JP5451495B2 - Image processing apparatus and image processing method - Google Patents

Description

  The present invention relates to an image processing apparatus and an image processing method.

There is a technique for suitably controlling the contrast of a display image by setting a γ curve (image processing parameter) for each frame based on a luminance feature amount such as an APL (average luminance level) or luminance histogram of the frame. Hereinafter, such gradation correction processing is referred to as dynamic γ processing. Such a technique is disclosed in Patent Document 1, for example. Normally, in dynamic γ processing, when the brightness of the input video signal changes abruptly, the γ curve to be set changes abruptly, and a constant time constant is given so as not to cause a feeling of interference such as flickering. The γ curve is changed slowly. Hereinafter, such processing is referred to as time direction LPF processing.
As an example of temporal direction LPF processing, the degree of scene change between temporally continuous frames is detected, weighting according to the detected degree of change is used as a luminance feature amount, and an image is obtained using the weighted luminance feature amount. A technique for obtaining a processing parameter is disclosed in Patent Document 2.

Japanese Unexamined Patent Publication No. 03-126377 JP 2004-282377 A

  However, when dynamic γ processing using time-direction LPF processing is performed during the fade scene period, application of the γ curve may be delayed, and blackout or overexposure may occur. Further, if the time-direction LPF processing is not performed, the detected change in the luminance feature amount for each frame is directly applied to the γ curve, so that a sense of interference such as flicker may occur. A fade scene is a scene in which the brightness of a frame monotonously increases or decreases, for example, a scene where a certain image is gradually switched to another image.

  An object of the present invention is to provide a technique capable of realizing a gradation correction process in which a delay in application of gradation correction parameters and a feeling of interference such as flickering are suppressed during a fade scene period.

  An image processing apparatus of the present invention is an image processing apparatus that dynamically switches image processing parameters used for gradation correction based on the luminance of an input video signal, and represents the luminance of the frame for each frame of the input video signal. A luminance feature quantity detecting means for detecting a feature quantity; a fade scene detecting means for detecting a fade scene period in which the brightness of the frame monotonously increases or decreases from the luminance feature quantities of a plurality of frames; An image processing parameter to be applied to the first frame is determined from the luminance feature amount of each frame, and an image processing parameter to be applied to a plurality of frames between the first and last frames is applied to the first frame Decide to gradually change from processing parameters to image processing parameters to be applied to the last frame With a constant section, and a correction means for gradation correction frames within a period of fade scenes by using the image processing parameter determined by said determining means.

An image processing method of the present invention is an image processing method executed by an image processing apparatus that dynamically switches image processing parameters used for gradation correction based on the luminance of an input video signal, and for each frame of the input video signal, A luminance feature amount detecting step for detecting a luminance feature amount representing the luminance of the frame; a fade scene detecting step for detecting a fade scene period in which the luminance of the frame monotonously increases or decreases from the luminance feature amounts of a plurality of frames; The image processing parameters to be applied to the first and last frames of the fade scene period are determined from the luminance feature values of the respective frames, and the image processing parameters to be applied to a plurality of frames between the first and last frames are determined. The image processing parameters applied to the last frame from the image processing parameters applied to the first frame. It has a determining step of determining to gradually change to chromatography data, and a correction step of tone correction frames within a period of fade scenes by using the image processing parameter determined by the determining step.

  According to the present invention, it is possible to realize a gradation correction process in which a delay in application of a gradation correction parameter and an interference feeling such as flickering are suppressed during a fade scene period.

1 is an example of a functional configuration of an image processing apparatus according to a first embodiment. An example of the relationship between time and APL. An example of the processing flow at the time of detecting the period of a fade scene. An example of APL of each frame. An example of the output of a fade scene detection part. The figure which shows an example of (gamma) curve. An example of the relationship between APL and a blend rate. An example of the relationship between the frame number and the blend ratio within the fade scene period. FIG. 10 illustrates an example of a functional configuration of an image processing apparatus according to a second embodiment. An example of the processing flow at the time of detecting the period of a fade scene.

<Example 1>
Hereinafter, an image processing apparatus according to Embodiment 1 of the present invention and an image processing method executed by the image processing apparatus will be described with reference to the drawings. The image processing apparatus according to the present embodiment dynamically switches image processing parameters used for gradation correction based on the luminance of an input video signal (input video signal).
FIG. 1 is a functional block diagram of the image processing apparatus according to the present embodiment. As shown in FIG. 1, the image processing apparatus 100 includes an input signal terminal 101, a video delay unit 102, an APL detection unit 103, an APL storage unit 104, a fade scene detection unit 105, and the like. Further, it includes a γ curve generation unit 106, a fade scene γ curve generation unit 107, a γ curve selection unit 108, a γ conversion unit 109, an output signal terminal 110, and the like.

An input video signal is input to the input signal terminal 101. Specifically, the input video signal is input in units of frames.
The video delay unit 102 stores a predetermined number of frames for detection of a fade scene period in which the luminance of the frame monotonously increases or decreases, and delays the input video signal by the predetermined number of frames to delay the input video signal. Output to 109. Specifically, the image processing apparatus 100 (video delay unit 102) has a buffer memory (not shown) capable of storing a predetermined number of frames, and the video delay unit 102 records the frames in the buffer memory. . In this embodiment, the delay amount is 300 frames (that is, 300 frames are stored). The amount of delay is not limited to this. It is only necessary to store frames for the fade scene period.

The APL detection unit 103 detects, for each frame of the input video signal, a luminance feature amount representing the luminance of the frame (luminance feature amount detection unit). Specifically, the APL of the frame is detected. The APL can be obtained by dividing the total luminance value of each pixel in the frame by the total number of pixels in the frame. In this embodiment, APL is detected with an accuracy of 8 bits (256 gradations) from 0 to 255. However, the present invention is not limited to this. For example, APL may be detected with an accuracy of 128, 512 gradations. Good. Further, the luminance feature amount is not limited to APL. The luminance feature amount may be data representing the luminance of the frame, such as a luminance histogram, the maximum luminance value of the frame, and the minimum luminance value.

The APL storage unit 104 stores a predetermined number of frames of APL (APL detected by the APL detection unit 103), delays the APL by the predetermined number of frames, and a γ curve generation unit 106 and a fade scene γ curve generation unit It outputs to 107. Specifically, the APL is recorded in a buffer memory (not shown). In this embodiment, the delay amount is set to 300 frames which is the same as the delay amount of the video delay unit 102. The frame and the APL may be stored in a common buffer memory, or may be stored in different buffer memories. FIG. 2 shows a relationship between time (frame number) and APL output from the APL detection unit 103 and the APL storage unit 104. As shown in FIG. 2, the APL is output from the APL storage unit 104 after being delayed by 300 frames (delay amount) from the APL detection unit 103.
In addition, when storing the APL for 300 frames, the APL storage unit 104 outputs the APL for 300 frames to the fade scene detection unit 105. That is, after storing 300 frames of APL, every time an APL is input, the stored 300 frames of APL are output to the fade scene detection unit 105.

The fade scene detection unit 105 detects a fade scene period from the luminance feature values of a plurality of frames (fade scene detection means). The detection result is output to fade scene γ curve generation unit 107 and γ curve selection unit 108. Specifically, the fade scene detection unit 105 acquires APL for 300 frames from the APL storage unit 104. Then, using the acquired APL, it is determined for each frame whether the frame is a fade scene, and the determination result is output to the fade scene γ curve generation unit 107 and the γ curve selection unit 108.
In this embodiment, a period in which APL continuously decreases monotonously or increases monotonously for 10 frames or more is detected as a fade scene period. Thereby, it is possible to detect a period such as a fade-in, fade-out, or cross-fade video scene.

Specifically, when the following conditional expressions 1 and 2 are satisfied continuously for 10 frames or more, it is determined as monotonic decrease, and when the following conditional expressions 3 and 4 are satisfied continuously for 10 frames or more, it is determined as monotonic increase. .
apl (i + f) −apl (i + 1 + f) ≧ th1 (1)
apl (i + f) −apl (i + 1 + f) ≦ th2 (2)
apl (i + 1 + f) −apl (i + f) ≧ th3 (3)
apl (i + 1 + f) −apl (i + f) ≦ th4 (4)

In equations 1 to 4, apl (i) is the APL of the i-th frame (frame i), and f is a variable used for determination. th1 is a threshold value for determining whether or not the APL of the next frame (subsequent frame) is lower than the APL of the determination target frame. th2 is a threshold value for determining whether or not the APL has largely changed between the determination target frame and the subsequent frame. th3 is a threshold value for determining whether or not the APL of the subsequent frame is higher than the APL of the determination target frame. th4 is a threshold value for determining whether or not the APL has largely changed between the determination target frame and the subsequent frame. The reason why Equations 2 and 4 are provided is that when the APL greatly changes between the determination target frame and the subsequent frame, it is highly possible that the scene has suddenly changed and the possibility of being a fade scene is low. It is.
In this embodiment, th1 = 1, th2 = 10, th3 = 1, and th4 = 10. This value is merely an example, and any value may be used as long as it can be determined whether APL is monotonously decreasing or monotonically increasing. Th1 and th3 and th2 and th4 may be different from each other.

Next, a processing flow for detecting a fade scene period will be described with reference to FIG.
Hereinafter, a variable indicating whether or not the frame is a fade scene is defined as flg (i). Specifically, if the frame is not a fade scene, flg (i) = 0, and if the frame is a fade scene, flg (i) = 1.

  In the present embodiment, not only the information indicating whether or not the frame is a fade scene, but further information is generated (detected) and output for processing by the fade scene γ curve generation unit 107. Specifically, when the number of frames of the fade scene is n and the i-th frame is a frame of the fade scene, f_no (i) (= 0, 1,..., N−1), apl_s , Apl_e are output. f_no (i) is a frame number within the fade scene period, and apl_s and apl_e are the APL of the first frame (start frame) of the fade scene and the APL of the last frame (end frame) of the fade scene, respectively. It is. For frames that are not fade scenes, n = f_no (i) = apl_s = apl_e = −1 is output.

In the present embodiment, as an example, the case where the APL of each frame has a value as shown in FIG. 4 will be described. The fade scene detection unit 105 determines whether or not it is a frame of a fade scene in accordance with the processing flow of FIG.
First, in step S101, i = 0 is set, and the process proceeds to step S102 (initialization process).
In step S102, it is determined whether determination processing for 300 frames has been performed. At this point in time, the determination target frame i is frame 0, and determination processing for 300 frames has not been performed (S102: no), and the process proceeds to step S103.
In step S103, f = 0 is set, and the process proceeds to step S104.
In step S104, it is determined whether apl (i + f) and apl (i + 1 + f) satisfy Expressions 1 and 2. Since the APL of frame 0 is 80, the APL of frame 1 is 80, and the APL of frames 0 and 1 does not satisfy Equations 1 and 2 (S104: no), the process proceeds to step S106. When apl (i + f) and apl (i + 1 + f) satisfy the expressions 1 and 2 (S104: yes), 1 is added to f, and the process proceeds to step S103.
In step S106, it is determined whether or not f ≧ 9 is satisfied. At this time, since f = 0 and f ≧ 9 is not satisfied (S106: no), the process proceeds to S107.
In step S107, f = 0 is set, and the process proceeds to step S108.
In step S108, it is determined whether apl (i + f) and apl (i + 1 + f) satisfy Expressions 3 and 4. Since the APL of frame 0 is 80, the APL of frame 1 is 80, and the APL of frames 0 and 1 does not satisfy Equations 3 and 4 (S108: no), the process proceeds to step S110. If apl (i + f) and apl (i + 1 + f) satisfy Expressions 3 and 4 (S108: yes), 1 is added to f, and the process proceeds to step S108.
In step S110, it is determined whether or not f ≧ 9 is satisfied. At this time, since f = 0 and f ≧ 9 is not satisfied (S110: no), the process proceeds to S112.
In step S112, it is determined that the determination target frame i is not a fade scene frame, and flg (i) = 0, n = −1, f_no (i) = − 1, apl_s = −1 with respect to the determination target frame i. , Apl_e = −1, the process proceeds to step S114. At this time, it is determined that the frame 0 is not a fade scene, and flg (0) = 0, n = −1, f_no (i) = − 1, apl_s = −1, and apl_e = −1.
In step S114, i = i + f + 1 is set, and the process proceeds to step S102. At this time, i = 1.

Thereafter, until i = 199, it is determined that the frame is not a fade scene.
Next, a processing flow for frames after frame 200 will be described.
First, in step S102, since the determination target frame i is the frame 200, and determination for 300 frames has not been performed, the process proceeds to step S103.
In step S103, f = 0 is set, and the process proceeds to step S104.
In step S104, the APL of the frame 200 is 78, the APL of the frame 201 is 76, and the APL of the frames 200 and 201 satisfies Expressions 1 and 2 (S104: yes), the process proceeds to step S105.
In step S105, 1 is added to f, and the process proceeds to step S103. At this time, f = 1.
In step S103, the APL of the frame 201 is 78, the APL of the frame 202 is 76, and the APL of the frames 201 and 202 satisfies Expressions 1 and 2, so the process proceeds to step S105.

Thereafter, expressions 1 and 2 are satisfied in step S104 until f = 14 in step S103.
When f = 14, the APL of the frame 214 is 50 and the APL of the frame 215 is 50 in step S104, and the expressions 1 and 2 are not satisfied. Therefore, the process proceeds to step S106.
Next, in step S106, since f = 14 and f ≧ 9 is satisfied (S106: yes), the process proceeds to step S113.
In step S113, the period of frames i to i + f is set as the fade scene period. Specifically, the period of frames 200 to 214 is a fade scene period.

In step S113, since the period of frames 200 to 214 is a fade scene period, flg (200) = flg (201) =... = Flg (214) = 1. The frame numbers in the fade scene period are f_no (200) = 0, f_no (201) = 1,..., F_no (214) = 14. The number n of fade scenes is 15, the APL (apl_s) of the start frame is 78, and the APL (apl_e) of the end frame is 50.
Then, the process proceeds to step S114.
In step S114, i = 215 is set, and the process proceeds to step S102. Then, processing for the frames after frame 215 is performed.
Thereafter, it is similarly determined whether or not the frame is a fade scene. In this embodiment, it is determined that the frames 215 to 300 are not fade scene frames. When the determination process for 300 frames is performed (S102: yes), the fade scene detection unit 105 outputs a determination result as shown in FIG. 5 to the fade scene γ curve generation unit 107 and the γ curve selection unit.

  The γ curve generation unit 106 generates an image processing parameter for each frame using the APL delayed from the APL storage unit 104 and outputs the image processing parameter to the γ curve selection unit 108. In this embodiment, a γ curve is generated as an image processing parameter. The image processing parameter is not limited to the γ curve. Any parameter may be used as long as it is used for gradation correction. Hereinafter, an example of processing of the γ curve generation unit 106 will be described in detail.

The γ curve generation unit 106 holds a γ curve G0 that floats a dark part and a γ curve G1 that sinks a dark part. The γ curve generation unit 106 calculates the blend rate α (x) using Equations 5-7. In Expressions 5 to 7, x is an AP that is output with a delay from the APL storage unit 104.
L, and a, b, and c are predetermined coefficients. Then, the γ curve G (x) is generated by synthesizing the two γ curves G0 and G1 with weights corresponding to the blend rate α (x).
For example, G0 is a γ curve as shown in FIG. 6A, and G1 is a γ curve as shown in FIG. 6B, for example.
When α (x) = a × x 0 ≦ x ≦ 64 (5)
α (x) = b × (x−64) + α (64) When 65 ≦ x ≦ 205 (6)
α (x) = c × (x−205) + α (205) When 206 ≦ x ≦ 255 (7)
In this embodiment, it is assumed that a = 0.0002, b = 0.0052, and c = 0.00252. FIG. 7 shows the relationship between APLx and blend rate α (x) in this case.

Formula 8 shows a method for generating the γ curve G (x) using the blend rate α (x).
G (x) = α (x) × G1 + (1−α (x)) × G0 (8)
According to Equation 8, when APLx is 0, α (0) = 0, so G (0) = G0. When APLx is 255, α (255) = 1, so G (255) = G1. When APLx is 60, α (60) = 0.132 from Equation 5, and thus G (60) = 0.132 × G1 + 0.868 × G0. FIG. 6C shows a γ curve of G (60) obtained by Expression 8.

  The fade scene γ curve generation unit 107 generates image processing parameters for frames in the fade scene period based on the APL output from the APL storage unit 104 with a delay and the detection result of the fade scene detection unit 105. (Decision means). Specifically, the fade scene γ curve generation unit 107 determines (generates) an image processing parameter to be applied to the first and last frames of the fade scene period from the luminance feature amount of each frame. Then, the image processing parameters (in-scene parameters) applied to a plurality of frames between the first and last frames are gradually changed from the image processing parameters applied to the first frame to the image processing parameters applied to the last frame. Decide to change. In this embodiment, the in-scene parameters are an image processing parameter determined from the luminance feature amount of the frame, and an image processing parameter when the image processing parameter is assumed to change linearly from the first frame to the last frame. Decide to take a value between. This will be described in detail below.

Formula 9 shows a calculation method of the blend rate α_f (i) for the frames within the fade scene period.
α_f (i) = d × α (x) + e × β (i) (9)
Note that x in Equation 9 is the APL of the frame i, and d and e are predetermined coefficients. In this embodiment, d = 0.5 and e = 0.5.
Β (i) in Expression 9 is obtained by blending ratio α of the end frame from blend ratio α (apl_s) (blending ratio obtained by Expressions 5 to 7) of the start frame of the fade scene.
This is the blend ratio of frame i when it is assumed that the linearity changes to (apl_e). The calculation method of the blend rate β (i) is shown in Equation 10.
β (i) = α (apl_s)
+ (Α (apl_e) −α (apl_s)) × (f_no (i) / (n−1))
(10)

Specifically, the blend rate α_f (i) of the frame 209 is calculated as follows. In this embodiment, as shown in FIG. 5, apl_s = 78, apl_e = 50, and n = 15. From Equation 5, α (64) = 0.1408. Therefore, from Expressions 5 and 6, α (apl_s) and α (apl_e) are 0.2136 and 0.11, respectively. The APL of the frame 209 is 60 and f_no (209) = 9.
β (209) = α (78) + (α (50) −α (78)) × (9/14)
= 0.2136+ (0.11-0.2136) × (9/14)
= 0.147
α_f (209) = 0.5 × α (60) + 0.5 × β (209)
= 0.5 x 0.132 + 0.5 x 0.147
= 0.1395

In the present embodiment, the fade scene γ curve generation unit 107 generates a γ curve using the blend rate α_f (i) in the same manner as the γ curve generation unit 106. Therefore, the γ curve of the frame 209 is 0.1395 × G1 + 0.8605 × G0.
Then, the frame number in the fade scene period and the γ curve generated by the above method are output to the γ curve selection unit 108.

In the present embodiment, since the period of frames 200 to 214 is a fade scene period, the relationship between the frame number and the blend rate α_f (i) within the fade scene period is as shown in FIG.
8B shows the blend rate obtained by performing the time-direction LPF processing, and FIG. 8C shows the blend rate obtained from the APL of the frame without performing the time-direction LPF processing. .

When performing the time-direction LPF process, the blend rate transitions as shown in FIG. 8B, so the application of the blend rate is delayed in the last frame (end frame) of the fade scene period, and blackout and overexposure occur. Will occur.
Further, when the time direction LPF processing is not performed, the blend rate transitions as shown in FIG. 8C, so that the frame number f_no (i) which is the blend rate change point (the point at which the blend rate becomes discontinuous). Flickers before and after 7.
On the other hand, in this embodiment, as shown in FIG. 8A, the blend rate gradually changes from the blend rate applied to the start frame of the fade scene period to the blend rate applied to the end frame. For this reason, it is possible to suppress the occurrence of flickering and the delay of parameter application (blackout and overexposure).

  The γ curve selection unit 108 selects the γ curve output from the fade scene γ curve generation unit 106 for frames within the fade scene period. The γ curve output from the γ curve generation unit 106 is applied to frames outside the fade scene period. Whether or not the frame is within the fade scene period is determined using the determination result output from the fade scene detection unit 105. Specifically, the γ curve selection unit 108 determines that the frame with flg (i) = 1 is a frame within the fade scene period, and the frame with flg (i) = 0 is out of the fade scene period. It is determined that it is a frame.

The γ conversion unit 109 performs gradation correction using a γ curve corresponding to the frame output from the video delay unit 102 (γ curve output from the γ curve selection unit 108), and outputs a frame after gradation correction. Output to the signal terminal (correction means). In other words, the frames within the fade scene period are subjected to gradation correction using the γ curve (gradation correction parameter) generated (determined) by the fade scene γ curve generation unit 106. Frames outside the fade scene period are subjected to gradation correction using the γ curve generated by the γ curve generation unit 106.
The output signal terminal 110 outputs the gradation-corrected frame to an image display device (television device) (not shown), a recording device, or the like.

As described above, according to the present embodiment, the image processing parameter applied to the first frame is gradually changed from the image processing parameter applied to the first frame between the first and last frames of the fade scene period. Image processing parameters are determined to change. As a result, delay in application of the γ curve and a sense of interference such as flickering can be suppressed.

<Example 2>
Hereinafter, an image processing apparatus according to Embodiment 2 of the present invention and an image processing method executed by the image processing apparatus will be described with reference to the drawings.

  FIG. 9 is a functional block diagram of the image processing apparatus according to the present embodiment. As illustrated in FIG. 9, the image processing apparatus 200 includes functional blocks similar to the functional blocks of the image processing apparatus 100 according to the first embodiment. In FIG. 9, the same functional blocks as those of the first embodiment are denoted by the same reference numerals as those of the first embodiment, and description thereof is omitted. This embodiment is different from the first embodiment in that the output of the fade scene detection unit 905 is reflected by the video delay unit 902 and the APL storage unit 904, and the delay amount is variable.

  The video delay unit 902 stores a plurality of frames, delays the input video signal by the plurality of frames, and outputs it to the γ conversion unit 109. In the present embodiment, the number of frames required for the fade scene detection unit 905 to first detect the fade scene period after the input video signal is input is used as the delay amount. The video delay unit 902 can store a predetermined number (for example, 300 frames) of frames. If a fade scene period is not detected by the fade scene detection unit 905 even if a predetermined number of frames are input, a predetermined number of frames (300 frames) is set as the delay amount.

The APL storage unit 904 stores APLs of a plurality of frames, delays the APL by the plurality of frames, and outputs the delayed APLs to the γ curve generation unit 106 and the fade scene γ curve generation unit 107. In this embodiment, the delay amount is the same as that of the video delay unit 902.
In this embodiment, when the APL storage unit 904 stores the APL for a predetermined number of frames, the APL for the predetermined number of frames is output to the fade scene detection unit 905. Specifically, when the APL storage unit 904 stores the APL for 100 frames, the APL for the 100 frames is output to the fade scene detection unit 905.
Therefore, in this embodiment, each frame is subjected to gradation correction at the timing when the first fade scene period is detected in the γ conversion unit 109 and is output.

  The fade scene detection unit 905 acquires 100 frames of APL from the APL storage unit 904, and detects the period of the fade scene. The detection result is output to fade scene γ curve generation unit 107 and γ curve selection unit 108. Further, when the fade scene detection unit 905 detects the period of the fade scene before the APL for 300 frames is input, the fade scene detection unit 905 determines the number of frames required to detect the period of the first fade scene. 902 and the APL storage unit 904.

Next, a processing flow for detecting a fade scene will be described with reference to FIG. In FIG. 10, the same processes as those in the first embodiment are denoted by the same reference numerals, and the description thereof is omitted.
In this embodiment, after step S101, it is determined whether or not determination processing for 100 frames has been performed (step S202). If the determination process for 100 frames has not been performed (S202: no), the process proceeds to step S103. If the determination process for 100 frames has been performed (S202: yes), the fade scene period is detected. The process ends.
After step S113, it is determined whether or not the fade scene period is detected for the first time (step S215). If the fade scene period is detected for the first time (S215: yes), the process proceeds to step S216. If not (S215: no), the process proceeds to step S114.
In step S216, the number of frames required to detect the fade scene period first is output to the video delay unit 902 and the APL storage unit 904.

As in the first embodiment, consider the case where the APL of each frame is as shown in FIG. In this case, since the period of the frames 200 to 214 is first detected as a fade scene period, 215 is output to the video delay unit 902 and the APL storage unit 904 as the number of frames required for detection.
As a result, the delay amount of the video delay unit 902 and the APL storage unit 904 is set to 215 frames. If the number of frames required for detection exceeds 300, the delay amounts of the video delay unit 902 and the APL storage unit 904 are set to 300 frames.

  As described above, in this embodiment, each frame is subjected to gradation correction at the timing when the period of the first fade scene is detected and output. Specifically, when the first fade scene period is detected before the predetermined number of frames are accumulated in the video delay unit 902 after the input of the first frame of the input video signal is started, the detection is performed. At each timing, each frame is subjected to gradation correction and output. Thereby, it is possible to shorten the time until the video is displayed as compared with the first embodiment.

DESCRIPTION OF SYMBOLS 100,200 Image processing apparatus 103 APL detection part 105,905 Fade scene detection part 107 Fade scene gamma curve generation part 109 γ conversion part

Claims (10)

An image processing apparatus that dynamically switches image processing parameters used for gradation correction based on luminance of an input video signal,
A luminance feature amount detecting means for detecting a luminance feature amount representing the luminance of the frame for each frame of the input video signal;
Fade scene detection means for detecting a fade scene period in which the luminance of the frame monotonously increases or decreases from the luminance feature quantity of a plurality of frames;
Image processing parameters to be applied to the first and last frames of the fade scene period are determined from the luminance feature values of the respective frames, and image processing parameters to be applied to a plurality of frames between the first and last frames are determined. Determining means for deciding to gradually change from an image processing parameter applied to the last frame to an image processing parameter applied to the last frame;
Correction means for correcting gradation using the image processing parameter determined by the determination means for a frame within a fade scene period;
An image processing apparatus comprising: The determination means includes an image processing parameter to be applied to a frame between the first and last frames, an image processing parameter determined from a luminance feature amount of the frame, and an image from the first frame to the last frame. The image processing apparatus according to claim 1, wherein the determination is made to take a value between the image processing parameter when the processing parameter is assumed to change linearly. It has a buffer memory that can store a predetermined number of frames for detection of fade scenes,
If the period of the first fade scene is detected by the fade scene detection means before the predetermined number of frames are accumulated in the buffer memory after the input of the first frame of the input video signal is started, 3. The image processing apparatus according to claim 1, wherein a process of correcting and outputting the gradation of each frame is started at a timing when the period of the first fade scene is detected.   The fade scene detection means includes
    When a predetermined number or more of frames having a value obtained by subtracting the luminance feature value of the next frame from the luminance feature value of the frame is equal to or greater than the first threshold value and equal to or less than the second threshold value, Judge the period as the period of the fade scene where the brightness of the frame decreases monotonously,
    When a predetermined number or more of frames having a value obtained by subtracting the luminance feature value of the frame from the luminance feature value of the next frame is not less than the third threshold value and not more than the fourth threshold value, Judge the period as the period of the fade scene where the brightness of the frame increases monotonously
The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus.   The determining means includes
    For the first frame of the fade scene period, the blending rate is determined from the luminance feature amount of the frame, and the first image processing parameter and the second image processing parameter are blended at the determined blending rate to perform image processing. Determine the parameters,
    For the last frame of the fade scene period, a blend rate is determined from the luminance feature amount of the frame, and the first image processing parameter and the second image processing parameter are blended at the determined blend rate. Determine image processing parameters,
    For each of a plurality of frames between the first and last frames, gradually change from the blend rate determined for the first frame to the blend rate determined for the last frame. And determining the image processing parameter by blending the first image processing parameter and the second image processing parameter at the determined blend rate.
The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. An image processing method executed by an image processing apparatus that dynamically switches image processing parameters used for gradation correction based on luminance of an input video signal,
A luminance feature amount detection step for detecting a luminance feature amount representing the luminance of the frame for each frame of the input video signal;
A fade scene detection step for detecting a period of a fade scene in which the luminance of the frame monotonously increases or decreases from the luminance feature amount of the plurality of frames;
Image processing parameters to be applied to the first and last frames of the fade scene period are determined from the luminance feature values of the respective frames, and image processing parameters to be applied to a plurality of frames between the first and last frames are determined. A determination step for deciding to gradually change from an image processing parameter to be applied to a frame of the image processing parameter to be applied to the last frame;
A correction step for correcting gradation using the image processing parameters determined in the determination step for frames within a fade scene period;
An image processing method comprising:   In the determining step, an image processing parameter to be applied to a frame between the first and last frames, an image processing parameter determined from a luminance feature amount of the frame, and an image from the first frame to the last frame Decide to take a value between the image processing parameters assuming that the processing parameters change linearly
The image processing method according to claim 6.   The image processing apparatus has a buffer memory capable of storing a predetermined number of frames for detecting a fade scene,
  When the period of the first fade scene is detected in the fade scene detection step before the predetermined number of frames are accumulated in the buffer memory after the input of the first frame of the input video signal is started, The timing at which the period of the first fade scene was detected
Processing to output each frame with gradation correction
The image processing method according to claim 6 or 7, wherein:   In the fade scene detection step,
    When a predetermined number or more of frames having a value obtained by subtracting the luminance feature value of the next frame from the luminance feature value of the frame is equal to or greater than the first threshold value and equal to or less than the second threshold value, Judge the period as the period of the fade scene where the brightness of the frame decreases monotonously,
    When a predetermined number or more of frames having a value obtained by subtracting the luminance feature value of the frame from the luminance feature value of the next frame is not less than the third threshold value and not more than the fourth threshold value, Judge the period as the period of the fade scene where the brightness of the frame increases monotonously
The image processing method according to claim 6, wherein:   In the determination step,
    For the first frame of the fade scene period, the blending rate is determined from the luminance feature amount of the frame, and the first image processing parameter and the second image processing parameter are blended at the determined blending rate to perform image processing. Determine the parameters,
    For the last frame of the fade scene period, a blend rate is determined from the luminance feature amount of the frame, and the first image processing parameter and the second image processing parameter are blended at the determined blend rate. Determine image processing parameters,
    For each of a plurality of frames between the first and last frames, gradually change from the blend rate determined for the first frame to the blend rate determined for the last frame. And determining the image processing parameter by blending the first image processing parameter and the second image processing parameter at the determined blend rate.
The image processing method according to any one of claims 6 to 9.

Images