JP4700557B2 - Imaging device - Google Patents

Description

  The present invention relates to an imaging apparatus that performs encoding based on correlation between frames for a moving image.

  When capturing a moving image, a scene change that switches the scene to be captured may occur. Specifically, such a scene change occurs when the imaging apparatus is panned or zoomed.

  As a technique for detecting such a scene change, for example, when a difference is obtained for each pixel data between two consecutive frames and the sum of absolute values of the obtained differences exceeds a predetermined threshold, a scene change is detected. A technique for determining that occurrence has occurred is known.

  By the way, an apparatus for encoding a moving picture using inter-frame correlation has been widely known. However, in such an encoding apparatus, when a scene change occurs and the inter-frame correlation is reduced, encoding is performed. The efficiency decreases and the data amount increases, or the image quality deteriorates when the data amount is maintained.

  In view of this, a technique has been proposed for preventing the picture quality from degrading due to a decrease in coding efficiency when a scene change occurs.

  For example, in Japanese Patent Application Laid-Open No. 2001-78196, when a scene change is detected when compressing a moving image, instead of performing inter-frame predictive coding that processes between frames that are temporally forward and backward, A technique for performing intra-frame coding to be processed within the frame is described.

The exposure time of the frame image in such a conventional technique is generally 1/30 seconds or 1/60 seconds.
JP 2001-78196 A

  However, in the case of imaging using an imaging apparatus using the conventional technique as described above and having a lens with a long focal length, for example, 1/60 seconds or 1/30 seconds as described above. In this exposure time, there is a possibility that image blur exceeding an allowable level may occur due to camera shake or the like. In such a case, since the image of each frame is blurred, it is impossible to prevent the deterioration of the image quality regardless of whether or not the moving image encoding is highly accurate.

  Further, even with an encoding method using both inter-frame prediction encoding and intra-frame encoding as described in the above-mentioned Japanese Patent Application Laid-Open No. 2001-78196, the image quality due to image blur as described above can be reduced. Deterioration cannot be prevented, and furthermore, this encoding method complicates the encoding device and the decoding device.

  The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an imaging apparatus capable of preventing deterioration in image quality of a frame image caused by a scene change without complicating encoding and decoding. It is said.

In order to achieve the above object, an image pickup apparatus according to a first aspect of the present invention provides an image pickup apparatus that performs encoding based on inter-frame correlation on a moving image composed of a group of frame images obtained by image pickup. An imaging unit that generates a plurality of time-division images to form the frame image by performing imaging with a plurality of time-division exposure times obtained by dividing the exposure time, and detects whether a scene change has occurred A scene change detection unit that generates a frame image by synthesizing the plurality of time-division images, and if the scene change detection unit detects that a scene change has occurred, An image composition unit that synthesizes the time-division image to generate a frame image so as to prevent image quality degradation of the resulting frame image, and an image composition unit. The generated frame image comprising an encoding unit for encoding, a based on inter-frame correlation, in the case where the image synthesizing unit has been detected that the scene change occurs by the scene change detection unit, Only one of the time-division image group before the scene change occurs in the time-division image group related to one frame image and the time-division image group after the scene change occurs are combined. By generating a frame image, image quality deterioration of the frame image due to the scene change is prevented.

The image pickup apparatus according to the second invention is the image pickup apparatus according to the first invention, wherein the image composition unit includes a time-division image group before the scene change occurs and a time after the scene change occurs. When only one of the divided image groups is combined, a frame image is generated by further amplifying the combined image to compensate for the lack of time-divided images.

  According to the imaging apparatus of the present invention, it is possible to prevent deterioration of the image quality of a frame image due to a scene change without complicating encoding and decoding.

  Embodiments of the present invention will be described below with reference to the drawings.

[Embodiment 1]
1 to 4 show a first embodiment of the present invention, and FIG. 1 is a block diagram showing the configuration of an imaging apparatus.

  The imaging apparatus is configured as a digital video camera having a digital camera function, for example.

  That is, as shown in FIG. 1, the imaging apparatus includes a lens 1, an imaging device 2, an imaging circuit 3, an A / D converter 4, a signal processing circuit 5, a frame memory 6, and a FIFO memory 7. An on-screen circuit 8, a TFT liquid crystal driving circuit 9, a TFT panel 10, a backlight unit 11, a video output circuit 12, a video output terminal 13, a recording buffer 14, and a recording medium interface (recording medium I). / F) 15, recording medium 16, actuator 17, actuator drive circuit 18, EEPROM 19, shake detection unit 21, external data interface (external data I / F) 22, key matrix 23, and LCD display A circuit 24, an LCD panel 25, a battery 26, a power supply circuit 27, a backup power supply 28, a battery state detection circuit 29, A CPU 31, is configured to include a first CPU 32, a.

  The lens 1 is for forming an optical subject image on the image sensor 2.

  The imaging device 2 is an imaging unit that photoelectrically converts an optical subject image formed by the lens 1 and outputs an electrical video signal. As long as the image pickup device 2 is of a type that can perform high-speed reading, any one of a CCD, a CMOS, or another type of image pickup device may be used.

  The imaging circuit 3 performs various analog signal processing on the output from the imaging device 2.

  The A / D converter 4 is for converting an analog video signal output from the imaging circuit 3 into a digital video signal.

  The signal processing circuit 5 is for performing various digital signal processings on the video signal output from the A / D converter 4.

  The frame memory 6 is a storage unit that stores various data and the like, and includes an exposure data holding unit 6a for storing a video signal (exposure data) processed by the signal processing circuit 5.

  The FIFO memory 7 is a memory for performing buffering when displaying or outputting a moving image obtained by imaging or a real-time video that becomes a monitor image at the time of still image imaging.

  The on-screen circuit 8 is a circuit for superimposing character data and other on-screen data on the video signal output from the FIFO memory 7. The data superimposed by the on-screen circuit 8 is output from the first CPU 31.

  The TFT liquid crystal drive circuit 9 controls the TFT panel 10 based on the video signal output from the on-screen circuit 8.

  The TFT panel 10 is capable of color display, and is based on the control of the TFT liquid crystal driving circuit 9 and is capturing a moving image, a real-time image, a still image after photographing, various information relating to the imaging device, and the like. Is a display means for displaying.

  The backlight unit 11 is provided on the back side of the TFT panel 10 and illuminates the TFT panel 10 from the back side.

  The video output circuit 12 is for outputting the video signal from the on-screen circuit 8 to the outside via the video output terminal 13.

  The video output terminal 13 is a connection terminal for connecting an external video cable or the like.

  The recording buffer 14 performs buffering when the video signal stored in the frame memory 6 is recorded on the recording medium 16. The recording buffer 14 also performs buffering of signals read from the recording medium 16.

  The recording medium interface 15 is for controlling data recording on the recording medium 16 and data reading from the recording medium 16.

  The recording medium 16 is a non-volatile recording medium for recording image data and other various data, and is configured to be detachable from the imaging apparatus, for example.

  The actuator 17 is a drive source for driving the lens 1 to perform autofocus, or to perform automatic zoom when an automatic zoom function is provided.

  The actuator drive circuit 18 controls and drives the actuator 17 based on the control of the first CPU 31.

  The EEPROM 19 is a non-volatile storage medium for storing various data relating to the imaging apparatus, a processing program executed by the first CPU 31 or the second CPU 32, and the like.

  The shake detection unit 21 includes, for example, a gyro sensor, detects angular velocities in the yaw direction and pitch direction generated in the imaging apparatus, and detects shakes of the imaging apparatus based on the detected angular velocities. Shake detection means.

  The external data interface 22 is configured by, for example, a USB or the like, and is an interface for transmitting / receiving data to / from an external device.

  The key matrix 23 is a general term for operation input means including various operation switches and operation buttons provided in the imaging apparatus. As a specific example, the key matrix 23 includes a power switch for turning on the power of the imaging apparatus, a photographing button for inputting a photographing operation, and the like. A signal generated by operating the key matrix 23 is input to the first CPU 31.

  The LCD display circuit 24 is a circuit for controlling the LCD panel 25 to perform various displays based on the control of the first CPU 31.

  The LCD panel 25 is configured to include, for example, a monochrome LCD, and displays information such as an operation mode such as a shooting mode set in the imaging apparatus and a recording time of a video recorded on the recording medium 16. Display means.

  A battery 26 is a main power source of the imaging apparatus.

  The backup power supply 28 is a power supply for supplying power for date display in this imaging apparatus.

  The power supply circuit 27 is a circuit that performs control to supply the power of the battery 26 and the backup power supply 28 to each circuit in the imaging apparatus based on a command from the first CPU 31.

  The battery state detection circuit 29 detects the voltage of the battery 26, etc., calculates the remaining battery level of the battery 26, and outputs the result to the first CPU 31.

  The first CPU 31 is a main CPU, and is a control means for comprehensively controlling each circuit of the imaging apparatus. The first CPU 31 includes a system control unit 31a for controlling the system of the imaging apparatus, and in the system control unit 31a, a divided exposure control unit 31b, a shake correction control unit 31c, and an I frame determination unit 31d. Is provided.

  Here, the division exposure control unit 31b calculates the exposure time (shutter time) of the frame image based on the photometric data and the like, and further divides this exposure time into a plurality of times to expose a plurality of time division images. To control.

  Further, the blur correction control unit 31c grasps the positional relationship (blurring state) of a plurality of time-division images photographed under the control of the above-described divided exposure control unit 31b based on the output from the blur detection unit 21. This is a shake correction means for controlling the shake correction amount when the image addition is performed by the divided exposure composition unit 32a described later of the second CPU 32.

  Furthermore, the I frame determination unit 31d is a reference frame determination unit that determines a target frame (I frame (frame of I picture)) to be subjected to intra-frame encoding. Here, the I frame is a reference frame that is referred to as a target of inter-frame correlation from other frames (P frame (P picture frame) and B frame (B picture frame)).

  Thus, the first CPU 31 mainly performs control, and the second CPU 32 mainly performs processing for handling image data.

  That is, the second CPU 32 processes the image data stored in the frame memory 6. The second CPU 32 includes a divided exposure combining unit 32a, an MPEG image compression / decompression unit 32b, a still image compression / decompression unit 32c, a scene change detection unit 32d, and a recording medium access unit 32e. .

  The divided exposure composition unit 32a performs alignment (blur correction) of each time-division image as necessary based on the control of the shake correction control unit 31c, and then converts each time-division image based on the control of the division exposure control unit 31b. This is an image composition unit that adds and generates a normal one-frame exposure image (an image corresponding to an image obtained by performing normal one-frame exposure by the image sensor 2), and serves as a shake correction unit.

  The MPEG image compression / decompression unit 32b performs a process of performing MPEG compression on each frame image generated by the divided exposure synthesis unit 32a or decompressing an MPEG compressed image recorded on the recording medium 16. It is.

  The still image compression / decompression unit 32c performs a process of, for example, JPEG compression or the like on the image data stored in the frame memory 6, or decompressing the compressed image data read from the recording medium 16. Is. Since the imaging apparatus also has a function as a digital camera as described above, the still image compression / decompression unit 32c is provided.

  The scene change detection unit 32d detects whether or not a scene change has occurred by calculating a correlation between exposure images that are temporally continuous among exposure images obtained by time-division exposure. is there. Note that, here, whether or not a scene change has occurred is detected based on the correlation, but other techniques may be used to detect whether or not a scene change has occurred. The scene change detection unit 32d can also detect whether or not image blur has occurred. At this time, the scene change detection unit 32d also serves as a blur detection unit (blur detection unit). .

  The recording medium access unit 32 e is for controlling access to the recording medium 16 by the recording medium interface 15.

  Next, FIG. 2 is a diagram for explaining the principle of image composition when a frame image is generated from a time-division image by this imaging apparatus.

  First, FIG. 2A shows frame data F1 to F3 obtained by normal photographing. These frame data F1 to F3 are obtained at a time interval (frame rate) of 1/30 seconds, for example.

  On the other hand, FIG. 2B shows each time-division image data obtained by time-division exposure. Here, the time-division image data D1-1 to D1-3 are obtained by dividing the exposure time of the frame data F1 into three based on the control of the division exposure control unit 31b and exposing each of the divided exposure times. Data. Similarly, the time-division image data D2-1 to D2-3 are data obtained by dividing the exposure time of the frame data F2 into three and exposing each of the divided exposure times, and time-division image data D3. −1 to D3-3 are data obtained by dividing the exposure time of the frame data F3 into three and exposing each of the divided exposure times. Here, exposure is performed by time-dividing one frame into three, but it is of course not limited to this, and exposure can be performed by time-sharing with an arbitrary number of divisions of 2 or more. .

  The time-division image data D1-1 to D1-3, D2-1 to D2-3, and D3-1 to D3-3 obtained by time division in this way are controlled based on the control of the division exposure control unit 31b. Then, the divided exposure combining unit 32a combines (adds) the respective frame data f1, f2, and f3 as shown in FIG. 2C. That is, the frame data f1 is synthesized by synthesizing the time-division image data D1-1 to D1-3, and the frame data f2 is synthesized by synthesizing the time-division image data D2-1 to D2-3. Frame data f3 is generated by combining 1 to D3-3. Thus, frame data f1 to f3 are generated as data corresponding to the frame data F1 to F3 obtained by performing normal exposure. The output rate of the frame data f1 to f3 is the same as that of the frame data F1 to F3, and is 1/30 second, for example.

  The frame data f1 to f3 generated in this way are compressed based on the interframe correlation by the MPEG image compression / decompression unit 32b.

  By the way, a scene change may occur when shooting. For example, it is assumed that a scene change occurs at time T1 shown in FIG.

  When this scene change occurs, the scene change detection unit 32d calculates the correlation between the time-division image data D1-1 and the time-division image data D1-2 (for example, the sum of absolute difference values for each pixel). When the calculated correlation indicates a correlation lower than a predetermined value (when the total value is larger than the predetermined value), it is detected.

  When such a scene change occurs, if the frame data f1 is generated by combining the time-division image data D1-1 to D1-3, images of different scenes are combined to obtain a high-quality image. I can't do that. Therefore, in the present embodiment, frame data is generated using only the time-division image group before the scene change occurs. Specifically, only the time-division image data D1-1 is used to generate the frame data f1, and the time-division image data D1-2 and D1-3 are not used. At this time, since the exposure amount is insufficient, the frame data f1 is obtained by amplifying the time-division image data D1-1 three times.

  Here, the frame data is generated using only the time-division image group before the scene change occurs. However, the present invention is not limited to this, and only the time-division image group after the scene change is used. Thus, frame data may be generated. At this time, the time division image data D1-2 and D1-3 are used to generate the frame data f1, and the time division image data D1-1 is not used. Furthermore, by comparing the number of time-division images related to one frame before the scene change occurs with the number after the scene change occurs, the time-division image group having the larger number is used. Frame data may be generated.

  Then, when the frame data f1 generated in this way is a frame that is referred to by another frame, the image quality deterioration also extends to the other frame. Therefore, the I frame determination unit 31d stores the frame data f1 in the frame data f1. The frame data f2 obtained by synthesizing the next time-division image data D2-1 to D2-3 is set as the I frame while not being set as the I frame.

  Similarly, for example, when a scene change occurs at time T2 shown in the figure, time-division image data D2-1 and D2-2 are used to generate frame data f2, and time-division image data D2-3 is It will not be used. Then, in order to compensate for the shortage of exposure amount, the image data obtained by synthesizing the time-division image data D2-1 and D2-2 is amplified 1.5 times to generate frame data f2.

  At this time, the frame data f2 is not set as an I frame, and frame data f3 obtained by synthesizing the next time-division image data D3-1 to D3-3 is set as an I frame.

  Next, FIG. 3 is a flowchart showing processing of photographing operation by the imaging apparatus.

  This photographing operation is started when a photographing button included in the key matrix 23 is operated.

  When this process starts, first, an AE operation is performed, and an exposure time of a frame to be imaged is calculated (step S1).

  Next, the division exposure control unit 31b calculates the number of time division exposures performed in order to finally obtain a frame image of this exposure time and the exposure time of each time division exposure (step S2).

  As processing in this step S2, time-division exposure time equal to or less than the camera shake limit second depending on the focal length of the lens 1 is calculated, and the exposure time calculated in step S1 is divided by this time-division exposure time, thereby time-division exposure. One example is to calculate the number of times. Accordingly, in the case where the exposure time calculated in step S1 is 1/15 seconds and the exposure time at the camera shake limit time is calculated as 1/60 seconds, the time-division exposure number is 4 times. Become.

  Subsequently, time-division exposure is performed according to the set time-division exposure time (step S3), and time-division exposure data is read every time one time-division exposure is completed (step S4).

  Then, it is determined whether or not the current time-division exposure is the first time-division exposure among a plurality of time-division exposures for dividing one frame (step S5).

  Here, when it is determined that the first time-division exposure is not performed, the scene change detection unit 32d calculates a pixel-by-pixel difference from the previous time-division image, and further sums the absolute values of the calculated differences. And scene change detection is performed based on whether the sum is greater than a predetermined value (step S6).

  Then, the system control unit 31a branches the process according to the result of the scene change detection by the scene change detection unit 32d (step S7).

  If it is determined that no scene change has occurred, the current time-division exposure data is further added to the time-division exposure data added up to the previous time (step S8).

  If it is determined in step S5 that it is the first time-division exposure, or if the process in step S8 is completed, whether or not the time-division exposure of the number of times calculated in step S2 has been completed. Is determined (step S9).

  If the predetermined number of time-division exposures have not yet been completed, the process returns to step S3 and the above-described processing is repeated.

  If it is determined in step S7 that a scene change has occurred, the scene change flag is turned on (step S10), and the insufficient exposure of the time-division exposure data added up to the previous time is digitally converted. It is amplified and corrected by processing (step S11).

  If it is determined that the process of step S11 is completed or a predetermined number of time-division exposures are completed in step S9 described above, the obtained frame image will be described later with reference to FIG. MPEG compression processing is performed as described above (step S12).

  Then, it is determined whether or not the shooting button operation state has changed and the shooting has ended (step S13). If it is determined that the shooting has not ended, the process returns to step S1 and the next frame is as described above. Process.

  Thus, if it is determined in step S13 that the photographing has been completed, the image is recorded on the recording medium 16 via the recording medium interface 15 under the control of the recording medium access unit 32e (step S14). The shooting operation ends.

  Next, FIG. 4 is a flowchart showing details of the MPEG compression processing in step S12 of FIG.

  When this process is started, it is determined whether or not a new GOP (Group of Pictures) flag is turned on (step S21). Here, GOP is a group including the above-described I frame, and is a group of frames obtained by referring to the I frame. The I-frame determining unit 31d turns on this new GOP flag when starting a new GOP.

  If it is determined that the new GOP (Group of Pictures) flag is turned on, the frame is registered as a new GOP, and the new GOP flag is cleared (step S22).

  On the other hand, if it is determined in step S21 that the new GOP (Group of Pictures) flag is not turned on, the frame is registered as an existing GOP (step S23).

  When the processing of step S22 or step S23 is thus completed, it is next determined whether or not the scene change flag is turned on (step S24).

  Here, if it is determined that the scene change flag is on, the I frame determination unit 31d registers the GOP to be completed up to this frame data, and a new GOP starts from the next frame. The new GOP flag is turned on (step S25).

  Then, the I frame determination unit 31d clears the scene change flag (step S26).

  When step S26 ends or when the scene change flag is OFF in step S24, the frame is processed according to the registered GOP configuration to generate compressed data (step S27), as shown in FIG. Return to the process.

  According to the first embodiment, when a frame is shot in a time division manner and a scene change occurs during the time division shooting, the time division image group before the scene change occurs. Or, since the frame images are synthesized only from the time-division image group after the scene change occurs, the images of different scenes will not be synthesized and the image quality will not deteriorate, and a good quality image frame image will be obtained Is possible.

  At this time, since the amount of exposure time is insufficient, the generated frame image is never darkened.

  Since the frame in which the scene change has occurred is not used as a reference frame that is referred to as a target of inter-frame correlation from other frames, the image quality of the frame image in which the scene change has occurred affects other frames. Therefore, the image quality of other frame images can be kept high.

[Embodiment 2]
FIG. 5 shows a second embodiment of the present invention, and is a flowchart showing processing of photographing operation by the imaging apparatus. In the second embodiment, parts that are the same as those in the first embodiment are given the same reference numerals and description thereof is omitted, and only differences are mainly described.

  In the present embodiment, when a frame image is generated by combining a plurality of time-division images, image blur between the time-division images is corrected and then combined. Note that the configuration of the imaging apparatus in the present embodiment is the same as that in the first embodiment described above.

  That is, as shown in FIG. 2, it is assumed that a scene change has occurred at time T1, for example. At this time, the shake correction control unit 31 c determines whether the generated scene change is a scene change that can correct the shake based on the output from the shake detection unit 21.

  When it is determined that the scene change is capable of correcting blur, processing different from that described above is performed.

  In other words, when it is determined that the scene change is capable of correcting blur, the divided exposure composition unit 32a performs time-division image data on the time-division image data D1-1 based on the control of the blur correction control unit 31c. D1-2 and time-division image data D1-3 are each subjected to shake correction and then combined to generate frame data. Here, the blur correction is performed by shifting the positions of the pixels to be added based on the blur amount detected by the blur detection unit 21 by an amount for correcting the blur amount.

  Thereafter, the I frame determination unit 31d sets the frame data f2 obtained by synthesizing the next time-division image data D2-1 to D2-3 as the I frame, as described above.

  The operation of the image pickup apparatus of this embodiment will be described with reference to FIG.

  It is assumed that the processing from step S1 to S7 is started, it is determined that a scene change has occurred in step S7, and the processing in step S10 is performed.

  Then, the shake correction control unit 31c determines whether or not a scene change is detected as a shake based on the output of the shake detection unit 21 (step S31).

  Here, when it is detected as a shake, the shake correction control unit 31c further determines whether or not the detected shake is a shake that can be corrected (step S32).

  If it is determined that the shake can be corrected, the data (that is, the amount and direction in which the pixel position of the current time-division image is shifted from the pixel position of the reference time-division image) Is generated by adding the inverse vector of the vector indicating the position vector of the current time-division image), and the process proceeds to step S8 described above.

  If it is determined in step S31 that a scene change has not been detected as a shake, or if it is determined in step S32 that the detected shake cannot be corrected, the process proceeds to step S11 described above. Processing as described above is performed.

  Other processes are the same as those shown in FIG. 3 of the first embodiment described above.

  According to the second embodiment, the same effects as those of the first embodiment described above can be obtained, and when shake correction is possible, the image of the time-division images is corrected and then combined to form a frame image. As a result of the generation, a high-quality frame image without blurring is obtained.

  Since the frame image in which the blur has occurred is not used as a reference frame that is referred to as a target of inter-frame correlation from other frames, the image quality of the frame image in which the blur has occurred affects the other frames. Therefore, the image quality of other frame images can be kept high.

  It should be noted that the present invention is not limited to the above-described embodiments, and various modifications and applications can be made without departing from the spirit of the invention.

  INDUSTRIAL APPLICABILITY The present invention can be suitably used for an imaging apparatus that performs encoding based on correlation between frames for a moving image.

1 is a block diagram illustrating a configuration of an imaging apparatus according to Embodiment 1 of the present invention. FIG. 3 is a diagram for explaining the principle of image composition when a frame image is generated from a time-division image by the imaging apparatus of the first embodiment. 5 is a flowchart illustrating processing of a shooting operation performed by the imaging apparatus according to the first embodiment. The flowchart which shows the detail of the MPEG compression process in step S12 of the said FIG. 9 is a flowchart showing processing of a photographing operation by the imaging apparatus according to the second embodiment of the present invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Lens 2 ... Image sensor (imaging part)
DESCRIPTION OF SYMBOLS 3 ... Imaging circuit 4 ... A / D converter 5 ... Signal processing circuit 6 ... Frame memory 6a ... Exposure data holding part 7 ... FIFO memory 8 ... On-screen circuit 9 ... TFT liquid crystal drive circuit 10 ... TFT panel 11 ... Backlight unit DESCRIPTION OF SYMBOLS 12 ... Video output circuit 13 ... Video output terminal 14 ... Recording buffer 15 ... Recording medium interface (recording medium I / F)
16 ... Recording medium 17 ... Actuator 18 ... Actuator drive circuit 19 ... EEPROM
21 ... blur detection unit 22 ... external data interface (external data I / F)
DESCRIPTION OF SYMBOLS 23 ... Key matrix 24 ... LCD display circuit 25 ... LCD panel 26 ... Battery 27 ... Power supply circuit 28 ... Backup power supply 29 ... Battery state detection circuit 31 ... 1st CPU
31a ... System control unit 31b ... Divided exposure control unit 31c ... Blur correction control unit 31d ... I frame determination unit (reference frame determination unit)
32 ... Second CPU
32a: Divided exposure composition unit (image composition unit)
32b... MPEG image compression / decompression unit (encoding unit)
32c: Still image compression / decompression unit 32d: Scene change detection unit 32e: Recording medium access unit

Claims (2)

In an imaging device that performs encoding based on inter-frame correlation for a moving image composed of a group of frame images obtained by imaging,
An imaging unit that generates a plurality of time-division images to form the frame image by performing imaging with a plurality of time-division exposure times obtained by dividing the exposure time of the frame image;
A scene change detection unit for detecting whether or not a scene change has occurred;
A frame image is generated by combining the plurality of time-division images. When the scene change detection unit detects that a scene change has occurred, the frame image resulting from the scene change An image synthesizing unit that synthesizes the time-division image to generate a frame image so as to prevent image quality degradation;
An encoding unit for encoding the frame image generated by the image combining unit based on the inter-frame correlation;
Equipped with,
When the scene change detection unit detects that a scene change has occurred, the image composition unit is a time-division image of the time-division image group related to one frame image before the scene change occurs. The frame image is generated by synthesizing only one of the group and the time-division image group after the scene change occurs, thereby preventing deterioration of the image quality of the frame image due to the scene change. There is an imaging apparatus. When only one of the time-division image group before the scene change occurs and the time-division image group after the scene change occurs, the image composition unit further adds the time-division image. The imaging apparatus according to claim 1, wherein the frame image is generated by amplifying the synthesized image by an amount that compensates for the shortage .

Images