JP5683255B2 - Image processing apparatus and method - Google Patents

Description

  The present invention relates to edge correction, and more particularly, to an edge correction method when the non-linear characteristic of gamma correction of a luminance system is changed.

  In an imaging apparatus such as a digital video camera, an output signal from the imaging element is subjected to image processing and output to a monitor or the like. Since the monitor has a gamma characteristic, the image processing unit includes a gamma correction circuit for correcting the gamma characteristic of the monitor. When the inverse gamma characteristic of the monitor is corrected for the luminance signal, the luminance signal of the monitor output video with respect to the incident light changes approximately linearly. However, if correction is performed with the reverse gamma characteristic of the monitor in a state where exposure is appropriate, the dynamic range becomes low, and the high-luminance portion is blown out relatively quickly. There is no problem with a low-contrast subject, but when the contrast is high, the whiteout of the image in the high-intensity part increases. Therefore, conventionally, there has been a method of widening the dynamic range by detecting brightness information and detecting a whiteout region and controlling the aperture based on the information (see Patent Document 1).

JP 2010-183460 A

  However, in the method of widening the dynamic range based on the brightness information, in order to keep the brightness of the output image other than the high brightness part substantially constant before and after the dynamic range change, the brightness other than the high brightness part is used. When the luminance gamma characteristic for the signal is changed, there is a difference in the edge correction amount determined before correction of the gamma characteristic, and there is a difference in the resolution of the output image due to the influence of the luminance gamma characteristic. There was a problem that.

  The present invention is for solving the above-described problems, and makes the sense of resolution of an output image substantially constant.

In order to solve the above problems, an image processing apparatus of the present invention generates brightness information of an image from image data captured using an image sensor that converts an optical image formed by an optical system into an electrical signal. Generating means for determining the dynamic range from the brightness information generated by the generating means, edge correcting means for performing edge correction on the image data, and image data edge-corrected by the edge correcting means relative conversion means for performing nonlinear transformation, depending on the dynamic range determined by the determining means, changing means for changing the characteristics of the nonlinear conversion by prior Symbol conversion means, the dynamic range determined by the determining means in response, have a, and control means for controlling the edge correction amount by said edge correcting means, the control means, the luminance level of the image signal The edge correction amount is determined in proportion to the amplitude of the edge correction amount, and the edge correction amount has a lower limit value and an upper limit value. If the amplitude is equal to or less than a predetermined brightness level difference, the edge correction amount is the lower limit value, and the predetermined brightness level Above the amplitude of the difference, the amount of edge correction is the upper limit value, and between the amplitude value of the luminance level difference that is the starting point of each of the lower limit value and the upper limit value, edge correction is performed with a characteristic that linearly complements between the starting points. It is characterized by performing .

  As described above, according to the present invention, in addition to changing the dynamic range, the non-linear characteristic is changed when the non-linear characteristic for converting the luminance signal is changed in order to keep the brightness other than the high luminance part substantially constant. The previous edge correction amount can be made appropriate, and the sense of resolution in substantially the same brightness region can be made substantially constant.

1 is an example of a diagram schematically illustrating a configuration of a digital video camera according to an embodiment of the present invention. It is a figure which shows the example of the nonlinear characteristic set to the luminance system gamma correction part at the time of dynamic range change. An example which shows transition of the edge correction amount at the time of dynamic range change. It is a figure which shows the relationship between the expansion amount of a dynamic range, and edge correction amount. It is a figure which shows the relationship of the edge correction before and behind dynamic range expansion.

Embodiments of the image processing apparatus of the present invention will be described below with reference to the drawings.
FIG. 1 shows a schematic configuration diagram of a digital video camera according to the present embodiment. The illustrated digital video camera includes an optical system 101 including an iris and a lens, an imaging unit 102, a signal processing unit 103, a luminance system signal processing unit 104, an edge correction unit 105, a luminance system gamma correction unit 106, a detection unit 107, and a D range. A determination unit 108 and a system control unit 109 are included.

  The system control unit 109 controls the optical system 101, the imaging unit 102, the edge correction unit 105, the luminance system gamma correction unit 106, the detection unit 107, and the D range determination unit 108. Further, the system control unit 109 obtains the detection signal from the detection unit 107 and the D range information determined by the D range determination unit 108.

  The illustrated imaging unit 102 includes an imaging device such as a complementary metal oxide semiconductor (CMOS) image sensor or a CCD image sensor (not shown). The imaging unit 102 also includes a timing generator (TG).

  When a subject (not shown) is photographed, an optical image that has passed through the optical system 101 is formed on an image sensor in the imaging unit 102. Then, the image sensor photoelectrically converts the image formed by imaging to generate an electrical signal, and outputs it as an analog image signal. At this time, an analog image signal is output from the image sensor at a shutter speed determined by the system control unit 109. Further, an AFE (AnalogFrontEnd not shown) is provided in the imaging unit 102 after the imaging device, and the analog image signal of the imaging device is converted from analog to digital (A / D) to be converted into a digital image signal (hereinafter referred to as image data and image data). Called). This image data is given to the signal processing unit 103.

  The signal processing unit 103 performs black level correction and white balance adjustment. The image data that has passed through the signal processing unit 103 is sent to the luminance signal processing unit 104.

  The luminance system signal processing unit 104 performs generation of a luminance signal, low-pass filter processing, and the like, and sends them to the edge correction unit 105. FIG. 2 is a block diagram in the edge correction unit. Reference numeral 201 denotes an edge detection unit, and 202 denotes an edge correction amount calculation unit. The edge detection unit 201 detects the amplitude value and the spatial frequency of the luminance level difference between adjacent regions from the image data, and detects the region having the high luminance level difference and the spatial frequency as an edge portion. With respect to the detected edge portion, the edge correction amount calculation unit 202 determines to increase the edge correction amount in proportion to the amplitude value of the luminance level difference between adjacent regions. The edge correction amount has an upper limit value and a lower limit value. The upper limit value is set above the amplitude value of a certain luminance level difference, and the lower limit value is set below the amplitude value of a certain luminance level difference. Control is performed so that the amplitude value of the luminance level difference that is the starting point of the lower limit value and the upper limit value has a characteristic that linearly complements both the starting points. The edge correction amount determined as described above is added to the edge portion of the image data and output from the edge correction unit 105.

  The image data subjected to the edge correction is sent to the luminance system gamma correction unit 106. The luminance system gamma correction unit 106 performs nonlinear conversion on the luminance signal and outputs it. In this embodiment, 11 types of luminance system gamma characteristics are prepared in advance according to the dynamic range.

  Also, the luminance signal created by the luminance system signal processing is sent to the detection unit 107. Based on the frame and threshold set by the system control unit 109, the detection unit 107 acquires an evaluation value for exposure control and an evaluation value for a high luminance signal from the image data. The evaluation value for exposure control is a value representing the exposure of the current video, such as an average value of the entire image data of one frame. The evaluation value for exposure control is sent from the detection unit 107 to the system control unit 109. The system control unit 109 controls the iris, shutter, and gain so that the exposure control evaluation value becomes a target value. On the other hand, the evaluation value of the high luminance signal is a value representing how much the signal of the high luminance portion exists in the entire image data. For example, it is set as the signal amount of the luminance signal exceeding the threshold value. If the signal amount is large, it can be seen that there are many signals in the high luminance part. The evaluation value of the high luminance signal is also sent from the detection unit 107 to the system control unit 109 in the same manner as the exposure control evaluation value. The system control unit 109 sets the nonlinear characteristic of the luminance system gamma correction unit 106 based on the evaluation value of the high luminance signal.

  The evaluation value obtained by the detection unit 107 is sent to the D range determination unit 108. The D range determining unit 108 controls the D range according to the D range table set in the system control unit 109 based on the evaluation value of the high luminance signal. For example, evaluation values that exceed the threshold are ranked between three levels, and dynamic range settings of 100%, 200%, and 300% are prepared. If the rank is 1, the dynamic range is determined as 100%, 200 if it is 2, 300% if it is 3, and so on. Information on the determined dynamic range is sent to the system control unit 109. The system control unit 109 controls the iris, shutter, gain, and the like according to the dynamic range. The system control unit 109 changes the luminance system gamma correction unit 106 according to the dynamic range. The luminance system gamma characteristic to be changed has a luminance system gamma characteristic that allows the gradation to be expressed in the high luminance part in advance as the dynamic range increases, and that the output for the input signals other than the high luminance part is almost the same. Keep it. As an example, a case where the dynamic range is expanded from 100% to 200% will be described. 3A shows the luminance system gamma characteristics when the dynamic range is 100% and 200%, and FIG. 3B shows the input / output characteristics of the luminance system gamma characteristics when the dynamic range is 100% and 200%.

  301 is a luminance system gamma characteristic when the dynamic range is 100%, 302 is a luminance system gamma characteristic when the dynamic range is 200%, 303 is an input / output characteristic when the dynamic range is 100%, and 304 is an input / output characteristic when the dynamic range is 200%. .

  The luminance system gamma characteristic 302 when the dynamic range is 200% is lower than the luminance system gamma characteristic when the dynamic range is 100%, and is higher in the area other than the high luminance part. Therefore, the gradation of the high luminance part can be expressed from the dynamic range of 100%, and the brightness of the output image can be made substantially the same except for the high luminance part. For this reason, the input / output characteristics 303 when the dynamic range is 100% and the input / output characteristics 304 when the dynamic range is 200% are substantially the same in the region other than the high luminance portion.

  Further, the system control unit 109 changes the edge correction amount before the luminance system gamma characteristic from the dynamic range information determined by the D range determination unit 108. The exposure and luminance system gamma characteristics are changed before and after the dynamic range change. Therefore, when the edge correction amount before performing the luminance system gamma correction is not changed, the luminance level is changed by the luminance system gamma correction processing in the subsequent stage, so that a difference occurs in the resolution of the image data. Therefore, as shown in FIG. 4, when expanding the dynamic range, it is necessary to reduce the edge correction amount to a value obtained by dividing the edge correction amount by the dynamic range.

  In order to realize the edge correction processing as described above in a brightness region other than the high luminance portion, the following processing is performed in the present embodiment. Divide the amplitude value of the brightness level difference that is the starting point of the lower limit value of the edge correction amount, the amplitude value of the brightness level difference that is the starting point of the upper limit value of the edge correction amount, and the upper limit value of the edge correction amount by the dynamic range. Value. Control is performed so that the amplitude value of the luminance level difference that is the starting point of the lower limit value and the upper limit value has a characteristic that linearly complements both the starting points. Therefore, the edge correction characteristics after the dynamic range change are based on the amplitude value of the luminance level difference that becomes the starting point of the lower limit value, the amplitude value of the luminance level difference that becomes the starting point of the upper limit value, and the upper limit value of the edge correction amount. To be derived.

  As a specific example, FIG. 5 shows the relationship of edge correction before and after dynamic range expansion when the dynamic range is changed from 100% to 200%. Reference numeral 501 denotes an edge correction amount characteristic when the dynamic range is 200% before expansion of the dynamic range, and 502 is an edge correction amount characteristic when the dynamic range is 100% after expansion of the dynamic range. Reference numeral 503 denotes an upper limit value of the edge correction amount when the dynamic range is 100%. Reference numeral 504 denotes an upper limit value of the edge correction amount when the dynamic range is 200%. Reference numeral 505 denotes an amplitude value of a luminance level difference serving as a starting point of the lower limit value of the edge correction amount when the dynamic range is 200%. Reference numeral 506 denotes an amplitude value of a luminance level difference serving as a starting point of the lower limit value of the edge correction amount when the dynamic range is 100%. Reference numeral 507 denotes an amplitude value of a luminance level difference serving as a starting point of the upper limit value of the edge correction amount when the dynamic range is 200%. Reference numeral 508 denotes an amplitude value of a luminance level difference serving as a starting point of the lower limit value of the edge correction amount when the dynamic range is 100%.

  The amplitude value 505, the upper limit value 504, and the amplitude value 507 are changed to values multiplied by 100/200 in order to obtain values obtained by dividing each value at the dynamic range of 100% by the dynamic range to be changed. Further, between the upper limit value and the lower limit value of the edge correction amount, the edge correction at the time of dynamic range 200% before the dynamic range expansion of 501 is performed by changing the characteristic to the linear interpolation of the starting points of the upper limit value and the lower limit value. Get quantity characteristics.

  If the edge correction amount is set according to the relationship as shown in FIG. 5, the value obtained by dividing the edge correction amount before the luminance system gamma characteristic at the time of changing the dynamic range by the timely dynamic range can be obtained. Therefore, the resolution of the image data portion having substantially the same brightness after luminance gamma correction when changing the dynamic range can be made substantially constant.

<Other embodiments>
In the above embodiment, as a method of changing the nonlinear conversion according to the dynamic range, a means for storing a plurality of nonlinear change characteristics in advance is described. However, the characteristic of the nonlinear conversion is changed from the exposure information changed according to the dynamic range. May be.

  Also, as a means for determining the dynamic range according to the amount of high-luminance part signal, it is determined by whether there are many high-luminance part signals that are equal to or higher than the threshold in the image data. The dynamic range may be determined from the partial signal value.

DESCRIPTION OF SYMBOLS 101 Optical system 102 Imaging part 103 Signal processing part 104 Luminance system signal processing part 105 Edge correction part 106 Luminance system gamma correction part 107 Detection part 108 D range determination part 109 System control part 201 Edge detection part 202 Edge correction amount calculation part 301 Dynamic Luminance gamma characteristics at 100% range 302 Luminance gamma characteristics at 200% dynamic range 303 Input / output characteristics at 100% dynamic range 304 Input / output characteristics at 200% dynamic range 501 Edge correction characteristics at 200% dynamic range 502 Edge correction characteristic when dynamic range is 100% 503 Edge correction upper limit value when dynamic range is 100% 504 Edge correction upper limit value when dynamic range is 200% 505 Edge when dynamic range is 200% Starting point of the positive lower limit of the start point 506 edge correction upper limit value of the starting point 508 when the dynamic range of 100% of the edge correction upper limit value when the start point 507 dynamic range to 200% of the edge correction lower value when the dynamic range of 100%

Claims (6)

Generating means for generating brightness information of the image from image data captured using an image sensor that converts an optical image formed by the optical system into an electrical signal;
Determining means for determining a dynamic range from the brightness information generated by the generating means;
Edge correction means for performing edge correction on the image data;
Conversion means for performing nonlinear conversion on the image data edge-corrected by the edge correction means;
Depending on the dynamic range determined by the determining means, changing means for changing the characteristics of the nonlinear conversion by prior Symbol conversion means,
Depending on the dynamic range determined by the determining means, we have a, and control means for controlling the edge correction amount by said edge correcting means,
The control means determines an edge correction amount in proportion to the amplitude of the luminance level difference of the image signal, the edge correction amount has a lower limit value and an upper limit value, and the lower limit value is below a predetermined luminance level difference amplitude The upper edge correction amount is equal to or greater than the predetermined luminance level difference amplitude, and the luminance level difference amplitude value that is the starting point for each of the lower limit value and the upper limit value is the start point. An image processing apparatus characterized in that edge correction is performed with a characteristic in which a gap is linearly complemented . The control means, the value of the amplitude value obtained by dividing the value of the dynamic range of the luminance level difference that is the starting point of the lower limit of the edge correction amount, the luminance level difference that is the starting point of the upper limit of the edge correction amount and amplitude values, the edge correction amount of the upper limit, the image processing apparatus according to claim 1, characterized in that to change the value obtained by dividing the dynamic range, respectively. 3. The image processing according to claim 1, wherein the changing unit changes the characteristic of the non-linear transformation so that the output with respect to the input signal becomes substantially constant before changing the dynamic range except for the high luminance part. 4. apparatus. Said determining means according to the signal information on the high-luminance portion in the image data, the image processing apparatus according to any one of claims 1-3, characterized in that to determine the dynamic range. The image processing apparatus has the imaging element,
The image processing apparatus according to claim 1, wherein the changing unit changes exposure in imaging using the imaging device. A generation step of generating brightness information of the image from image data captured using an image sensor that converts an optical image formed by the optical system into an electrical signal;
A determination step of determining a dynamic range from the brightness information generated in the generation step;
An edge correction step for performing edge correction on the image data;
A conversion step for performing nonlinear conversion on the image data edge-corrected by the edge correction step;
Depending on the dynamic range determined by the determining step, a changing step of changing the characteristics of the nonlinear conversion by pre Symbol conversion step,
Depending on the dynamic range determined by the determining step, we have a, and a control step of controlling the edge correction amount by the edge correction step,
The control step determines an edge correction amount in proportion to the amplitude of the luminance level difference of the image signal, the edge correction amount has a lower limit value and an upper limit value, and the lower limit value is less than or equal to a predetermined luminance level difference amplitude The upper edge correction amount is equal to or greater than the predetermined luminance level difference amplitude, and the luminance level difference amplitude value that is the starting point for each of the lower limit value and the upper limit value is the start point. An image processing method characterized by performing edge correction with a characteristic in which a gap is linearly complemented .

Images