KR101417818B1 - Apparatus and method for recognizing scene mode in digital image processing device - Google Patents

Abstract

The present invention relates to a digital image processing apparatus and method, and more particularly, to a scene recognition apparatus and method in a digital image processor capable of quickly recognizing a scene of a live view image using a face recognition function. In the digital image processor, the scene recognizing device detects a face from a live view image. When a face is detected, the live view image is classified into a night view mode, a backlight mode, and a portrait mode. Landscape mode, macro mode, and automatic exposure mode in that order.

A face recognition, a first scene recognition unit (night view mode, backlight mode, portrait mode), a second scene recognition unit (night view mode, landscape mode, macro mode,

Description

[0001] Apparatus and method for recognizing a scene in a digital image processor [0002]

The present invention relates to a digital image processing apparatus and method, and more particularly, to a scene recognition apparatus and method in a digital image processor capable of quickly recognizing a scene of a live view image using a face recognition function.

The current digital image processor sets the scene mode for the user who does not handle it well and sets the optimal setting condition for the scene. Therefore, it is self-evident that many scene modes are required to obtain a better image.

However, as more scenes can be selected, ordinary users who have little knowledge about the digital image processor are rather inconvenient to choose, and in the automatic setting mode, the digital image processor recognizes the optimal scene mode You want a digital image processor.

In order to recognize various scenes such as a night view, a night view person, a portrait, a back light person, a landscape, and a macro, basically a quick execution time is required. However, if the scene is judged sequentially every time for each live view image, fast recognition speed can not be expected, and a smart function may be felt as inconvenience to the user.

An aspect of the present invention is to provide a scene recognition apparatus and method in a digital image processor capable of quickly recognizing a scene of a live view image using a face recognition function.

There is provided a digital image processor for detecting a face from a live view image and detecting the face when the face is detected, And a digital signal processing means for classifying the live view image into a night view mode, a landscape mode, a macro mode and an automatic exposure mode in the case where the face is not detected, A face recognizing unit for recognizing the face from a view image; A first scene recognition unit for classifying the live view image into a night view mode, a backlight mode, and a portrait mode in a case where the face is detected from the face recognition unit; And a second scene recognition unit for classifying the live view image into a night view mode, a landscape mode, a macro mode, and an automatic exposure mode when the face is not detected from the face recognition unit.
According to another aspect of the present invention, there is provided a method of recognizing a scene in a digital image processor including the steps of: (a) detecting a face from a live view image; (b) classifying the live view image in the order of a night view mode, a backlight mode, and a portrait mode when the face is detected; And c) classifying the live view image into a night view mode, a landscape mode, a macro mode, and an automatic exposure mode if the face is not detected.
Wherein the step (b) includes the steps of: (b-1) classifying the live view image into the night view mode when the live view image is similar to previously stored night view images; (b-2) classifying the live view image into the backlight mode if the live view image is not similar to the stored night view images; And (b-3) classifying the live view image into the person mode if the exposure value of the live view image is within the first exposure value range.
The step (c) includes the steps of: (c-1) classifying the live view image into the night view mode if the live view image is similar to previously stored night view images; (c-2) classifying the live view image into the landscape mode if the live view image is not similar to the stored night view images; (c-3) classifying the live view image into the macro mode when the focus peak value of the live view image is within a macro area; And (c-4) classifying the live view image into the automatic exposure mode if the focus peak value of the live view image is outside the macro region.

delete

delete

delete

delete

As described above, according to the present invention, a scene of a live view image can be quickly recognized using a face recognition function.

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

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a perspective view showing the front and top contours of a digital image processor.

The shutter-release button 11 is opened and closed to expose the CCD or film to light for a predetermined time, and the image is recorded on the CCD by appropriately exposing the subject in cooperation with the aperture (not shown).

The shutter-release button 11 generates the first and second image pickup signals by the photographer input. When the first shutter-release button 11 is input as a half-shutter signal, the digital image processing apparatus focuses and adjusts the amount of light. At this time, the display unit 25 is lit green. When the focus of the first shutter-release button 11 is focused and the amount of light is adjusted, the second shutter-release button 11 is input as a full shutter signal to photograph the image.

The power button 13 is inputted to supply power to the digital image processing apparatus and operate it.

The flash 15 illuminates bright light momentarily when shooting in a dark place. The flash mode includes auto flash, forced flash, no flash, red-eye reduction, and slow synchro flash.

The auxiliary light 17 supplies light to the subject so that the digital image processing apparatus automatically catches the focus quickly and accurately when the light amount is insufficient or at night.

The lens unit 19 receives light from an external light source and processes the image.

FIG. 2 is a rear view showing the rear external shape of the digital image processor shown in FIG. 1, and includes a wide angle-zoom button 21w, a telephoto-zoom button 21t, a display unit 23 and a touch sensor or a contact switch Input buttons B1 to B14 (hereinafter referred to as buttons B1 to B14).

The wide-angle zoom button 21w or the telephoto-zoom button 21t widens the view angle or narrows the view angle depending on the input, in particular, when the size of the selected exposure area is to be changed. When the wide angle-zoom button 21w is input, the size of the selected exposure area is reduced, and when the telephoto-zoom button 21t is input, the size of the selected exposure area is increased.

The buttons B1 to B14 are provided in the rows and columns of the display unit 23. The buttons B1 to B14 provided in the row and column of the display unit 23 are provided with a touch sensor (not shown) or a contact type switch (not shown).

That is, the touch sensors are provided on the buttons B1 to B14 to move up / down / left / right while touching the buttons B1 to B7 of the horizontal row or the buttons B8 to B14 of the vertical row, It is possible to select any of the items (for example, color or brightness) or activate the submenu icon included in the main menu icon.

The buttons B1 to B14 are provided with contact switches so that the main menu icon and the sub menu icon can be directly selected to execute the corresponding function. The touch sensor requires a relatively weak touch compared to the contact switch input, but the contact switch input requires a relatively strong touch compared to the touch sensor input.

FIG. 3 is a block diagram showing a configuration of a scene recognition apparatus in a digital image processor according to the present invention, which includes a display unit 23, a user input unit 31, an image pickup unit 33, an image processing unit 35, 37 and a digital signal processing unit 39. [

The user input unit 31 includes a shutter-release button 11 that is opened and closed to expose a CCD or a film to light for a predetermined period of time, a power button 13 that inputs power to supply power, A wide angle-zoom button 21w and a telephoto-zoom button 21t for narrowing the angle of view and a touch sensor or a contact-type switch provided in the horizontal and vertical columns around the display unit 23 for character input or menu selection and execution, And buttons B1 to B14 provided with buttons.

The imaging section 33 includes a shutter, a lens section, a diaphragm, and a charge coupled device (CCD) and an ADC (not shown). The shutter is a mechanism that adjusts the amount of light that is exposed with the iris. The lens unit receives light from an external light source and processes the image. At this time, the diaphragm adjusts the amount (light amount) of the incident light according to the degree of opening and closing. The opening / closing degree of the diaphragm is controlled by the digital signal processing section 39. [

The CCD accumulates the amount of light input through the lens unit and outputs the image captured by the lens unit according to the accumulated light amount in accordance with the vertical synchronization signal. Image acquisition of a digital image processing apparatus is performed by a CCD which converts light reflected from a subject into an electrical signal. In order to obtain a color image using a CCD, a color filter is required, and a filter (not shown) called a color filter array (CFA) is employed in most cases. CFA has a regularly arranged structure that passes only light that represents one color per pixel, and it has various forms according to the arrangement structure. The ADC converts the analog video signal output from the CCD into a digital signal.

The image processing unit 35 processes the digitally converted RAW data so that it can be displayed. The image processing unit 35 removes a black level due to the dark current generated in the CCD and the CFA filter which are full of temperature change. The image processing unit 35 performs gamma correction to encode information in accordance with the nonlinearity of human vision. The image processing unit 35 performs CFA interpolation for interpolating a Bayer pattern implemented with RGA and RGB lines of gamma-corrected predetermined data into RGB lines. The image processing unit 35 converts the interpolated RGB signals into YUV signals, performs edge compensation for filtering the Y signals by a high-pass filter, and calculates color values of the U and V signals using a standard color coordinate system Corrects the color correction, and removes these noises. The image processing unit 35 compresses and processes the Y, U, and V signals from which the noise has been removed to generate a JPEG file, and the generated JPEG file is displayed on the display unit 23 and stored in the storage unit 37 . All the operations of the image processing unit 35 operate under the control of the digital signal processing unit 39. [

When the face is detected, the digital signal processing unit 39 classifies the live view image into the night view mode, the back light mode, and the portrait mode in that order. When the face is not detected, Mode, landscape mode, macro mode, and automatic exposure mode.

To this end, the digital signal processing unit 39 includes a face recognition unit 39-1, a first scene recognition unit 39-2, a second scene recognition unit 39-3, and a control unit 39-4.

In the present invention, scenes that can be obtained for efficient scene recognition are classified into seven types, which are night view portrait mode, night view mode, backlight portrait mode, portrait mode, landscape mode, macro mode, and default (auto exposure) mode.

In the present invention, when a specific scene is recognized in the default mode, the parameter values such as the exposure time, ISO, and aperture at that time are set to the values of the automatic exposure algorithm for the corresponding tuned mode. This enables us to implement the expert knowledge based scene recognition technology of image quality tuning.

The face recognition unit 39-1 detects at least one face information, that is, the face size, position and direction, from an arbitrary live view image under the control of the control unit 39-4. The face recognition unit 39-1 can detect the face area based on the color or detect the face area based on the edge at the time of detecting the face size and when detecting the face position, And the center of the display unit 23, and the face direction can be detected in the form of a triangle composed of the eyes and mouth from the extracted face at the time of detecting the direction of the face. Since details of the face information detection by the face recognition unit 39-1 are well known in the art, a detailed description thereof will be omitted.

Since most of the character images are captured when the video is shot, the face detection information is used first as the scene recognition information. When there is a person as shown in Figs. 4A and 4C, the night view mode, the back light mode, and the portrait mode are categorized in this order, and if there is no character as shown in Figs. 4B and 4C, , Landscape mode, macro mode, and automatic exposure mode.

If the face detection is modularized by software, the part that takes the longest operation time except the macro mode is the face detection part. Therefore, scene classification using face detection information is set as the upper classification condition.

When the face is detected from the live view image based on the face detection information received by the control unit 39-4, the first scene recognition unit 39-2 selects the live view image as the night view mode, backlight mode, .

The first scene recognition unit 39-2 compares a number of night view images stored in a database (not shown) with a live view image at the time of classifying the night view mode, and a similar live view image corresponding to the night view mode exists in the database , Classifies the live view image into the night view mode.

The first scene recognizing unit 39-2 recognizes the difference between the brightness of the face detected from the live view image and the brightness of the background portion excluding the detected face in a state in which it is determined that the live view image is not the night view mode And if the background portion is lighter than the detected face, it is classified into the back light mode.

The first scene recognizing unit 39-2 classifies the person view mode into the portrait mode when the live view image is not the night view mode and the backlight mode is not.

When the face is detected, the detected face is subjected to auto focus and automatic exposure, and the night view or back light has a feature that the brightness of the face is dark. Therefore, if the exposure time that can express the automatic exposure information is used, it is possible to detect the person without sequentially recognizing the night view or backlight.

For example, if it takes 200 msec for face detection and 50 msec for night view or backlight, respectively, the scene recognition time can be within 250 msec under person detection.

If the face is detected from the live view image based on the face detection information received by the control unit 39-4, the second scene recognizing unit 39-3 recognizes the live view image as a night view mode, a landscape mode, a macro mode, Automatic exposure mode.

The second scene recognition unit 39-3 compares a number of night view images stored in a database (not shown) with a live view image at the time of classifying the night view mode, and a similar live view image corresponding to the night view mode exists in the database , Classifies the live view image into the night view mode.

The second scene recognizing unit 39-3 determines that the live view image is not in the night view mode at the time of the landscape mode classification and if the total brightness of the live view image is equal to or larger than a predetermined value (for example, 12 LV (light value) If it is, it is classified into landscape mode.

When the focus peak value is located within the macro area while the live view image is not in the night view mode and the landscape mode is not in the macro area, the second scene recognition unit 39-3 sets the live view image to the macro mode Classify.

The macro mode recognition is classified using whether the focus peak value is within the macro area using the detection data obtained by moving the focus motor. This is because the operation time is long due to the restriction of moving according to the VD synchronization, and there is a problem that the live view image is languid because it uses the hill climb method. In addition, when recognizing another scene after recognizing the macro mode, the sharpness of the image displayed on the live view image should be increased by moving the focus motor to the focal distance of the scene.

Since the operation is the longest and there is a change of the live view image due to the movement of the focus motor, the recognition is performed last and the efficiency is improved. In addition, since both the night view mode and the landscape mode are mainly located at the far portion, it is possible to minimize the movement of the focus motor when the first shutter-release button 11 is input after recognizing the scene.

The second scene recognition unit 39-3 classifies the automatic exposure mode into the automatic exposure mode when the live view image is not a night view mode, nor is it a landscape mode or a macro mode. In Auto Exposure mode, Auto Exposure allows you to adjust the optimum brightness for recognized scenes.

Hereinafter, a scene recognition method in a digital image processor according to the present invention will be described in detail with reference to FIG. The scene recognition method in the digital image processor according to the present invention can be performed in the digital image processor as shown in FIG. 3. According to the embodiment, the main algorithm of the operation method is digital And may be performed within the signal processing unit 39.

When the live view image is displayed on the display unit (step 501), the digital signal processing unit 39 detects face presence / absence through face recognition of the live view image (steps 503 and 505). The digital signal processing unit 39 detects at least one face information, that is, the size, position, and direction of the face from the live view image. The details of the face detection are described above and will be omitted here.

If the facial recognition result face of the digital signal processing unit 39 is detected, it is determined whether the live view image is an appropriate exposure (step 507). If it is not the proper exposure, it is determined whether it is a night view (step 509) (Step 511). The digital signal processing unit 39 compares a number of night view images stored in a database (not shown) with a live view image at the time of classifying the night view mode with respect to the live view image, not the proper exposure, If there is a live view image, classify the live view image into a night view mode.

If it is determined that the live view image is not the night view mode in a state in which the live view image is not properly exposed, the digital signal processor 39 classifies the live view image into a backlight mode (step 513). The digital signal processing unit 39 compares the brightness of the face detected from the live view image and the brightness of the background portion excluding the detected face to determine whether the background portion is brighter than the detected face, Classify.

When the face is detected, the detected face is subjected to auto focus and automatic exposure, and the night view or back light has a feature that the brightness of the face is dark. Therefore, if the exposure time that can express the automatic exposure information is used, it is possible to detect the person without sequentially recognizing the night view or backlight.

For example, if it takes 200 msec for face detection and 50 msec for night view or backlight, respectively, the scene recognition time becomes less than 250 msec under portrait detection.

If the live view image is in the proper exposure state, the digital signal processor 39 classifies the live view image into the portrait mode (step 515). The digital signal processing unit 39 classifies the person view mode into the portrait mode when the live view image is not the night view mode and the back view mode is not appropriate.

If the face recognition result face of the digital signal processing unit 39 is not detected, it is determined whether the live view image is a night view (Step 517), and the night view mode is classified (Step 519). The digital signal processing unit 39 compares a number of night view images stored in a database (not shown) with a live view image at the time of classifying the night view mode, and when a similar live view image corresponding to the night view mode exists in the database, The view image is classified into the night view mode.

If the digital signal processing unit 39 determines that the live view image is not the night view mode and the total brightness of the live view image is equal to or greater than 12 LV (step 521), the digital signal processing unit 39 classifies the live view image into a landscape mode (Step 523).

The digital signal processing unit 39 classifies the live view image into a macro mode when the total brightness of the live view image is 12 LV (light value) or less and the focus peak value is located within the macro area (step 525) ). The macro mode recognition is classified using whether the focus peak value is within the macro area using the detection data obtained by moving the focus motor. This is because the operation time is long due to the restriction of moving according to the VD synchronization, and there is a problem that the live view image is languid because it uses the hill climb method. In addition, when recognizing another scene after recognizing the macro mode, the sharpness of the image displayed on the live view image should be increased by moving the focus motor to the focal distance of the scene. Since the operation is the longest and there is a change of the live view image due to the movement of the focus motor, the recognition is performed last and the efficiency is improved. In addition, since both the night view mode and the landscape mode are mainly located at the far portion, it is possible to minimize the movement of the focus motor when the first shutter-release button 11 is input after recognizing the scene.

If the focus peak value of the live view image is not located within the macro area, the digital signal processor 39 classifies the live view image into an automatic exposure mode (step 529). The digital signal processing unit 39 classifies the live view image into an automatic exposure mode when the live view image is not a night view mode, nor is it a landscape mode or a macro mode. In Auto Exposure mode, Auto Exposure allows you to adjust the optimum brightness for recognized scenes.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a perspective view showing the front and top contours of a digital image processor.

FIG. 2 is a rear view showing the rear external shape of the digital image processor shown in FIG. 1; FIG.

3 is a block diagram showing a configuration of a scene recognition apparatus in a digital image processor according to the present invention.

4 is a view for explaining scene recognition in the apparatus of FIG.

5 is a flowchart illustrating an operation of a scene recognition method in a digital image processor according to the present invention.

Claims (5)

In a digital image processor, as a scene recognition device, The live view image is classified into a night view mode, a backlight mode, and a portrait mode in the case where the face is detected, and when the face is not detected, the live view image is divided into a night view mode, , A macro mode, and an automatic exposure mode, The digital signal processing means A face recognition unit for recognizing the face from the live view image; A first scene recognition unit for classifying the live view image into a night view mode, a backlight mode, and a portrait mode in a case where the face is detected from the face recognition unit; And And a second scene recognition unit for classifying the live view image into a night view mode, a landscape mode, a macro mode, and an automatic exposure mode when the face is not detected from the face recognition unit. delete A scene recognition method in a digital image processor, (a) detecting a face from a live view image; (b) classifying the live view image in the order of a night view mode, a backlight mode, and a portrait mode when the face is detected; And (c) if the face is not detected, classifying the live view image into a night view mode, a landscape mode, a macro mode, and an automatic exposure mode. 4. The method of claim 3, wherein step (b) (b-1) classifying the live view image into the night view mode if the live view image is similar to previously stored night view images; (b-2) classifying the live view image into the backlight mode if the live view image is not similar to the stored night view images; And (b-3) classifying the live view image into the person mode when the exposure value of the live view image is within the first exposure value range. 4. The method of claim 3, wherein step (c) (c-1) classifying the live view image into the night view mode if the live view image is similar to previously stored night view images; (c-2) classifying the live view image into the landscape mode if the live view image is not similar to the stored night view images; (c-3) classifying the live view image into the macro mode when the focus peak value of the live view image is within a macro area; And (c-4) classifying the live view image into the automatic exposure mode if the focus peak value of the live view image is outside the macro area.

Images