KR101593502B1 - Method for taking a photograph of face - Google Patents

Abstract

The present invention relates to a portrait shooting method, which divides an input image into a plurality of window areas, counts the number of high brightness areas that are window areas that are equal to or larger than the total brightness average value among the brightness average values of the plurality of window areas, It is possible to correct the distortion of the image data caused by the backlight by correcting the character area and the area outside the character area separately by judging that the backlight is generated in the image, The boundary between the peripheral areas of the person becomes clearer and the image quality of the image can be improved.

Backlight, Image, Pixel, Brightness, Saturation, Correction, Iris

Description

[0001] The present invention relates to a method for taking a portrait,

The present invention relates to a method of photographing a person using a camera module.

The camera's auto exposure control (Auto Exposure) or Auto Iris is a function for optimizing the light entering the CCD (Charge-coupled Devices).

That is, the exposure of the camera is controlled by setting an optimum iris value according to the amount of incident light.

In the automatic exposure control described above, blooming, which is a phenomenon in which a signal charge is overflowed when strong light is incident at a saturation light amount or more, and a surplus charge is mixed in surrounding pixels to brighten a portion where the light is not reflected, The present invention also provides a function of eliminating or preventing a phenomenon such as smear caused by incomplete shading of the vertical transfer portion of the CCD.

However, for a dark subject to be dark and a bright subject to be bright, the dynamic range of the CCD must be wide enough, but the saturation output of a high resolution CCD (more than 3 million pixels) 200mV ~ 500mV, and the SN ratio (signal-to-noise ratio) can not be increased. Therefore, it is impossible to completely cope with the above-mentioned blooming and smear.

Particularly, in the case where the face of the person is not darkened under the backlight and the picture is taken in a clear view, the picture is taken while looking at the light directly, the noise can not be seen even in the case where the automobile can be photographed even when the headlight is turned on at night.

In order to prevent the above-mentioned noise phenomenon, the conventional technique uses a screen average method and a screen division averaging method which assigns a weight to each screen.

The screen average method is a technique of controlling the iris by comparing the average value of the brightness levels of the whole screen or a part of the screen. The screen division averaging method is a technique of dividing a video signal into a plurality of screens, And detects the distribution and corrects it automatically.

However, the above method is not an effective solution against the noise, and particularly when applied to a camera for a mobile device such as a mobile phone or a notebook computer, , And ND (Neutral Density) filters are very difficult to apply.

In particular, when the image data is distorted as in the backlight image, it is necessary to correct the image data to improve the image quality.

The present invention provides a portrait photographing method for correcting a portrait area of an image in which backlight is generated.

According to an aspect of the present invention, there is provided a method of controlling an image processing apparatus, the method comprising: dividing an input image into a plurality of window regions; calculating a total luminance average value of the plurality of window regions; The method comprising the steps of: counting the number of high luminance regions which are window regions that are equal to or larger than the total luminance average value among the average luminance values of the respective window regions; if the number of high luminance regions is greater than or equal to the reference number, Detecting, as a face region, pixels included in a reference saturation range in which a saturation is set among a plurality of pixels included in the intermediate luminance region, Detecting, as an iris region, pixels less than two reference luminance values, Detecting a plurality of edges by using a luminance difference between two adjacent pixels included in the intermediate luminance region when the iris region is detected in a middle luminance region of the intermediate luminance region, A step of detecting a character area having an edge as a boundary line, and a step of setting a luminance of a pixel included in the character area by raising the luminance by a predetermined level.
The method may further include converting R, G, and B data of the image into a luminance signal (Y) and a chroma signal (Cb, Cr) 180 < / RTI > range.
Also, the second reference luminance value may be 10 of the luminance signal (Y).

In the embodiment of the present invention, distortion of the image data caused by backlight can be corrected by separately correcting the person area and the person peripheral area in the backlit image, and the boundary between the person area and the person peripheral area becomes clearer, Can be improved.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the invention is not intended to be limited to the particular embodiments, but includes all modifications, equivalents, and alternatives falling within the spirit and scope of the invention.

Terms including ordinals, such as first, second, etc., may be used to describe various elements, but the elements are not limited to these terms. The terms are used only for the purpose of distinguishing one component from another.

For example, without departing from the scope of the present invention, the second component may be referred to as a first component, and similarly, the first component may also be referred to as a second component.

It is to be understood that when an element is referred to as being "connected" or "connected" to another element, it may be directly connected or connected to the other element, . On the other hand, when an element is referred to as being "directly connected" or "directly connected" to another element, it should be understood that there are no other elements in between.

The terminology used in this application is used only to describe a specific embodiment and is not intended to limit the invention. The singular expressions include plural expressions unless the context clearly dictates otherwise. In the present application, the terms "comprises", "having", and the like are used to specify that a feature, a number, a step, an operation, an element, a component, But do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, or combinations thereof.

Hereinafter, a person shooting method according to an embodiment of the present invention will be described in detail with reference to the drawings, and the same or corresponding elements are denoted by the same reference numerals regardless of their reference numerals, and redundant explanations thereof are omitted.

1 is a flowchart showing a person photographing method according to an embodiment of the present invention.

1 is a block diagram illustrating a configuration of an apparatus including a processor, and FIG. 1 is a block diagram illustrating a configuration of an image processing apparatus according to an embodiment of the present invention. And the like. In particular, the present invention is applicable to a camera module built in a mobile communication terminal.

As shown in FIG. 1, the present invention includes a step of inputting an acquired image (S100), and a step of determining whether or not the input image is a backlit image (S200). If it is determined that the backlight image is not a backlight image, a step of detecting and correcting an area other than a person area and a character area in the image (S300) is performed. If it is determined that the backlight image is not a backlight image, a step S400 of outputting the image without correction to a display unit .

FIG. 2 is a flowchart illustrating a backlight image determination step (S100) according to an embodiment of the present invention. FIG. 3 is a diagram illustrating a method of dividing an image according to an embodiment of the present invention.

As shown in Figs. 2 and 3, the acquired image is first divided into a plurality of window regions W1 to Wn (S101). In the present invention, the image data is divided into 16 window regions for convenience of explanation. One window region includes a plurality of pixels.

The total luminance average value Yo of the entire image is calculated (S103), and the divided luminance values of the plurality of window regions (Cb, Cr) are calculated The luminance average values Y1 to Yn of each of the luminance signals W1 to Wn are calculated (S104).

A window area having a total luminance average value Yo or more among a plurality of luminance average values Y1 to Yn of a plurality of window areas is compared with a luminance range average value Y1 to Yn and a total luminance average value Yo, (S106), and detects a window area having a brightness average value (Y1 to Yn) less than the total average value (Yo) as a low brightness area (S107).

Then, the number of high brightness areas is counted (S108), and the number of high brightness areas is compared with the set reference number (S109).

If the number of high luminance regions is equal to or more than the reference number, it is determined that backlight has occurred in the image (S110), and the step of detecting the character region (S200) is performed.

However, if the number of high brightness regions is less than the reference number, it is determined that backlight does not occur in the image (S111), and the image is output to the display unit (not shown) (S112).

4 is a flowchart illustrating a step S200 of detecting a person area according to an embodiment of the present invention.

The luminance average values Y1 to Yn of the plurality of window regions are compared with the first reference luminance range that is set (S201). When the luminance average value ( Y1 to Yn) are included in the first reference luminance range as an intermediate luminance region (S202).

Then, the saturation of the plurality of pixels included in the intermediate luminance region is compared with the set reference saturation range (S203), and the pixels having the saturation included in the reference saturation range are detected as the face region (S204). If there is no pixel whose saturation is within the reference saturation range, it is determined that it is not a portrait, and the image is output to the display unit (S205).

In this case, the reference saturation range is a saturation range indicating flesh color, and the saturation signal (Cb) is included within the range of 69 to 114, and the saturation signal (Cr) is within the range of 119 to 177.

Next, the luminance of the pixels included in the face region is compared with the set second reference luminance value (S206), and the pixels less than the second reference luminance value are detected as the iris region (S207). At this time, if there is no pixel less than the second reference luminance value, it is determined that it is not a portrait, and the image is output to the display unit (S208).

The second reference luminance value is a reference value of a luminance value indicating a black color, and the luminance signal (Y) is a value of 10.

More specifically, in the step of detecting the iris region, a luminance histogram of pixels included in the face region is modeled as a mixture model of N normal distribution functions, and pixels having luminance lower than the second reference luminance value are extracted, .

When the face region is detected and the iris region is detected in the face region, it means that the image contains a person.

Accordingly, a plurality of edges are detected using the luminance difference between two adjacent pixels included in the intermediate luminance region (S209), and a character region having the edge nearest to the edge of the intermediate luminance region among the detected plurality of edges is detected (S210).

At this time, a plurality of edges can be detected using a Haar wavelet transform technique.

Finally, the luminance of the pixels included in the character area is increased by a predetermined luminance level (S211), and the luminance of the pixels included in the area excluding the character area in the intermediate luminance area is decreased by the set luminance level (S212).

At this time, the luminance level for increasing the luminance of the pixel and the luminance level for decreasing the luminance of the pixel are values that can be arbitrarily adjusted by the user, such that the boundary between the character area and the out-of-character area is sharpened.

Then, the backlight generated in the person area can be corrected. Further, distortion of the image data caused by backlight can be corrected by separately correcting the person area and the person peripheral area in the backlit image, and the boundary between the character area and the person peripheral area becomes clearer, can do.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, It belongs to the scope of right.

1 is a flowchart showing a person photographing method according to an embodiment of the present invention.

2 is a flowchart illustrating a backlight image determination step according to an embodiment of the present invention.

3 is a view for explaining a method of dividing an image according to an embodiment of the present invention.

4 is a flowchart illustrating a step of detecting a person area according to an embodiment of the present invention.

Claims (7)

Dividing an input image into a plurality of window regions, Calculating a total luminance average value of the plurality of window regions and calculating a luminance average value for each of the plurality of window regions, Counting the number of high luminance regions which are window regions that are equal to or greater than the total luminance average value among the luminance average values of the plurality of window regions, Detecting, as an intermediate luminance region, the window region included in the first reference luminance range in which the luminance average value of each region is set, if the number of the high luminance regions is equal to or greater than the reference number; Detecting pixels of a plurality of pixels included in the intermediate luminance region included in a reference saturation range in which the saturation is set as a face region, Detecting, as an iris region, pixels having a luminance lower than a set second reference luminance value among a plurality of pixels included in the face region,  Detecting a plurality of edges using the luminance difference between two adjacent pixels included in the intermediate luminance region when the iris region is detected in the intermediate luminance region, Detecting a character area having a border nearest to an edge of the intermediate luminance area among the plurality of edges;  And setting the brightness of the pixels included in the person area to be higher by a predetermined level. The method according to claim 1, Further comprising the step of setting the luminance of the pixels of the intermediate luminance region other than the character region by a predetermined level. The method according to claim 1, Wherein detecting the edge comprises detecting an edge using a Haar wavelet transform technique. The method according to claim 1, Wherein the detecting the iris region comprises: modeling a luminance histogram of pixels included in the face region as a mixture model of N normal distribution functions; extracting pixels of the mixture model whose luminance is less than the second reference luminance value, Area detecting method. The method according to claim 1, Converting the R, G and B data of the image into a luminance signal (Y) and a chroma signal (Cb, Cr) Wherein the first reference luminance range is included within a range of 140 to 180 of the luminance signal (Y). 6. The method of claim 5, Wherein the reference saturation range is included within a range of 69 to 114 of the saturation signal (Cb) and is included within a range of 119 to 177 of the saturation signal (Cr). 6. The method of claim 5, And the second reference luminance value is the luminance signal (Y).

Images