KR101663042B1

KR101663042B1 - Photo fusing apparatus

Info

Publication number: KR101663042B1
Application number: KR1020150082194A
Authority: KR
Inventors: 김태성; 김혜영; 이한수; 김혜경
Original assignee: 주식회사 아이오로라
Priority date: 2015-06-10
Filing date: 2015-06-10
Publication date: 2016-10-07

Abstract

A photo synthesizing apparatus is disclosed. A camera for photographing a moving picture; And a process of performing a channel-by-channel image correction in the photo object and the moving image and compositing the combined image. Accordingly, the user can feel new pleasure in the process of issuing a ticket through instant photo shooting with a photo object (star / character / background, etc.) selected by the user.

Description

[0001] PHOTO FUSING APPARATUS [0002]

Field of the Invention [0002] The present invention relates to a photo-synthesizing apparatus, and more particularly, to a photo-synthesizing apparatus for generating a synthesized image.

Unmanned ticket ticketing machines are devices that issue tickets on behalf of people at cinemas, sporting venues, concert halls, exhibitions, museums, and sightseeing spots. Ticket shows viewing information such as date, movie name / game name / performance name / exhibition name, viewing place, viewing position and the like.

A general unattended ticketing machine consists of a PC, a monitor, a receipt printer, a card reader, a bill recognizer and a fee settlement department. In recent years, with the development of image data processing and composition techniques, the user is provided with an instant photography service with a star / character / background using a camera, thereby providing additional enjoyment to the user. It is possible to collect the tickets by printing the photographed pictures with the printer attached. In addition, the photographed pictures can be transmitted to a mobile phone, an e-mail, and an SNS via a network connection and can be shared.

The photo synthesizer generates an image of a composite image of a camera image and an object image and prints it as a ticket. At this time, there is a problem that the image obtained through the camera is not provided with satisfactory image quality due to the color mismatch with the photo object to be synthesized due to the influence of the ambient light source.

SUMMARY OF THE INVENTION It is an object of the present invention to solve the problems described above, and to provide a photo synthesizing apparatus for synthesizing two image-corrected images and printing them on paper or a media medium.

According to an aspect of the present invention, And a background image (hereinafter referred to as a " photo object ") and a camera moving image.

Here, the printer further includes a printer for printing the composite image on the ticket.

At this time, the image correction includes at least one of histogram matching, filtering, and color correction. The histogram matching estimates minimum and maximum values for each channel, performs histogram stretching based on the estimated minimum and maximum values, The image processing is performed using a lateral filter. The color correction corrects the color of the entire image, or the skin color is corrected through face detection.

At this time, the color correction is performed by extracting the saturated component using the CIELAB and HSV color space conversion of the image and increasing the extracted saturated component.

At this time, the skin color correction corrects the skin color and the gray tone.

At this time, the process estimates and corrects the light source for the moving image.

At this time, the process synthesizes the camera image and the photo object by improving color and contrast.

When the photo synthesizing apparatus according to the present invention as described above is used, the user can feel new pleasure in the process of issuing a ticket through instant photographing with a star / character / background selected by the user.

In addition, it is possible to add the value as a collection and record the memories of the user by leaving the photographs taken with the star / character / background as a ticket.

In addition, by transferring photographs taken on mobile phones, e-mails, and social networking sites, it is possible to pursue advertising effects such as movies, sports events, performances, and exhibitions through photo sharing.

In addition, based on the power of the Internet (SNS), it can be expected to develop into a new trend influencing the movies, sports events, performances, and exhibition publicity markets.

In addition, it can be developed as an application on smart phone, and it can develop new pleasure and new advertisement trend through synthesis of camera image and photo object.

1 is a block diagram showing a configuration of a photo synthesizing apparatus according to an embodiment of the present invention.
2 is a diagram illustrating a process of combining two images by improving color and contrast.
3 is a diagram illustrating a process of estimating brightness of a light source.
4 is a diagram illustrating a process of correcting brightness of a light source.
5 is a diagram illustrating a process of synthesizing a camera image and a photo object.
6 is an exemplary view showing photographing through a camera connection.
FIG. 7 is an exemplary view showing a process of reproducing a preferred skin color.
8 is an exemplary view showing an input image and a skin color correction image through the reproduction of a preferred skin color.

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

1 is a block diagram showing a configuration of a photo synthesizing apparatus according to an embodiment of the present invention.

The image unit 110 prepares a user image and a photo object. The photographing unit 111 photographs a user image, and the selecting unit 112 selects a photo object. The photo (star / character / background) object is a pre-made star / character / background image. The light source correction unit 120 estimates the light source from the user image obtained from the camera and corrects the light source. The fusion unit 130 combines the user image and the photo object. The filter processing unit 140 filters the synthesized image. The output unit 150 outputs the synthesized image to a paper or media medium. The output unit 150 may include a printer. The printer is a photo printer. The communication unit 160 transmits the synthesized image to the mobile phone, the e-mail, and the SNS through the network.

The photo synthesizer processes synthesis between the photo object and the acquired image. The photo synthesizer acquires the image from the camera and reproduces it on the display. In order to simultaneously reproduce the photo object and the real-time camera image, the photo-synthesizing apparatus overlays the photo objects until the photographing is performed on the camera image in an animation form and displays it on the screen.

2 is a diagram illustrating a process of combining two images by improving color and contrast.

- Histogram Matching: The photo synthesizer estimates the minimum and maximum values for each channel in the photo object and the acquired image. Based on the minimum and maximum values of the minimum and maximum values, the histogram matching method uses the histogram stretching technique to set the value between channels .

- Filtering and Color Correction: The photo synthesizer performs additional filtering and color correction processing on the images acquired by the camera before composing the matched images. Filtering and color correction were performed globally or locally on the image. The whole region means the entire image, and the region applies only a specific part of the image, such as skin color correction through face detection of the image. Filtering uses a bilateral filter to reduce edge noise, blurring, and color noise in the flat area while preserving the edges of the image. The color correction is performed by extracting only the saturation component using the color space conversion, which can separate chrominance and brightness components such as CIELAB and HSV of the image, and increasing it.

The photo synthesizer fuses the photo object with the image through color and contrast enhancement. The photo synthesizer synthesizes the image obtained by the camera with the photo object to generate a composite image. The photo synthesizer acquires the image from the camera, processes it, composites it with the photo object, and outputs it as a photo card.

3 is a diagram illustrating a process of estimating brightness of a light source.

The photo synthesizer obtains the brightness image (Y) through the YCbCr color space transformation to extract the brightness component of the input image. At this time, HSV, HIS, CIELAB and the like can be used for the color space for obtaining the brightness image.

- Estimation of Brightness Components of Light Sources Using Gaussian Kernels: The photo synthesis apparatus generates three Gaussian kernels (G) by setting the standard deviations of the Gaussian functions to be different from each other, and then applies them to the brightness component Y of the image . Then, the photo synthesizer estimates the brightness component (S) of the light source in the image by summing weights (w) of the images to which the kernel is applied.

Equation 1

Technique to correct brightness of light source

4 is a diagram illustrating a process of correcting brightness of a light source.

The photo synthesizer compensates the brightness of the light source in the image. The photo synthesizer compensates the brightness of the light source by dividing the estimated light source S using Equation (2) into each channel (I) of the input RGB image of the image.

Equation 2

The photo synthesizer fuses the photo object with the image through color and contrast enhancement. The photo synthesizer adjusts the color and contrast in each image to synthesize the image that has undergone the preprocessing, such as photo object and light source correction, and synthesizes it by similar conversion. In order to analyze the contrast between the photo object and the image obtained from the camera, the photo synthesizer estimates the minimum and maximum values by calculating the histogram between the respective channels, and performs histogram matching to match the two images. In addition, the photo synthesizer adjusts the filter and the color in addition to the image acquired from the camera to synthesize the image with the photo object.

5 is a diagram illustrating a process of synthesizing a camera image and a photo object.

The photo synthesizer acquires the image from the camera and reproduces it on the display. The photo synthesizer estimates the light source from the image obtained from the camera and corrects the light source. The photo synthesizer generates a composite image through image fusion technology. Composite images include image fusion with image object through enhancement of color and contrast, and image fusion through correction of skin color and gray gradation. The photo composite apparatus transmits the photo card to the printer to print the photo card. The photo synthesizer acquires the image from the camera and reproduces it on the display.

The photo synthesizing device captures the user's appearance and synthesizes and corrects the user with the previously produced object (character, background, etc.). To this end, the photo synthesizer acquires an image from a camera and outputs it to the screen.

6 is an exemplary view showing photographing through a camera connection.

In order to reproduce a photo object and a real-time camera image simultaneously, the photo synthesizer connects the photo object using a moving image. The photo synthesizer opens the selected file and overlays it on the camera image in an animation form until photographing, and displays it on the screen.

The photo synthesizer estimates the light source from the image obtained from the camera and corrects the light source. The photo synthesizer acquires an image from the camera and outputs it to the display.

When the camera is connected to the photo synthesizer, the camera plays the real-time image and finishes the photographing. The photo composition device connects the object video file to reproduce the object video. The photo composition device overlaps and photographs the camera image and the object image.

The photo synthesizer estimates the brightness of the light source on the image obtained from the camera and corrects the light source.

The image obtained through the camera can not provide satisfactory image quality due to color mismatch with the photo object to be synthesized due to the influence of the ambient light source. The photo synthesizer estimates and corrects the brightness of the light source in the image to reduce the influence of the ambient light source. The photo synthesizer estimates and compensates the light source brightness of the image through Retinex technology based on the color homeostasis which recognizes the unique color of the object by canceling the effect of the ambient light source by using light energy reflected from the object.

To this end, the photo synthesizer converts the RGB image into a YCbCr space, and calculates the brightness component of the light source by applying a Gaussian kernel of various sizes to the brightness channel. In order to compensate the brightness, the photo synthesizer compensates the brightness of the light source in the image by dividing the brightness of the calculated light source by the brightness channel of the image.

The photo synthesizer adjusts the color and contrast in each image to synthesize the image that has undergone the preprocessing, such as photo object and light source correction, and synthesizes it by similar conversion.

a) Contrast control

In order for the image to be well represented, the gradation of each channel must be adjusted to be evenly distributed. However, in the RGB color space used in general, the chromaticity of the image changes due to the correlation between the channels. Therefore, in order to reduce distortion of the color, the photo synthesizer separates brightness and chroma through color space conversion and extracts only brightness components.

After extracting only the brightness components, the photo synthesizer analyzes the distribution of each step of the gray level by increasing the gray level by comparing the histogram of the brightness. Thus, the photo-compositor reduces the over-enhancement defect that occurs globally while keeping the shape of the histogram similar.

b) Saturation control

It enhances the color through saturation control in the image with improved brightness in the photo synthesizer. The correction is performed by extracting only the saturation component according to the brightness using the color space conversion which can separate chromaticity and brightness components such as CIELAB and HSV as contrast adjustment. Also, the skin color is estimated locally and corrected to reproduce the color with the preferred skin color.

FIG. 7 is an exemplary view showing a process of reproducing a preferred skin color.

Therefore, a technology to reduce the heterogeneity of the skin color of the photo object and the skin color of the photographed image is developed. In order to reproduce the preferred color in the image, the preferred skin color is analyzed as shown in FIG. 7 and applied to the detected skin color region in the image to reproduce the preferred skin color.

First, the preferred skin color is defined in the chromaticity coordinate system through the preferred skin color experiment. Next, the skin color region is detected with respect to the photo object and the photographed image. Generally, a skin color region in an image is detected using a color space. The skin color exists within a certain range in various color spaces. For example, in the RGB color space, the value of the R channel has a smaller distribution than the value of the G channel or the B channel, and has a high correlation in the R channel than the G channel or the B channel . Therefore, the skin color region in the image is detected in consideration of the characteristics in the color space.

Next, the photo-synthesizing apparatus smoothly reproduces the skin color by proportionally calculating the difference between the input skin color and the preferred skin color with respect to the preferred skin color obtained through the preferred skin color experiment.

The photo synthesizer outputs a synthesized image through correction of skin color and gray gradation.

8 is an exemplary view showing an input image and a skin color correction image through reproduction of a preferred skin color.

It is the memory color of the person that has the most influence on the image quality evaluation of the image. Memory color plays an important role in improving user's preference and image quality. As shown in FIG. 8, the skin color of the image is reproduced in the preferred skin color, so that the user can obtain a preferred image.

The photo synthesizer may be implemented as an app, which is an application form of a smart phone. The application can consist of a part that composes a camera image and a photo object to create a composite image, and a part that prints a composite image on another medium. Other media may be various digital media, such as paper, image memory, and the like, in which composite images are output in the form of images.

110: image part 111: photographing part
112: selection unit 120: light source correction unit
130: fusion unit 140: filter processing unit
150: output unit 160: communication unit

Claims

A camera for photographing a moving picture; And
And performing a channel-by-channel image correction on the photo object and the moving image,
In the image correction,
Histogram matching, filtering, and color correction,
The histogram matching estimates minimum and maximum values for each channel, performs histogram stretching based on the estimated minimum and maximum values,
The filtering is performed by using a bilateral filter,
Wherein the color correction is color correction for the whole image, or the skin color correction is performed through face detection.

The method according to claim 1,
And a printer that prints the image generated by the composition on a ticket.

delete

The method according to claim 1,
Wherein the color correction is performed by extracting a saturated component using color space conversion in which brightness and chrominance are separated, such as CIELAB and HSV of an image, and correcting the saturated component by increasing the extracted saturated component.

The method according to claim 1,
The skin color correction is performed by detecting a skin color region with respect to the image, computing the brightness of the skin color region, calculating the difference between the input skin color and the preferred skin color for the preferred skin color obtained through the preferred skin color experiment, And correcting the skin color and the gray gradation.

The method according to claim 1,
The process comprises:
A photo-synthesis apparatus for estimating and correcting a component of a light source by summing weights of light sources of respective scales by applying a multi-scale retinex algorithm of an image.

The method according to claim 6,
And estimating and correcting the components of the light source to obtain a resultant image using the estimated light source component by performing multiple retinex.

The method according to claim 1,
The process comprises:
Separating brightness and chrominance through color space conversion, extracting and analyzing only brightness components, and enhancing color and contrast to synthesize the moving image and the photo object.

9. The method of claim 8,
Extracting only the brightness components, analyzing histograms of the brightness by histogram comparison, and analyzing the distribution of each gray level of each step, and correcting and combining the saturation correction and the skin color estimation with the preferred skin color.

10. The method of claim 9,
Wherein the histogram comparison is performed to maintain the shape of the histogram and to reduce the over enhancement defect that occurs when the histogram is performed globally.

10. The method of claim 9,
The saturation correction according to the brightness and the correction to the preferred skin color through the estimation of the skin color,
The image obtained by the camera is corrected by the local average image in the multiple retinex algorithm, and the effect of the light source is reduced. The resultant images of the Retinex algorithm using the Gaussian filters of various sizes and the reduction of the saturation occurring in the averaging process, And performing additional compensation for chroma coincidence with chroma saturation.