference image; a fourth step of changing the color information of the reference image to the color information of the target image; and a fifth step of changing all of the color information of all the images constituting the selected shot to the changed color information of the reference image.

Description

Batch color conversion method of video and its recording medium

The present invention relates to a color conversion method, and more particularly, to a computer-readable recording medium having a batch color conversion method of a moving picture for collectively converting color of a moving picture and a program for executing the method on a computer. It is about.

Due to the development of internet technology and stream transmission technology, various multimedia contents are actively produced and distributed. In particular, video contents are showing a great growth trend due to the proliferation of smart devices and social networks.

Most of these moving images are made by combining and editing partial images divided into various scenes and cuts. At this time, the plurality of partial images forming one video is very different from each other in color information depending on the equipment or weather shooting conditions, in some cases it is necessary to maintain a uniform color in the entire video Occurs. The technology that can be used in this case is the color conversion technology of moving pictures.

The video is composed of a number of consecutive frames. For the color conversion of the entire video, all the frames are stored in RAW and then subjected to proper image processing (eg, white balance, gamma correction, etc.). However, since RAW has a very large data capacity, it is virtually difficult to have a video in RAW format, and it takes a very long time because many frames need to be color-transformed in order to convert the color of the entire video.

There are editing tools for converting the color of the video. However, these tools convert the absolute value of the color, which completely corrects the non-uniform color difference caused by the difference in shooting conditions between images. There is a limit because it is difficult.

Embodiments of the present invention provide a batch color converting method of a moving picture and a recording medium thereof capable of batch converting colors for a whole image or a partial image of a moving picture.

According to an aspect of the present invention, a step of dividing a video into a plurality of shot (SHOT); Selecting at least one shot from among the shots and designating at least one reference video from among the images forming the shot; Inputting a target image including the same image as the reference image and having color information different from the reference image; Converting color information of the reference image into color information of the target image; And converting all the color information of all the images constituting the selected shot into a conversion relationship between the converted reference image and the target image.

In addition, after the step 5, it may further include a six step to repeat the steps from step 2 to step 5.

The video segmentation in the first step may be performed by calculating a difference between pixel values of two adjacent images and determining that the difference is the same shot if the difference is less than a threshold value, and determining the same shot if the difference is greater than or equal to the threshold value. It is available.

Also, the reference image in step 2 may be the first image of the selected shot.

The method may further include a step 2-1 of calculating a difference between pixel values of the first image and other images of the selected shot, and adding the image having the highest difference as a new reference image among the images having the difference greater than or equal to a threshold value. can do.

Further, after step 5, step 5-1 of comparing the color information of the converted entire images with the color information of the target image, respectively; Adding an image having the largest difference in color information to a reference image when an image having a difference in color information from the target image is greater than or equal to a threshold value among all the converted images; The method may further include steps 5-3 for performing steps 4 and 5 again.

The target image may be an image taken by using the same camera as the reference image, but may be an image having a different image setting, an image taken by using a camera different from the reference image, or include the same scene as the reference image. The image may be an image downloaded through the Internet, or may be an image obtained by processing the reference image using image editing software.

In addition, the step 4, the same physical position in the reference image and the target image

Step 4-1 of searching for a corresponding point having the same feature in the device; Radiating calibration of the reference image and the target image so that the camera response functions have a linear function; And 4-3 calculating the color conversion matrix to convert the color information.

According to another aspect of the present invention, a computer-readable recording medium having recorded thereon a program for executing the method according to one aspect of the present invention may be provided.

According to the exemplary embodiments of the present invention, when the color of at least one image is converted for a shot that constitutes a video, the color of the video may be converted very efficiently since the color may be converted collectively for the other image forming the shot. Batch conversion is possible.

In addition, since it is possible to use an image processed by another camera setting, another camera, or image editing software as a target image having color information to be converted, it is not necessary to separately store RAW of each image of the video.

1 is a flowchart illustrating a batch color conversion method of a video according to an embodiment of the present invention.

FIG. 2 is a flowchart illustrating an operation S120 in FIG. 1 in detail.

3 is a flowchart illustrating an operation S140 in detail in FIG. 1.

4 is a flowchart illustrating in detail the steps added after S150 in FIG. 1.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a flowchart illustrating a batch color conversion method (hereinafter, referred to as a batch color conversion method) of a video according to an embodiment of the present invention. Hereinafter, the present invention will be described in each step shown in FIG. 1.

(1) Step 1 (S110)

Step 1 is a step of dividing a video into a plurality of shots.

Briefly describing the hierarchical structure of a moving picture described in the present specification, a moving picture is made up of a large number of continuous frames (FRAME). In other words, the frame corresponds to the basic unit forming the video. In the present specification, the frame has been described as having the same meaning as 'image'.

The plurality of consecutive frames may constitute one shot. A shot may be defined as a collection of frames with a similar image as a whole. In other words, the video is composed of a plurality of shots. In general, the scene change in the video may also be understood as a shot (SHOT) has changed.

The plurality of shots may constitute one scene according to a semantic relationship. Such scenes may vary depending on the intention of the video editor (or director). That is, the scenes may vary according to how the editor groups a plurality of shots. Therefore, completely different images may exist in one scene.

Therefore, in the first step of the present invention, the segmentation of the video does not mean the division of the scene, but rather the division of the shot, which is a set of frames having similar images as a whole.

A method of dividing a video into a plurality of shots will be described.

There may be various methods of dividing a video into a plurality of shots. For example, if the difference is less than a threshold value by calculating a difference between pixel values of two adjacent images (frames), the difference is determined as the same shot. If this is the case, a method of judging and dividing into another shot may be used (shot boundary detection). Herein, the threshold may be preset.

Alternatively, the method may compare the entire color histogram between the images, determine that the shot is the same shot if the difference between the color histograms is less than the threshold value, and determine if the difference is the other shot. .

Alternatively, if there are not many scene changes in the video or the video length is relatively short, the user may manually divide the shot.

Alternatively, when audio information is included in the video, the audio information may be used. For example, since a person's voice has unique pitch information for each individual, a method of dividing a shot by analyzing a change in pitch and detecting a change in a speaker may be used.

As such, a method of dividing a video into a plurality of shots may use various known methods, and is not limited to the above-mentioned method (S110).

(2) Step 2 (S120)

In the second step, any one shot is selected from the divided shots, and at least one reference image is designated among the images (frames) constituting the shot.

As described above, one shot is composed of a plurality of images. The images have similar pixel values and color histogram values.

The reference image may be at least one of a plurality of images forming a shot. For example, the reference image may be the first image (frame) of the selected shot (DEFAULT value). In the present specification, the first image of the selected shot is referred to as a first reference image.

A plurality of reference images may be designated. This is because an image having a large difference in pixel value and color histogram value from another image may belong to the same shot, and in this case, a color conversion result may be deteriorated. In the present specification, an image added as a reference image in addition to the first reference image will be referred to as a second reference image.

FIG. 2 is a flowchart illustrating in detail S120 of FIG. 1. The second reference image may be designated through the same steps as in FIG. 2.

In detail, when a first reference image is designated (S121), a difference in color information between the first reference image and other images of the selected shot is calculated. Images in the same shot have relatively similar color information, but there may be a difference in color information between the images, and in some cases, the difference may be large (S122). Meanwhile, in the present specification, color information refers to color characteristics such as brightness, saturation, color, etc. of an image as well as a color model such as RGB (Red-Green-Blue) obtained from an image.

At this time, when compared with a predetermined threshold value (for example, the threshold value is 10%) (S123), if the difference between the color information (between the first reference image and other images of the selected shot) is less than the threshold value between the images Since color information is at a relatively similar level, there is no need to additionally specify a second reference image. Therefore, the process may proceed to S130. However, when the color information difference is greater than or equal to a threshold, reliability of the color conversion result to be described later may be improved by adding an image having the largest color information difference as the second reference image (S124).

Meanwhile, the second reference image may be additionally designated even after the color conversion is performed, and this will be described later (above S120).

(3) step 3 (S130)

Step 3 is a step of inputting a target image. In the present specification, the target image includes the same image as the reference image, and means an image having color information different from that of the reference image. That is, the target image is an image having color information which is a target of conversion of color information of the reference image.

The target image may include the same image as the reference image and is not limited to a specific kind.

For example, the target image is an image taken using the same camera as the reference image, but the image has a different image setting (where the image setting means image setting such as white balance, gamma correction, picture style, brightness adjustment, etc.). Can be.

In addition, the target image may be an image taken using a different camera from the reference image (different cameras have different spectral sensitivity, gamma, style, etc.).

In addition, the target image may be an image downloaded through the Internet when it includes the same scene as the reference image.

In addition, the target image may be an image obtained by processing the reference image using image editing software (editing tools such as Photoshop and paint shop).

For example, an image taken using a digital single-lens reflex camera (DSLR) is different from an image taken using a smartphone. That is, since the photographing equipment is different, the colors of the two people are different. At this time, assuming that the image taken using a DSLR is a reference image and the image taken using a smartphone is a target image, the reference image is converted into a target image through a batch color conversion method of a video according to an embodiment of the present invention. Can be converted to have the same color. In other words, the video taken with the DSLR can be converted like the video taken with a smartphone. Of course the reverse is also possible (more than S130).

(4) Step 4 (S140)

Step 4 is a step of converting color information of the reference image into color information of the target image when the target image is input as in S130 (color correction). The color information conversion method may use various color correction methods known in the art.

Related Korean Patent No. 10-1227936, filed and registered by the inventors of the present invention, describes a color correction method of a corresponding image. In the present invention, step S140 may be applied to the color correction method of the corresponding image described in the patent. have. That is, step S140 of the present invention may include the contents of the patent, which will be briefly described below.

FIG. 5 is a flowchart illustrating an operation S140 in detail in FIG. 1.

Referring to FIG. 5, step 4 is a step 4-1 of searching for a corresponding point having the same feature while having the same physical location in the reference image and the target image, and a camera response function of the reference image and the target image. Radiation calibration means that the camera response function of the image is calculated to have a linear function, and the inverse function of the image is added to the image to calibrate the camera response function to be a linear function. and a 4-3 step of converting color information by calculating a color conversion matrix. A detailed description of each step is described in detail in Korean Patent Registration No. 10-1227936, and will be omitted here.

As a result, the color information of the reference image is converted into the color information of the target image (S140).

(5) Step 5 (S150)

Step 5 is a step of converting all of the color information of all the images forming the selected shot in a conversion relationship between the converted reference image and the target image. That is, when the color information of the reference image is converted in step 4, a conversion relationship between the reference image and the target image is derived, and color information of all images in the same shot may be collectively converted.

As described above, all images in the same shot are images having relatively similar color information to the reference image. Therefore, converting all the images in the same shot to the color information of the converted reference image can be relatively fast and reliable.

On the other hand, after the color information is collectively converted for all the images in the same shot, it is possible to perform the color conversion again by adding a reference image (second reference image) according to the conversion result.

As mentioned above, images in the same shot have relatively similar color information, but there may be a difference in color information between the images, and in some cases, the difference may be large. Therefore, a plurality of reference images may be designated from the beginning, but after converting the color information for all images in the same shot, it is possible to additionally specify the reference image according to the conversion result.

6 is a flowchart illustrating in detail the steps added after S150 in FIG. 1.

Referring to FIG. 6, color information of the entire images (images in the same shot) converted in operation S150 is compared with color information of the target image, respectively (S151). In this case, when the difference in color information from the target image is less than a threshold value among the entire images from which the color information is converted, it may be considered that the color conversion of the entire images is properly performed. Therefore, it may proceed to the next step (S160).

However, when there is an image having a color information difference greater than or equal to a threshold value among all images in which color information is converted (S152), it may be considered that color conversion for all images is not properly performed. Therefore, it is necessary to specify additional reference images. Therefore, in such a case, the image having the largest difference (corresponding to the image having the worst color conversion result) among the images whose color information difference is greater than or equal to the threshold value is added as the second reference image (S153), and then the image is re-started from step S140. By doing this, the reliability of color conversion can be improved. Since the target image is not changed even when the second reference image is added, the image having the worst color conversion result is converted to be closer to the target image (S150).

(6) step 6 (S160)

When the process up to step 5 is completed, the color conversion is performed for the selected shot among the plurality of shots that constitute the video. Therefore, when a color conversion is to be performed for other shots, a new shot is selected and repeated steps S120 to S150 are performed. By repeating the above process, images may be collectively converted to have the same or similar color information as the desired target image for the entire video.

As described above, in the embodiments of the present invention, when the color of at least one image is converted for a shot that constitutes a video, the color may be collectively converted for other images forming the shot. You can batch convert the colors of the video. In addition, since it is possible to use an image processed by another camera setting, another camera, or image editing software as a target image having color information to be converted, there is an advantage that it is not necessary to separately store the RAW of each image of the video.

The batch color converting method for moving images according to the present invention described above may be embodied as computer readable codes on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like, and may also be implemented in the form of a carrier wave. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. In addition, functional programs, codes, and code segments for implementing the color correction method of the corresponding image may be easily inferred by programmers in the art.

As described above, embodiments of the present invention have been described, but those skilled in the art may add, change, delete, or add components within the scope not departing from the spirit of the present invention described in the claims. The present invention may be modified and changed in various ways, which will also be included within the scope of the present invention.

Claims (9)

Dividing the video into a plurality of shots; Selecting at least one shot from among the shots and designating at least one reference video from among the images forming the shot; Inputting a target image including the same image as the reference image and having color information different from the reference image; Converting color information of the reference image into color information of the target image; And And converting all the color information of all the images constituting the selected shot into a conversion relationship between the converted reference image and the target image. The method according to claim 1, After step 5, Batch color conversion method of the video further comprising the step of repeating from step 2 to step 5. The method according to claim 1, The video segmentation in the first step is to calculate the difference between the pixel values of the two adjacent images to determine the same shot if the difference is less than the threshold value, and to determine if the video is divided into different shots if the threshold value is greater than the threshold value Batch color conversion method. The method according to claim 1, And the reference image in step 2 is the first image of the selected shot. The method according to claim 4, And a step 2-1 of calculating a difference between pixel values of the first image and other images of the selected shot, and adding the image having the highest difference as a new reference image among the images having the difference greater than or equal to a threshold value. Batch color conversion method. The method according to claim 4, After step 5, Step 5-1 of comparing color information of the converted entire images with color information of the target image; Adding an image having the largest difference in color information to a reference image when an image having a difference in color information from the target image is greater than or equal to a threshold value among all the converted images; The batch color conversion method of the video further comprising a step 5-3 performing the steps 4 and 5. The method according to claim 1, The target image is an image taken using the same camera as the reference image but has a different image setting, an image taken using a different camera from the reference image, or includes the same scene as the reference image. A batch color converting method of a video, which is a video downloaded through the video or a video processed by the video editing software. The method according to claim 1, The fourth step, Searching for corresponding points having the same features in the same physical location in the reference image and the target image; Radiating calibration of the reference image and the target image so that the camera response functions have a linear function; And A batch color conversion method of a video, comprising 4-3 steps of converting color information by calculating a color conversion matrix. A computer-readable recording medium having recorded thereon a program for executing the method according to any one of claims 1 to 8.

Images