JP5441656B2 - Image processing apparatus, image processing method, and program - Google Patents

Description

The present invention relates to an image processing apparatus , an image processing method, and a program such as a digital still camera and a digital video camera. The present invention relates to an image processing apparatus that performs processing, an image processing method, and a program.

  Conventionally, in this type of image processing apparatus, the appropriate image quality correction varies depending on the scene to be shot. Therefore, the shooting scene is automatically determined by analyzing the color information and luminance information included in the image, and depending on the determination result. A device that performs image quality correction has been proposed (Patent Document 1).

  For example, when it is determined that the scene is “blue sky”, image quality correction and shooting control are performed in which the noise removal level of an area considered to be blue sky is set strongly and the target luminance is set so that the saturation of the blue sky is reflected. If it is determined that the scene is an “evening scene”, image quality correction and shooting control are performed so as to increase the noise removal level of the area considered to be an evening scene and to keep the target luminance of the entire image low. If it is determined that the scene is a “night scene”, image quality correction and shooting control are performed so as to suppress the noise and increase the exposure time.

  Also, combining background judgment results for background scenes such as “blue sky” and “night view” with subject judgment results for main subject scenes such as whether the subject is close (“macro”) or the subject is “person” A technique for determining one scene has also been proposed.

  For example, when shooting a person against a blue sky, the background determination result is “blue sky” and the subject determination result is “person”. In addition to the control when the blue sky is determined as described above, Set the noise reduction level of the area you think is strong. Then, the weighting of the brightness of the face area is increased, and image quality correction and shooting control are performed so that the brightness of the face area approaches the target brightness more preferentially.

  In addition, when shooting a close subject with a blue sky as the background, the background determination result is “blue sky” and the subject determination result is “macro”. In addition to the control when the blue sky is determined as described above, camera shake compensation is performed. Shooting control is performed to widen the shift range of the target image.

  That is, even if the same “blue sky” background determination result is obtained, the subject determination result is “person”, “macro”, or “person” or “macro”. The parameters used for shooting control and image quality correction are changed depending on the case where it is not performed.

JP 2009-116691 A

  In the conventional scene determination described above, image quality correction and shooting control according to the scene can be performed. However, if the scene determination result changes frequently, parameters used for shooting control and image quality correction change. Images (especially moving images) may be difficult to see.

  In addition, in the conventional scene determination described above, there may be inconveniences by combining the subject determination result regarding “macro” or “person” with the background determination result such as “blue sky” or “night view”.

  For example, when “evening scene” is selected as the background determination result, as described above, in order to emphasize the atmosphere of the evening scene, the noise removal level of the area considered to be the evening scene is increased, and the target luminance of the entire image is reduced. Thus, image quality correction and shooting control are performed.

  On the other hand, when “person” is detected as the subject determination result, image quality correction and shooting control for increasing the brightness of the face of the person are performed. That is, if the face of a person is dark, image quality correction and shooting control are performed to adjust the brightness of the entire image so that the face becomes bright.

  For this reason, when trying to shoot a person with a sunset background, the target luminance of the entire image is suppressed according to the background determination result, whereas the target luminance of the entire image is increased according to the subject determination result. A malfunction occurs.

  Also, for example, when a person is photographed in a shop with many red walls and objects and is generally dark, the background determination result other than “evening scene” and the background determination result “dark (indoor)” are easily interchanged. The situation can be considered.

  At this time, if a background determination result of “dark (indoor)” is obtained, the presence / absence of “person” may be determined as subject determination. For this reason, although the subject itself has not changed much, a situation occurs in which the incompatible scene determination results of “evening scene” and “dark (indoor) + person” are interchanged. As a result, there is a possibility that the contents of the image quality correction and the shooting control are greatly changed despite the same situation.

Accordingly, the present invention is incompatible scene between the determination result to avoid frequent changes, the scene determination stability image processing apparatus that can be improved, and aims to provide an image processing method and program And

In order to achieve the above object, an image processing apparatus according to the present invention performs a determination on a first scene of an image captured by an imaging unit, and each scene included in the first scene is determined. The number of times the scene condition is continuously satisfied is counted, and when the count reaches a set value, the determination result of the first scene is switched to the scene where the count has reached the set value. A determination processing unit and a second scene of the image captured by the imaging unit are determined, and the scene condition is continuously satisfied for each of the scenes included in the second scene. A second determination processing means for switching the determination result of the second scene to a scene where the count has reached the set value when the count reaches a set value; Based on the determination result obtained by the first determination processing unit and the determination result obtained by the second determination processing unit, an image processing unit that performs image processing on the image captured by the imaging unit is controlled. Control means, wherein the second determination processing means determines that a condition of a specific scene included in the second scene is satisfied, or counts a specific scene included in the second scene Is determined to have reached a predetermined value smaller than the set value, the first determination processing unit resets the count of scenes included in the first scene.

  According to the present invention, it is possible to avoid frequent changes in determination results between incompatible scenes, and to improve the stability of scene determination.

1 is a perspective view illustrating an overview of an imaging apparatus according to an example of an embodiment of an image processing apparatus of the present invention. It is a schematic block diagram for demonstrating the control system of an imaging device. It is a flowchart for demonstrating the background determination process by a camera control part. It is a flowchart figure for demonstrating the switching process of the background determination result by a camera control part. It is a flowchart for demonstrating the object determination process by a camera control part. It is a flowchart for demonstrating the scene determination process by a camera control part. It is a figure which shows the example of a combination of a background determination result and a subject determination result.

  Hereinafter, an example of an embodiment of the present invention will be described with reference to the drawings.

  FIG. 1 is a perspective view showing an overview of an imaging apparatus according to an example of an embodiment of an image processing apparatus of the present invention.

  As shown in FIG. 1, the imaging apparatus 120 according to the present embodiment includes a lens unit 121 disposed on the front surface, and an operation for instructing the start and end of the electronic viewfinder (EVF) 123 and moving image shooting on the rear surface. A moving image trigger switch 124 as a switch is arranged. On the upper surface of the imaging device 120, a microphone 122 that records sound at the time of shooting and a still image trigger switch 125 that is an operation switch for instructing start and end of still image shooting are arranged.

  In addition, a mode dial 126 is disposed on the left side as viewed from the lens unit 121 side of the imaging device 120, and an operation switch group 127, a display unit 128, and a speaker 129 for outputting reproduced sound are provided on the right side. A battery 130 (detachable) is arranged.

  The operation switch group 127 includes keys for inputting a white balance mode, and other keys for menu operation, reproduction operation, and the like. The display unit 128 is configured by a liquid crystal panel or the like, and is used for confirming a subject image when photographing with a camera and displaying a reproduced image during reproduction, as with the electronic viewfinder 123. The display unit 128 is supported so as to be openable and closable with respect to the main body of the imaging device 120 and to be rotatable in the open state.

  FIG. 2 is a schematic block diagram for explaining a control system of the imaging apparatus 120.

  In FIG. 2, a diaphragm 102 is disposed on the image plane side of the lens 101 constituting the lens unit 121 of FIG. The subject light that has passed through the lens 101 and the diaphragm 102 forms an image on the image sensor 105 such as a CCD sensor or a CMOS sensor, undergoes photoelectric conversion, and is output to the image processing unit 108 of the signal processing unit 112 as an image signal.

  The image processing unit 108 performs white balance correction processing, noise removal processing, edge enhancement processing, saturation correction processing, gamma correction processing, and the like on the image signal acquired from the image sensor 105, and then displays the information on the display unit 128. Output to the unit 110 or the recording medium 117. Examples of the recording medium 117 include a magnetic tape, a DVD disk, or a memory card.

  The information detection unit 110 detects camera information such as luminance, color, focus, subject distance, image stabilization, and face from the image data acquired from the image processing unit 108 and outputs the detection result to the camera control unit 111.

  As a face detection process, a known method, for example, a method of extracting a skin color region from the gradation color of each pixel represented by image data and detecting the face with a matching degree with a face contour plate recorded in advance. There is. In addition, there is a method for detecting a face by extracting feature points of a face such as eyes, nose, and mouth using a known pattern recognition technique, but the method is not particularly limited.

  The camera control unit 111 performs background determination processing and subject determination processing based on detection results of camera information such as luminance, color, focus, subject distance, image stabilization, and face acquired from the information detection unit 110, and performs background determination and A shooting scene is determined by a combination with subject determination. The camera control unit 111 calculates at least one of a white balance correction value, a noise removal level, an edge enhancement value, a saturation correction level, and a gamma correction value based on the determined scene.

  Then, the camera control unit 111 performs white balance control for capturing an image using the calculated white balance correction value, and controls the image processing unit 108 so that the white balance of the image becomes appropriate. In addition, the camera control unit 111 controls the image processing unit 108 so as to suppress image noise and perform edge enhancement using the calculated noise removal level and edge enhancement value so that the sense of resolution of the image becomes appropriate. To do.

  Furthermore, the camera control unit 111 uses the calculated gamma correction value to control the image processing unit 108 so that the gradation and contrast of the image are appropriate. Note that background determination processing, subject determination processing, and scene determination processing by the camera control unit 111 will be described later.

  Further, the camera control unit 111 performs control such as charge accumulation and reading of the image sensor 105 via the sensor driving unit 106, and performs control such as focusing and zooming of the lens 101 via the lens driving unit 103.

  The camera control unit 111 controls the aperture 102 via the aperture drive unit 104 using the luminance information of the subject obtained from the image data captured by the image sensor 105, and also uses the sensor drive unit 106. The exposure is controlled by controlling the shutter speed. Further, the camera control unit 111 corrects camera shake during shooting by driving the lens 101 via the lens driving unit 103 or by controlling the signal processing unit 112.

  Next, background determination processing by the camera control unit 111 will be described with reference to FIG. Each process in FIG. 3 is executed by the CPU or the like of the camera control unit 111 by loading a program stored in a storage unit such as a ROM (not shown) into the RAM.

  In step S301, the camera control unit 111 determines whether or not the luminance value (BV value) of the subject acquired from the information detection unit 110 exceeds a preset threshold value Th1.

  If the luminance value (BV value) exceeds the threshold value Th1, the camera control unit 111 proceeds to step S307. If the luminance value (BV value) is equal to or less than the threshold value Th1, the camera control unit 111 proceeds to step S302.

  In step S <b> 302, the camera control unit 111 determines whether or not it is a night scene by determining whether or not a luminance histogram created from the entire image satisfies a predetermined condition.

  If the camera control unit 111 determines that the scene is a night view, the background determination result is “night scene” in step S304. If the camera control unit 111 determines that the scene is not a night scene, the process proceeds to step S303.

  In step S <b> 303, the camera control unit 111 determines whether or not the spotlight is a local light by determining whether or not there is a region with high brightness.

  If the camera control unit 111 determines that the light is a spotlight, the background determination result is “spotlight” in step S305. If the camera control unit 111 determines that the light is not a spotlight, the background determination result is “dark” in step S306. Indoor) ”.

  On the other hand, in step S307, the camera control unit 111 determines whether or not the red component area and the dark area whose luminance value is equal to or less than the threshold value are arranged in a predetermined order and area on the image. Determine whether.

  If it is an evening scene, the camera control unit 111 sets the background determination result to “evening scene” in step S308. If not, the process proceeds to step S309.

  In step S309, the camera control unit 111 determines whether the backlight is back by determining whether the brightness value at the center of the image is darker than the brightness value at the periphery by a predetermined threshold or more.

  If the camera control unit 111 is backlit, the process proceeds to step S310. If not, the process proceeds to step S315.

  In step S310, the camera control unit 111 determines whether there is a blue sky by determining whether a region that satisfies the color as a blue sky exists in a predetermined area or more.

  If it is a blue sky scene, the camera control unit 111 sets the background determination result to “backlit blue sky” in step S311. If it is not a blue sky scene, the camera control unit 111 proceeds to step S312.

  In step S <b> 312, the camera control unit 111 generates a hue histogram from the image, and determines whether or not the hue is vivid by determining whether or not a hue that is equal to or greater than a predetermined threshold exists.

  If the camera control unit 111 determines that the background is bright, the background determination result is “backlit vivid” in step S313, and if it is not bright, the background determination result is “backlight” in step S314.

  On the other hand, in step S315, the camera control unit 111 determines whether there is a blue sky by determining whether a region satisfying a color as a blue sky exists in a predetermined area or more.

  If the camera control unit 111 determines that the sky is blue, the background determination result is “blue sky” in step S316. If the camera control unit 111 determines that the sky is not blue, the process proceeds to step S317.

  In step S <b> 317, the camera control unit 111 generates a hue histogram from the image, and determines whether the hue that is equal to or greater than the threshold is greater than or equal to a predetermined value, thereby determining whether the image is vivid.

  If the camera control unit 111 determines that the background is bright, the background determination result is “bright” in step S318. If the camera control unit 111 determines that the background is bright, the camera control unit 111 sets the background determination result to “bright” in step S319.

  Here, in the present embodiment, in order to prevent the scene determination result from being frequently replaced, when the same background determination result is continuously obtained a predetermined number of times or more, the image quality corresponding to the background determination result of the scene is not changed for the first time. Switch to correction and shooting control.

  Next, a background determination result switching process performed by the camera control unit 111 will be described with reference to FIG. Each process in FIG. 4 is executed by the CPU or the like of the camera control unit 111 by loading a program stored in a storage unit such as a ROM (not shown) into the RAM. 4 exemplifies a case where the processes in steps S304, S305, and S308 in FIG. 3 are replaced with steps S401, S404, and S407, respectively. Here, the processing in FIGS. 3 and 4 corresponds to an example of the first determination processing means of the present invention for determining the first scene of the image.

  In step S401 in FIG. 4A, since the background determination result is “night scene”, the camera control unit 111 increments the night scene counter Cn, and the process proceeds to step S402.

  In step S402, the camera control unit 111 determines whether or not the value of the night view counter Cn exceeds a preset setting value N.

  If the value of the night view counter Cn exceeds the set value N, the camera control unit 111 proceeds to step S403. In addition, when the value of the night view counter Cn is equal to or less than the set value N, the camera control unit 111 does not change the background determination result and maintains the image quality correction parameter and the shooting control parameter according to the previous background determination result.

  In step S403, the camera control unit 111 switches the background determination result from the previous one to “night scene”, and changes the image quality correction parameter and the shooting control parameter to values corresponding to “night scene”.

  In step S404 of FIG. 4B, since the background determination result is “spotlight”, the camera control unit 111 increments the spotlight counter Csp and proceeds to step S405.

  In step S405, the camera control unit 111 determines whether or not the value of the spotlight counter Csp exceeds a preset setting value N.

  If the value of the spotlight counter Csp exceeds the set value N, the camera control unit 111 proceeds to step S406. Further, when the value of the spotlight counter Csp is equal to or smaller than the set value N, the camera control unit 111 does not change the background determination result and maintains the image quality correction parameter and the shooting control parameter according to the previous background determination result. .

  In step S406, the camera control unit 111 switches the background determination result from the previous one to “spotlight”, and changes the image quality correction parameter and the shooting control parameter to values corresponding to “spotlight”.

  In step S407 of FIG. 4C, since the background determination result is “evening scene”, the camera control unit 111 increments the evening scene counter Csu, and the process proceeds to step S408.

  In step S408, the camera control unit 111 determines whether or not the value of the evening scene counter Csu exceeds a preset setting value N.

  If the value of the evening scene counter Csu exceeds the set value N, the camera control unit 111 proceeds to step S409. Further, when the value of the evening scene counter Csu is equal to or smaller than the set value N, the camera control unit 111 does not change the background determination result, and maintains the image quality correction parameter and the shooting control parameter according to the previous background determination result.

  In step S409, the camera control unit 111 switches the background determination result from the previous one to “evening scene”, and changes the image quality correction parameter and the shooting control parameter to values corresponding to “evening scene”.

  The same applies to the other background determinations in FIG. 3, and the dark counter Cd, the backlight blue sky counter Cgs, the backlight bright counter Cga, the backlight counter Cg, the blue sky counter Cs, and the bright counter according to the respective determination results. Ca, bright counter Cb is counted. When the value of the counter exceeds the set value N, the background scene indicated by the background determination result is switched, and the setting is changed to the image quality correction parameter and the shooting control parameter corresponding to the switched background scene.

  In this embodiment, the counter value in each background determination result is incremented only when the same background determination result is obtained continuously, and is reset when another background determination result is obtained midway. Is done. This is to improve the reliability and stability of the background determination result. When the value of the counter exceeds the set value N, that is, the background determination result is switched only after the same determination result continues a predetermined number of times.

  When the background determination result is a scene of “night scene”, the camera control unit 111 sets the noise suppression level by the image processing unit 108 to be higher than a normal value and sets the upper limit of the exposure time by the sensor driving unit 106. Set longer.

  When the background determination result is a scene of “spotlight”, the camera control unit 111 refers to only the luminance level of the area determined as the spotlight, and the sensor driving unit 106 so that this luminance level becomes the target luminance. To control.

  When the background determination result is “dark (indoor)”, the camera control unit 111 sets the noise suppression level and the edge enhancement level by the image processing unit 108 higher than normal values.

  When the background determination result is a scene of “evening scene”, the camera control unit 111 increases the noise suppression level by the image processing unit 108 for a color region that is considered to be an evening scene to be higher than a normal value, and sets the target of the entire image. The sensor driving unit 106 is controlled so as to keep the luminance low.

  If the background determination result is a scene of “backlit blue sky”, the camera control unit 111 raises the noise suppression level by the image processing unit 108 for a color region that seems to be blue sky higher than a normal value. Further, the camera control unit 111 increases the weighting of the luminance at the center of the image, and controls the sensor driving unit 106 so that the luminance of the obtained entire image reaches the target luminance.

  When the background determination result is a scene with “brilliant backlight”, the camera control unit 111 increases the saturation of the entire screen by the image processing unit 108 and increases the weighting of the luminance at the center of the image to obtain the entire image. The sensor driving unit 106 is controlled so that the luminance reaches the target luminance.

  When the background determination result is a “backlit” scene, the camera control unit 111 controls the sensor driving unit 106 so that the luminance of the entire image obtained by increasing the luminance weight of the center of the image reaches the target luminance. To do.

  If the background determination result is a “blue sky” scene, the camera control unit 111 sets the noise suppression level by the image processing unit 108 for a color region that seems to be blue sky higher than a normal value.

  When the background determination result is a “bright” scene, the camera control unit 111 increases the saturation of the entire screen by the image processing unit 108. When the background determination result is a “bright” scene, the camera control unit 111 sets the noise suppression level by the image processing unit 108 to be lower than a normal value.

  Next, a subject determination process by the camera control unit 111 will be described with reference to FIG. Each process in FIG. 5 is executed by the CPU or the like of the camera control unit 111 by loading a program stored in a storage unit such as a ROM (not shown) into the RAM. The subject determination process here is started simultaneously with the background determination process described above. Here, the processing in FIG. 5 corresponds to an example of the second determination processing means of the present invention for determining the second scene of the image.

  In step S501, the camera control unit 111 determines the presence / absence of a person in the image based on the face detection result acquired from the information detection unit 110. If there is a person, the process proceeds to step S502. The process proceeds to S510.

  In step S502, the camera control unit 111 resets only the evening scene counter Csu, the night scene counter Cn, and the spotlight counter Csp, and the process proceeds to step S503.

  In step S503, the camera control unit 111 determines whether there is a person (face) movement. If there is a movement, the process proceeds to step S504. If there is no movement, the process proceeds to step S507.

  In step S504, the camera control unit 111 increments the value of the person movement counter Cmv, and the process proceeds to step S505.

  In step S505, the camera control unit 111 determines whether or not the value of the person movement counter Cmv exceeds a preset setting value N. If the value of the person movement counter Cmv exceeds the setting value N, The process proceeds to step S506. Further, when the value of the person movement counter Cmv is equal to or smaller than the set value N, the camera control unit 111 does not change the subject determination result and maintains the image quality correction parameter and the shooting control parameter according to the previous subject determination result. .

  In step S506, the camera control unit 111 switches the subject determination result from the previous one to “person movement”, and changes the settings of the image quality correction parameter and the shooting control parameter to those corresponding to “person movement”.

  In step S507, the camera control unit 111 increments the value of the person counter Cp, and proceeds to step S508.

  In step S508, the camera control unit 111 determines whether or not the value of the person counter Cp exceeds a preset set value N. If the value of the person counter Cp exceeds the set value N, step S509 is performed. Proceed to Further, when the value of the person counter Cp is equal to or smaller than the set value N, the camera control unit 111 does not change the subject determination result and maintains the image quality correction parameter and the shooting control parameter according to the previous subject determination result.

  In step S509, the camera control unit 111 switches the subject determination result from the previous one to “person” and changes the settings of the image quality correction parameter and the shooting control parameter to those corresponding to “person”.

  On the other hand, in step S510, the camera control unit 111 uses the subject distance information acquired from the information detection unit 110 to determine whether or not the macro shooting is performed.

  The information detection unit 110 calculates the subject distance from the positions of the focus lens and the zoom lens when the subject is in focus by autofocus control. If the subject distance is shorter than the threshold, the camera control unit 111 determines that the macro shooting is performed, and proceeds to step S511. If not, the process proceeds to step S514.

  In step S511, the camera control unit 111 increments the value of the macro counter Cmc, and proceeds to step S512.

  In step S512, the camera control unit 111 determines whether or not the value of the macro counter Cmc exceeds a preset setting value N. If the value of the macro counter Cmc exceeds the setting value N, step S513 is performed. Proceed to In addition, when the value of the macro counter Cmc is equal to or smaller than the set value N, the camera control unit 111 does not change the subject determination result and maintains the image quality correction parameter and the shooting control parameter according to the previous subject determination result.

  In step S513, the camera control unit 111 switches the subject determination result from the previous one to “macro”, and changes the settings of the image quality correction parameter and the imaging control parameter to those corresponding to “macro”.

  On the other hand, in step S514, the camera control unit 111 increments the value of the landscape counter Cv, and the process proceeds to step S515.

  In step S515, the camera control unit 111 determines whether or not the value of the landscape counter Cv exceeds a preset setting value N. If the value of the landscape counter Cv exceeds the set value N, step S516 is performed. Proceed to Further, when the value of the landscape counter Cv is equal to or less than the set value N, the camera control unit 111 does not change the subject determination result and maintains the image quality correction parameter and the shooting control parameter according to the previous subject determination result.

  In step S516, the camera control unit 111 changes the subject determination result from the previous one to “landscape (auto)”.

  In this embodiment, the counter value in each subject determination result is incremented only when the same subject determination result is obtained continuously, and is reset when another subject determination result is obtained midway. Is done. This is to improve the reliability and stability of the subject determination result. When the value of the counter exceeds the set value N, that is, the subject determination result is switched only after the same determination result continues for a predetermined number of times. .

  Further, when the subject determination result is a scene of “person”, the camera control unit 111 increases the noise suppression level by the image processing unit 108 for a region that seems to be a person's skin color to be higher than a normal value. In addition, the camera control unit 111 increases the weighting of the luminance of the face area, and controls the sensor driving unit 106 so that the luminance of the face area approaches the target luminance more preferentially.

  When the subject determination result is a scene of “person movement”, the camera control unit 111 gives priority to shortening the exposure time than usual when performing exposure control in addition to the control of “person”. Thus, the sensor driving unit 106 is controlled.

  When the subject determination result is a “macro” scene, the camera control unit 111 controls the anti-vibration lens included in the lens 101 by the lens driving unit 103 so as to widen the shift range of the image subjected to camera shake compensation. To do.

  When the subject determination result is a “landscape (auto)” scene, the camera control unit 111 performs image correction and shooting control based on the background determination result without performing control based on the subject determination result.

  As described above, in the present embodiment, regardless of the subject determination result, when a person is detected even once by the information detection unit 110, the evening scene counter Csu, the night scene counter Cn, the spotlight is forcibly detected. The counter Csp is reset.

  The fact that a person is detected by the information detection unit 110 is considered to have already been determined as a scene in which “person” exists, or that there is a high probability of shifting to a scene in which “person” exists. Therefore, at the stage where a person is detected, the count of “sunset”, “night view”, and “spotlight”, which is incompatible with the scene where there is a “person” that tries to bring the brightness of the person's face to the target brightness positively, is counted. Reset to make it difficult to move to these scenes. Here, if a person is detected even once, the counts of “Evening Scene”, “Night Scene”, and “Spotlight” are reset. When the count reaches a predetermined value, these counts may be reset. In this case, if the predetermined value is set to be equal to or less than the set value N, it is difficult to shift from a scene where “person” exists to a scene of “evening scene”, “night scene”, and “spotlight” which are incompatible with this scene. Can be obtained.

  On the contrary, even when a person is detected by the information detection unit 110, the counters other than the evening scene counter Csu, the night scene counter Cn, and the spotlight counter Csp are not reset. As described above, in this embodiment, when a person is detected, the count of a part of the scene that is incompatible with the scene determination result in which “person” exists is reset, whereas the count of the part other than the part of the scene is reset. It is configured not to reset. This makes it possible to adjust the ease of transition between scenes according to the combination of scenes.

  Next, the scene determination process by the camera control unit 111 will be described with reference to FIG. Each process in FIG. 6 is executed by the CPU or the like of the camera control unit 111 by loading a program stored in a storage unit such as a ROM (not shown) into the RAM. The scene determination process by the camera control unit 111 is started after the background determination process and the subject determination process.

  In step S601, the camera control unit 111 determines whether the background determination result is an evening scene, a night scene, or a spotlight.

  Then, when the background determination result is any of the sunset scene, the night scene, and the spotlight, the camera control unit 111 does not refer to the subject determination result in step S602 and uses only the background determination result as the scene determination.

  In addition, when the background determination result is other than the sunset scene, the night scene, and the spotlight, the camera control unit 111 performs scene determination by combining the background determination result and the subject determination result. For example, if the background determination result is “blue sky” and the subject determination result is “person”, the scene determination result is “ Blue sky + person ".

  FIG. 7 shows a combination example of the background determination result and the subject determination result.

  In FIG. 7, the background determination results of “night scene”, “spotlight”, and “evening scene” do not have a combination with the subject determination result. Combined with any of “person movement”, “macro”, and “landscape”. However, when the subject determination result is “landscape”, the result of the background determination process is reflected as it is, and therefore “landscape” is not described in FIG.

  In addition, the background determination results “night view”, “spotlight”, and “evening scene” are not compatible with the subject determination results “person”, “person movement”, and “macro”. For this reason, in this embodiment, priority is provided to one determination result, and when one determination result is selected, the other determination result is not referred to.

  As described above, in the present embodiment, the determination result is switched only after the same determination result for the background determination process and the subject determination process is continued a predetermined number of times or more. Thereby, it is possible to avoid frequent changes in the scene determination result, and it is possible to improve the reliability and stability of the scene determination. As a result, it is possible to suppress significant fluctuations in the image quality correction parameter and the imaging control parameter, and it is possible to maintain the image quality level.

  In this embodiment, when it is detected that the scene is a specific scene, the counter of the scene that is incompatible with the scene is reset. This improves the stability of scene determination for a plurality of incompatible scenes, avoids duplication of correction, and makes it possible to perform appropriate correction processing and shooting control on the shooting scene. As a result, it is possible to avoid repetitive changes in the image quality correction and the contents of the imaging control in spite of the same situation.

  In addition, this invention is not limited to what was illustrated to the said embodiment, In the range which does not deviate from the summary of this invention, it can change suitably.

  For example, in this embodiment, the counter for “evening scene”, “night scene”, and “spotlight” is reset when it is determined that the subject is determined to be “person”. Even if it is determined that there is, the counter may be similarly reset. On the other hand, the counter of “person” and “macro” may be reset when it is determined that the background determination result is “evening scene”, “night scene”, or “spotlight”.

  The present invention can also be realized by executing the following processing. That is, software (program) that realizes the functions of the above-described embodiments is supplied to a system or apparatus via a network or various storage media, and a computer (or CPU, MPU, or the like) of the system or apparatus reads the program. It is a process to be executed. It can also be realized by executing software (program) acquired via a network or various storage media on a personal computer (CPU, processor).

105 Image Sensor 108 Image Processing Unit 110 Information Detection Unit 111 Camera Control Unit 120 Imaging Device

Claims (6)

The determination is made for the first scene of the image captured by the imaging means, and for each of the scenes included in the first scene, the number of times that the scene condition is continuously satisfied is counted, First determination processing means for switching a determination result of the first scene to a scene where the count has reached a set value when the count has reached a set value;
The second scene of the image captured by the imaging means is determined, and the number of times that the scene condition is continuously satisfied is counted for each of the scenes included in the second scene. A second determination processing means for switching the determination result of the second scene to a scene where the count has reached the set value when the count reaches a set value;
Based on the determination result obtained by the first determination processing unit and the determination result obtained by the second determination processing unit, an image processing unit that performs image processing on the image captured by the imaging unit is controlled. Control means,
The second determination processing unit determines that the condition of the specific scene included in the second scene is satisfied, or the count of the specific scene included in the second scene is smaller than the set value The image processing apparatus according to claim 1, wherein when it is determined that the predetermined value has been reached, the first determination processing unit resets the count of a scene included in the first scene. The second determination processing unit determines that the condition of the specific scene included in the second scene is satisfied, or the count of the specific scene included in the second scene is smaller than the set value When it is determined that the predetermined value has been reached, the first determination processing unit resets the count of some scenes included in the first scene, and resets the count of scenes other than the some scenes. The image processing apparatus according to claim 1, wherein the image processing apparatus is not.   The first scene is a scene related to one of a background and a subject, and the second scene is a scene related to the other of the background and the subject. Or the image processing apparatus of 2.   4. The specific scene is a person or macro scene, and the scene whose count is reset by the first determination processing unit is an evening scene, a night scene, or a spotlight scene. An image processing apparatus according to 1. The determination is made for the first scene of the image captured by the imaging means, and for each of the scenes included in the first scene, the number of times that the scene condition is continuously satisfied is counted, A first determination processing step of switching a determination result of the first scene to a scene where the count has reached a set value when the count has reached a set value;
The second scene of the image captured by the imaging means is determined, and the number of times that the scene condition is continuously satisfied is counted for each of the scenes included in the second scene. A second determination processing step of switching the determination result of the second scene to a scene where the count reaches the set value when the count reaches a set value;
Based on the determination result obtained in the first determination processing step and the determination result obtained in the second determination processing step, an image processing unit that performs image processing on the image captured by the imaging unit is controlled. A control step,
In the second determination processing step, it is determined that the condition of the specific scene included in the second scene is satisfied, or the count of the specific scene included in the second scene is smaller than the set value An image processing method comprising: resetting a count of scenes included in the first scene in the first determination processing step when it is determined that a predetermined value has been reached. The determination is made for the first scene of the image captured by the imaging means, and for each of the scenes included in the first scene, the number of times that the scene condition is continuously satisfied is counted, A first determination processing step of switching a determination result of the first scene to a scene where the count has reached a set value when the count has reached a set value;
The second scene of the image captured by the imaging means is determined, and the number of times that the scene condition is continuously satisfied is counted for each of the scenes included in the second scene. A second determination processing step of switching the determination result of the second scene to a scene where the count reaches the set value when the count reaches a set value;
Based on the determination result obtained in the first determination processing step and the determination result obtained in the second determination processing step, an image processing unit that performs image processing on the image captured by the imaging unit is controlled. A control step,
In the second determination processing step, it is determined that the condition of the specific scene included in the second scene is satisfied, or the count of the specific scene included in the second scene is smaller than the set value If it is determined to have reached the predetermined value, the first determination processing step, the program for causing executed to reset the count of the scene on a computer included in the first scene.

Images