JP5398610B2 - Solid-state electronic imaging device and operation control method thereof - Google Patents

Description

  The present invention relates to a solid-state electronic imaging device and an operation control method thereof.

  An image sensor is used in a digital camera, and there is one that mixes signal charges accumulated in a photodiode in order to obtain a composite image (Patent Document 1). In addition, the use of a C-MOS sensor as an image sensor has been put into practical use. Among such C-MOS sensors, there is one that specifies how the object moves (Patent Document 2). In the C-MOS sensor, since the signal charges accumulated in the photodiode are sequentially read out for each row, the exposure timing is different for each row. For this reason, when a moving subject is imaged, distortion may occur (so-called rolling distortion). There is also one that corrects such rolling distortion (Patent Document 3).

  However, there is still room for improvement in order to reduce rolling distortion.

Japanese Unexamined Patent Publication No. 2006-270658 JP 2006-217213 JP JP 2009-141717 A

  An object of the present invention is to reduce rolling distortion relatively easily.

  In the solid-state electronic imaging device according to the present invention, a photoelectric conversion element having a color filter having a characteristic of transmitting red, green or blue or cyan, magenta or yellow light components formed on the light receiving surface has odd-numbered rows, odd-numbered columns and even-numbered columns. Color filters arranged in rows even columns or odd rows even columns and even rows and odd columns and formed in two photoelectric conversion elements obliquely in the 4n (n is an integer of 0 or more) +1 and 4n + 2 rows And the characteristics of the color filters formed in the two photoelectric conversion elements in the oblique direction in the 4n + 3 and 4n + 4 lines are respectively stored in the C-MOS sensor and the odd-numbered photoelectric conversion elements. The signal charges are sequentially read from the m-th row (m is a positive odd number) row one by one in the row direction, and from the first row to the first row in response to the end of reading of the last row Read one line in the row direction, or read one line from the m-th line in the reverse direction to the row direction, and in reverse from the row direction one line at a time from the last line in response to the completion of reading the first line And the odd-numbered row signal charge reading control means for controlling the C-MOS sensor so as to output it as an odd-numbered video signal, and the signal charges accumulated in the odd-numbered photoelectric conversion elements one by one from the m-th row. When read in the row direction in order, the signal charges accumulated in the photoelectric conversion elements in the even-numbered rows are read one by one from the m + 1th row in the direction opposite to the row direction, and the reading of the first row is completed. Depending on the situation, reading from the last row one by one is performed in the direction opposite to the row direction, and the signal charges accumulated in the odd-numbered photoelectric conversion elements are read one by one from the mth row in the direction opposite to the row direction. If so, even rows The signal charges accumulated in the photoelectric conversion elements are read out in the row direction one by one from the (m + 1) th row, and read out in the forward direction one by one from the first row in response to the end of the reading of the last row. An even-numbered row signal charge readout control means for controlling the C-MOS sensor so as to output it as a signal is provided.

  The present invention also provides an operation control method suitable for the solid-state electronic imaging device. That is, a photoelectric conversion element in which a color filter having a characteristic of transmitting light components of red, green or blue or cyan, magenta or yellow is formed on the light receiving surface is an odd-numbered row odd-numbered column and an even-numbered row even-numbered column or an odd-numbered row even-numbered. Characteristics of the color filters formed in the two photoelectric conversion elements in the diagonal direction in the 4n (n is an integer of 0 or more) +1 row and the 4n + 2 row, and the 4n + 3 row, In an operation control method for a solid-state electronic imaging device including a C-MOS sensor having the same characteristics of color filters formed in two photoelectric conversion elements in oblique directions in the 4n + 4th row, odd-numbered row signal charge readout The control means reads out the signal charges accumulated in the odd-numbered photoelectric conversion elements in the row direction one by one from the m-th row (m is a positive odd number). Depending on the end of reading of the last row, reading from the first row one row at a time in the row direction, or reading from the mth row one row at a time in the direction opposite to the row direction, reading of the first row is completed In response to this, the C-MOS sensor is controlled so as to read out one row at a time from the last row in the direction opposite to the row direction and output as an odd row video signal, and the even row signal charge read control means has the odd number. When the signal charges accumulated in the photoelectric conversion elements in the rows are read in the row direction one by one from the m-th row, the signal charges accumulated in the photoelectric conversion elements in the even-numbered rows are changed to 1 from the (m + 1) -th row. Each row is read in the direction opposite to the row direction, and in response to the end of reading the first row, the rows are read one by one from the last row in the direction opposite to the row direction, and stored in the odd-numbered photoelectric conversion elements. Signal charge is 1 row from the m-th row When read in the direction opposite to the one row direction, the signal charges accumulated in the photoelectric conversion elements in the even number rows are read in the row direction one by one from the (m + 1) th row, and the reading of the last row is completed. Accordingly, the C-MOS sensor is controlled so as to read out from the first row one by one in the forward direction and output it as an even-numbered video signal.

  According to the present invention, the C-MOS sensor is obtained based on the odd-numbered video signal obtained based on the signal charges accumulated in the odd-numbered photoelectric conversion elements and the signal charge accumulated in the even-numbered photoelectric conversion elements. Even rows and video signals are obtained. An image represented by an odd-numbered video signal (odd-numbered image) is composed of pixels in odd-numbered rows of an image represented by a video signal obtained based on signal charges accumulated in photoelectric conversion elements in all rows. It is an image. An image represented by an even-numbered video signal (even-numbered row image) is composed of pixels in even-numbered rows of an image represented by a video signal obtained on the basis of signal charges accumulated in photoelectric conversion elements of all rows. It is an image to be. These two odd row images and even row images can be regarded as substantially the same. When half of the image is synthesized using half of the odd-numbered row image and the other half of the image is synthesized using the even-numbered row image, the signal charges accumulated in all the photoelectric conversion elements of the C-MOS sensor are read and the image is read. The time required to generate an image is shorter than when generating. Therefore, rolling distortion is reduced.

  A motion determining means for determining whether or not the subject has a motion with a speed greater than or equal to a predetermined speed, and the odd-numbered row signal when the motion determining means determines that the subject has a motion with a speed with a speed greater than or equal to a predetermined speed. Expressed by the upper or lower half of the odd-numbered row image represented by the odd-numbered row video signal obtained under the control of the readout control means and the even-numbered row video signal obtained under the control of the even-numbered row signal readout control means. An image generation means for generating an image of one frame from the lower half or the upper half of the even-numbered line image may be further provided.

  You may further provide the moving direction detection means which detects the moving direction of a to-be-photographed object. In this case, in response to the movement direction detection means detecting that the subject is moving in the left direction toward the left, the odd-numbered row signal charge read-out control means sends the photoelectric conversion element in the first row or the final row. The C-MOS sensor will be controlled so as to read out signal charges from the photoelectric conversion elements in the row.

  When the moving direction detecting means detects that the subject is moving in a direction other than the left direction, the odd-numbered row signal charge reading control means is configured to detect the photoelectric conversion element of the C-MOS sensor. The signal charges may be read from the photoelectric conversion elements in half of all the rows.

The schematic diagram of C-MOS is shown. The C-MOS read order is shown. The circuit diagram of C-MOS is shown. The C-MOS read order is shown. It shows how a composite image is generated. 2 shows the electrical configuration of a digital camera. It is a flowchart which shows the process sequence of a digital camera. The C-MOS read order is shown. The C-MOS read order is shown. It is a flowchart which shows the process sequence of a digital camera. It is a flowchart which shows the process sequence of a digital camera. It is an example of a to-be-photographed image. It is an example of a to-be-photographed image.

  FIG. 1 shows an embodiment of the present invention and shows a schematic diagram of a part of a C-MOS (Complementary Metal Oxide Semiconductor) sensor 40.

  In the C-MOS sensor 40, a large number of photodiodes 2 and 3 are arranged in the row direction and the column direction. The photodiodes 2 arranged in the odd rows are denoted by “2”, and the photodiodes 3 disposed in the even rows are denoted by “3”. In this embodiment, the photodiodes 2 or 3 are arranged in odd rows and odd columns and even rows and even columns, but photodiodes 2 or 3 may be arranged in odd rows and even columns and even rows and odd columns. Good.

  On the light receiving surface of the photodiode 2 or 3, a color filter having a characteristic of transmitting a red component, a green component, or a blue component is formed. Filters that transmit red light components are labeled with “R” or “r”, filters that transmit green light components are labeled with “G” or “g”, and blue light components The filter that passes through is labeled “B” or “b”. Characteristics of the color filters formed in the two photoelectric conversion elements in the diagonal direction in the 4nth (n is an integer of 0 or more) +1 row and the 4n + 2th row, and 2 in the diagonal direction in the 4n + 3th row and the 4n + 4th row. The characteristics of the color filters formed in the two photodiodes 2 and 3 are the same. Instead of a color filter having a characteristic of transmitting a red component, a green component, or a blue component, a color filter having a characteristic of transmitting a cyan component, a magenta component, or a yellow component may be provided. Even in such a case, as described above, the characteristics of the color filters formed in the two photoelectric conversion elements in the oblique direction are the same.

  FIG. 2 shows the reading order of the C-MOS sensor 40.

  In this embodiment, the signal charges accumulated in the photodiodes 2 in the odd-numbered rows are read in the row direction in order from the first row (every other row). On the other hand, the signal charges accumulated in the photodiodes 2 in the even-numbered rows are read in the direction opposite to the row direction in order (every other row) from the bottom last row. An image obtained based on the signal charges accumulated in the odd-numbered photodiodes 2 is referred to as an odd-numbered image 4, and an image obtained based on the signal charges accumulated in the even-numbered photodiodes 3 is referred to as an even-numbered image 5. To do.

  As described above, in this embodiment, the odd-numbered row image 4 and the even-numbered row image 5 are obtained by one imaging performed using the C-MOS sensor 40. In this embodiment, as can be seen from FIG. 1, the characteristics of the color filter formed in the two photoelectric conversion elements in the oblique direction in the 4n (n is an integer of 0 or more) +1 row and the 4n + 2 row and the 4n + 3 Since the characteristics of the color filters formed in the two photoelectric conversion elements in the oblique direction in the row and the 4n + 4th row are the same, the color characteristics (C of each pixel in the odd row image 4 and the even row image 5 are the same. The appearance order of the color filter of the MOS sensor 40 is the same.

  FIG. 3 is a circuit diagram of the C-MOS sensor 40.

  A source terminal of a first field effect transistor (FET) 21 is connected to the photodiodes 2 in the odd rows described above. A read line 25 is connected to the gate terminal of the first FET 21. The drain terminal of the first FET 21 is connected to the source terminal of the second FET 22 and the gate terminal of the third FET 23. A reset line 24 and a drain line 26 are connected to the gate terminal and the source terminal of the second FET 22, respectively. A signal line 27 and a drain line 26 are connected to the drain terminal and the source terminal of the third FET 23. The photodiode 2 and the first to third FETs 21 to 23 correspond to one pixel of the odd image. A group of the photodiode 2 and the first to third FETs 21 to 23 is referred to as an odd-numbered pixel unit 20.

  Similarly, the source terminals of first field effect transistors (FETs) 31 are connected to the photodiodes 3 in the even rows. A read line 35 is connected to the gate terminal of the first FET 31. The drain terminal of the first FET 31 is connected to the source terminal of the second FET 32 and the gate terminal of the third FET 33. A reset line 34 and a drain line 36 are connected to the gate terminal and the source terminal of the second FET 32, respectively. A signal line 37 and a drain line 36 are connected to the drain terminal and the source terminal of the third FET 33. The photodiode 3 and the first to third FETs 31 to 33 correspond to one pixel of the even image. A group of the photodiode 3 and the first to third FETs 31 to 33 is referred to as an even-numbered row pixel unit 30.

  Constant current sources 28 and 38 are connected to one ends of the signal lines 27 and 37.

  The first shift register 25 controls the accumulation and discharge of charges in the odd-numbered photodiodes 2, and the reset line 24 and the readout line 25 are connected to the first shift register 25. Unnecessary charges accumulated in the odd-numbered photodiodes 2 are drained by outputting reset pulses and readout pulses to the reset line 24 and readout line 25 in the row direction in order from the first shift register 25. Swept from. By outputting a read pulse from the first shift register 25 to the read line 25 in the row direction one row at a time, the signal charges accumulated in the photodiodes 2 in the odd rows represent the first image representing the odd row images from the signal lines 27. Are output as video signals.

  The second shift register 35 controls the accumulation and discharge of charges to the photodiodes 3 in even rows, and is connected to the reset line 34 and the readout line 35. Unnecessary charges accumulated in the photodiodes 3 in the even rows are drained by outputting reset pulses and readout pulses to the reset line 34 and the readout line 35 in the row direction in order from the second shift register 35. Swept from. By outputting a read pulse from the second shift register 35 to the read line 35 in the column direction one column at a time, the signal charges accumulated in the photodiodes 2 in the even rows represent the second row images representing the even row images from the signal lines 37. Are output as video signals.

  FIG. 4 shows the exposure time of the photodiode 2 or 3 of the C-MOS sensor 40.

  As described above, the signal charges accumulated in the photodiodes 2 in the odd-numbered rows are read in the row direction sequentially (every other row) from the first row. On the other hand, the signal charges accumulated in the photodiodes 2 in the even-numbered rows are read in the direction opposite to the row direction in order (every other row) from the bottom last row. The number of rows of the C-MOS sensor 40 is N + 1. To make it possible to distinguish between the exposure time for odd lines and the exposure time for even lines, the odd-numbered exposure time is marked with "odd" and the even-numbered exposure time is marked with "even". It is attached.

  When only the odd-numbered photodiodes 2 are considered, the exposure time for the photodiodes 2 in the first row is from time t1 to time t3. The photodiodes 2 in the third row are from time t2 to time t4. The exposure start timing of the photodiodes 2 in the first row is time t1, whereas the exposure start timing of the photodiodes 2 in the third row is time t2, which is higher than the time t1 that is the exposure start timing of the first row. It is shifted by time Δt. The exposure start timing of the photodiodes 2 in the fifth and subsequent rows is also shifted by a time Δt from the exposure start timing of the photodiodes 2 in the odd-numbered rows immediately before the respective rows. The exposure start timing of the odd-numbered photodiodes 2 in the N-th row is time t (N + 1), which is T1 = t (N + 1) −t1 from the time t that is the exposure start timing of the photodiodes 2 in the first row. The time is off.

  As described above, the signal charges are read from the photodiodes 3 in the even-numbered rows in the upward direction (the direction opposite to the row direction) from the last row. The photodiode 3 in the (N + 1) th row, which is the last row, has an exposure time from time t1 to time t3. The photodiodes 3 in the (N−1) th row are from time t2 to time t4. Similar to the readout of the signal charges of the odd-numbered photodiodes 3, the exposure start timing of the photodiodes 3 in each row is also shifted by the time Δt from the exposure start timing of the even-numbered photodiodes 3 that start exposure immediately before. The exposure start timing of the photodiodes 3 in the even-numbered rows of the second row is time t (N + 1), which is only T1 = t (N + 1) −t1 from the time t that is the exposure start timing of the photodiodes 3 in the N + 1-th row. The time is off.

  In this embodiment, an image (upper half image of one frame) represented by the signal charges accumulated in the upper half of the photodiodes 2 in all odd rows and the lower half of the photodiodes 3 in all even rows. The image represented by the accumulated signal charge (the lower half image of one frame) is combined to generate an image of one frame. That is, the signal charges accumulated in the photodiodes 2 from the first row to the N / 2 rows for the odd rows and the photodiodes 3 from the (N + 1) / 2 rows to the (N + 1) rows for the even rows. An image of one frame is generated from the accumulated signal charges. Time lag from time t1 which is the exposure start timing of the odd-numbered first row or even-numbered row to the (N + 1) -th row to the exposure start timing of odd-numbered row N / 2 or the even-numbered (N + 1) / 2-th row Becomes (T1) / 2 = t1-t (N / 2), which is relatively short. Therefore, rolling distortion is reduced. The lower half image of the odd row image and the upper half image of the even row image may be combined.

  FIG. 5 shows how image synthesis is performed.

  As described above, the upper half image 52 of the upper half image 52 and the lower half image 53 constituting the odd row image 51, and the upper half image 62 and the lower half of the even row image 61 are constituted. The synthesized image 70 is obtained by synthesizing the lower half image 63 of the image 63. The composite image 70 has reduced rolling distortion.

  FIG. 6 is a block diagram showing the electrical configuration of the digital camera.

  The overall operation of the digital camera is controlled by the control device 43.

  The digital camera includes the C-MOS sensor 40 described above. Reading of the C-MOS sensor 40 is controlled by a reading control device 41. In the imaging mode, the subject is imaged at a constant cycle by the C-MOS sensor 40, and so-called through images are obtained at a constant cycle. As described above, the C-MOS sensor 40 outputs the first video signal representing the odd-numbered image and the second video signal representing the even-numbered row image at a constant cycle. These first video signal and second video signal are input to the subject motion detection device 42. When the subject moves, the shape of the subject represented by the odd-numbered image and the shape of the subject represented by the even-numbered row image change. It can be seen that the subject has a movement because the shape of the subject represented by the odd image and the shape of the subject represented by the even row image are different. The subject motion detection device 42 includes an imaging device, and the subject is imaged twice by the imaging device. The direction of movement of the subject is detected from the two frames of the subject image obtained by the two imaging operations. Data indicating the amount and direction of movement output from the subject movement detection device 42 is input to the control device 43.

  The odd-numbered image is obtained based on the signal charges read out in order from the upper row to the lower row of the photodiodes 2 in the odd-numbered row as described above, whereas the even-numbered row image is obtained. Is obtained on the basis of the signal charges read out in order from the lower row to the upper row of the photodiodes 3 in the even rows. For this reason, when the subject moves, a shift occurs between the subject image represented by the odd-numbered row image and the subject image represented by the even-numbered row image. Whether there is a movement at a speed, movement speed, presence / absence of movement). For example, when an object moving in the horizontal direction is imaged, the odd-line image is more distorted in the horizontal direction in the lower image part, whereas the even-line image is distorted in the horizontal direction in the upper image part. Becomes larger. Thus, the amount of motion of the subject can be calculated by comparing the distorted portion or the non-distorted portion. Data representing the amount of movement is input to the control device 43.

  When a shutter release button (not shown) is pressed, the subject is imaged by the C-MOS sensor 40, and the first video signal representing the odd-numbered row image and the second image representing the even-numbered row image from the C-MOS sensor 40. Video signal is output. The first video signal and the second video signal output from the C-MOS sensor 40 are input to the image composition / selection device 44.

  The image composition / selection device 44 also receives an image composition / selection signal from the control device 43. When the movement of the subject is detected by the subject motion detection device 42, as described above, a composite image of one frame is generated from the upper half image of the odd row image and the lower half image of the even row image. The image composition / selection signal is input to the image composition / selection device 44. On the other hand, when the movement of the subject is not detected, the image composition / selection signal is such that either the first video signal or the second video signal (or both) is output. Is input to the image composition / selection device 44.

  The video signal output from the image composition / selection device 44 is subjected to predetermined signal processing in the signal processing device 45. The video signal (image data) output from the signal processing device 45 is recorded on the memory card 47 by the recording control device 46.

  FIG. 7 is a flowchart showing the processing procedure of the digital camera.

  When the shutter release button is pressed and the subject is imaged as described above, the signal charges are read in the row direction in order from the first row for the photodiodes 2 in the odd-numbered rows of the C-MOS sensor 40 (scanning). In the even-numbered photodiodes 3, signal charges are read from the last row to the upper row (step 81).

  If a fast moving subject (or simply a moving subject) is detected (YES in step 82), the upper half (or lower half) of the odd-numbered image and the lower half (upper half) of the even-numbered row image may be used. ) And the synthesized image are obtained (step 83). If a fast moving subject (or simply a moving subject) is not detected (NO in step 82), it is selected whether to record an odd row image or an even row image (step 84).

  The composite image or the selected odd row image or even row image is recorded on the memory card (step 85).

  8 to 10 show another embodiment.

  FIG. 8 shows the reading order of the C-MOS sensor 40.

  In this embodiment, the odd-numbered photodiodes 2 are read upward from the middle row N / 2, and when the first row of photodiodes 2 is read, the photodiodes 2 in the last row are read upward. Is read out. Further, the photodiodes 3 in the even-numbered rows are read from the middle row (N + 1) / 2 rows in the row direction (downward), and when the photodiodes 3 in the last row are read out, the photodiodes in the second row 3 is read out in the row direction.

  FIG. 9 shows the exposure time in the photodiode 2 or 3 of the C-MOS sensor 40. In this figure as well, as shown in FIG. 4, in order to distinguish between the exposure time for odd-numbered lines and the exposure time for even-numbered lines, the odd-numbered line exposure times are marked with “odd” characters. “Even” characters are added to the exposure time of even lines.

  When only the odd-numbered photodiodes 2 are considered, the exposure time for the photodiodes 2 in the N / 2-th line is from time t1 to time t3. The photodiodes 2 in the (N-2) / 2 rows are from time t2 to time t4. As described above, the exposure start timing of the photodiodes 2 in each row is shifted by the time Δt from the exposure start timing of the odd-numbered photodiodes 2 immediately before each row.

  The photodiodes 3 in the even-numbered rows are sequentially read in the row direction from the (N + 1) / 2th row which is the middle row. The exposure start timing of the photodiodes 3 in each row is also shifted by the time Δt from the exposure start timing of the photodiodes 3 in the even-numbered rows that start exposure immediately before.

  Also in this embodiment, the image (the upper half image of one frame) represented by the signal charges accumulated in the upper half of the photodiodes 2 in all odd rows and the lower half of the photodiodes 3 in all even rows. Are combined with the image (the lower half image of one frame) represented by the signal charge accumulated in the frame to generate an image of one frame. That is, the signal charges accumulated in the photodiodes 2 from the first row to the N / 2 rows for the odd rows and the photodiodes 3 from the (N + 1) / 2 rows to the (N + 1) rows for the even rows. An image of one frame is generated from the accumulated signal charges. As described above, a row for obtaining a composite image of one frame from time t1, which is the exposure start timing of the first row of the odd-numbered photodiodes 2 and the even-numbered photodiodes 3 (the first row for the odd-numbered rows, The time lag until the exposure start timing for the even-numbered row is (T1) / 2 = t1-t (N / 2), which is relatively short, as shown in FIG. Therefore, rolling distortion is reduced.

  FIG. 10 is a flowchart showing the processing procedure of the digital camera. In this figure, the same processes as those shown in FIG.

  When the shutter release button is pressed and the subject is imaged as described above, the signal charges are read (scanned) upward from the middle row of the photodiodes 2 of the odd-numbered rows of the C-MOS sensor 40. ), When the photodiodes 2 in the first row are read, the signal charges are read upward from the photodiodes 2 in the last row. For even-numbered photodiodes 3, signal charges are read in the row direction from the middle row, and when the last row of photodiodes 3 is read, the signal charges are sequentially read in the row direction from the photodiodes 3 in the second row. Is read (step 81A). In this way, an odd line image and an even line image are obtained.

  If a fast moving subject (or simply a moving subject) is detected (YES in step 82), the upper half (or lower half) of the odd-numbered image and the lower half (upper half) of the even-numbered row image may be used. ) And the synthesized image are obtained (step 83). If a fast moving subject (or simply a moving subject) is not detected (NO in step 82), it is selected whether to record an odd row image or an even row image (step 84).

  The composite image or the selected odd row image or even row image is recorded on the memory card (step 85).

  11 to 13 show another embodiment.

  FIG. 11 is a flowchart showing the processing procedure of the digital camera.

  When the imaging mode is set, the subject is imaged and a through image is obtained (step 91).

  If a fast moving subject (or simply a moving subject) is detected (YES in step 82), it is determined whether or not the traveling direction of the subject is the left direction (step 93). In the case of the left direction, the signal charges are read (scanned) in the row direction from the first row for the photodiodes 2 in the odd-numbered rows of the C-MOS sensor 40, and the last row for the photodiodes 3 in the even-numbered rows. The signal charge is set so as to be read out from the upper line toward the upper line (step 94). For the odd-numbered photodiodes 2 of the C-MOS sensor 40, signal charges are read (scanned) upward from the middle row, and when the photodiodes 2 in the first row are read out, the photodiodes 2 in the final row are read out. The signal charge is read upward from. For even-numbered photodiodes 3, signal charges are read in the row direction from the middle row, and when the last row of photodiodes 3 is read, the signal charges are sequentially read in the row direction from the photodiodes 3 in the second row. Is set to be read (step 95).

  The subject is imaged (step 96), the signal charge is read by the set readout method, and the upper half of the obtained odd-numbered row image and the lower half of the even-numbered row image are synthesized to obtain a synthesized image ( Step 97). Predetermined image processing is performed on the image data representing the obtained composite image and is recorded on the memory card (step 98).

  If an object that moves at high speed is not detected (NO in step 92), scanning is set to sequentially scan one line at a time (step 99). The subject is imaged, and predetermined image processing is performed on the obtained image data and recorded on the memory card (step 98). Image data representing either an odd-numbered image or an even-numbered row image may be recorded on a memory card, or an odd-numbered image and an even-numbered row image are synthesized to generate a high-quality one-frame synthesized image. Then, the image data representing the generated composite image may be recorded on the memory card.

  FIG. 12 is an example of a subject image in which the subject is moving in the left direction.

  When the rectangular object is moving in the left direction, as described above, the signal charges are read (scanned) in the row direction in order from the first row for the photodiodes 2 in the odd-numbered rows of the C-MOS sensor 40. For the photodiodes 3 in the even rows, the signal charges are read from the last row to the upper row, and the composite image 111 is obtained. The moving subject image 112 included in the composite image 111 has little distortion in the upper part and the lower part. On the other hand, if the reading order is not changed in the odd row image or the even row image, a large distortion occurs in the lower portion as shown in the image 113 indicated by a broken line.

  FIG. 13 is an example of subject increase in which the subject is moving in the right direction.

  When the rectangular object is moving in the right direction, the signal charges are read (scanned) upward from the middle row of the photodiodes 2 in the odd-numbered rows of the C-MOS sensor 40 as described above. When the photodiodes 2 in the first row are read out, signal charges are read upward from the photodiodes 2 in the last row to obtain an odd-numbered row image. For the photodiodes 3 in the even-numbered rows, the row direction from the middle row is obtained. When the signal charges are read out toward the first row and the photodiodes 3 in the last row are read out, the signal charges are read out in the row direction sequentially from the photodiodes 3 in the second row, and an even row image is obtained. A composite image 114 is obtained from these odd and even line images. The moving subject image 115 included in the composite image 114 has little distortion in the upper part and the lower part. On the other hand, if the reading order is not changed in the odd row image or the even row image, a large distortion occurs in the lower portion as shown in the image 116 indicated by a broken line.

2, 3 Photodiode (photoelectric conversion element)
40 C-MOS sensor
41 Read control device
42 Subject motion detection device (motion determination means, movement direction detection means)
43 Image composition / selection device R, G, B, r, g, b Color filter

Claims (5)

A photoelectric conversion element in which a color filter having a characteristic of transmitting light components of red, green or blue or cyan, magenta or yellow is formed on the light receiving surface is an odd-numbered row odd-numbered column and an even-numbered row even-numbered column or an odd-numbered even-numbered column and The characteristics of the color filters arranged in the even rows and the odd columns, and formed in the two photoelectric conversion elements obliquely in the 4n (n is an integer of 0 or more) +1 row and the 4n + 2 row, the 4n + 3 row and the 4n + 4 row C-MOS sensors having the same characteristics of the color filters formed in the two photoelectric conversion elements in the diagonal direction in the row,
The signal charges accumulated in the photoelectric conversion elements in the odd-numbered rows are sequentially read in the row direction from the mth row (m is a positive odd number) row in the row direction, and from the first row to the first row in response to the end of reading of the last row Read sequentially in the row direction, or read from the m-th row one by one in the opposite direction to the row direction, and in response to the end of reading the first row, one row from the last row in the opposite direction to the row direction And the odd-numbered row signal charge readout control means for controlling the C-MOS sensor so as to output it as an odd-numbered video signal, and the signal charges accumulated in the odd-numbered photoelectric conversion elements one by one from the m-th row. When read in the row direction in order, the signal charges accumulated in the photoelectric conversion elements in the even-numbered rows are read one by one from the m + 1th row in the direction opposite to the row direction, and the reading of the first row is completed. Possibly from the last line When the signal charges stored in the odd-numbered photoelectric conversion elements are read one by one from the m-th row in the reverse direction to the row direction, one row at a time The signal charges accumulated in the photoelectric conversion elements in the row are read out in the row direction one by one from the (m + 1) th row, and are read in the forward direction one by one from the first row in response to the end of reading out the last row. Even row signal charge readout control means for controlling the C-MOS sensor to output as a row video signal;
A solid-state electronic imaging device. A motion determining means for determining whether or not the subject has a motion of a predetermined speed or more, and the odd row signal in response to the motion determining means determining that the subject has a motion of a predetermined speed or more. The image of the upper half or the lower half of the odd row image represented by the odd row video signal obtained under the control of the charge readout control means and the even row obtained under the control of the even row signal charge readout control means. Image generating means for generating an image of one frame from the lower half image or the upper half image represented by the video signal;
The solid-state electronic imaging device according to claim 1, further comprising: A moving direction detecting means for detecting the moving direction of the subject;
In response to detecting that the subject is moving in the left direction toward the moving direction detecting means, the odd-numbered row signal charge reading control means is configured to detect the photoelectric conversion elements in the first row or the photoelectric conversion elements in the last row. The C-MOS sensor is controlled so as to read the signal charge from the conversion element.
The solid-state electronic imaging device according to claim 2. When the moving direction detecting means detects that the subject is moving in a direction other than the left direction, the odd-numbered row signal charge reading control means is configured to detect the photoelectric conversion element of the C-MOS sensor. The signal charge is read from the photoelectric conversion elements in half of all the rows of
The solid-state electronic imaging device according to claim 3. A photoelectric conversion element in which a color filter having a characteristic of transmitting light components of red, green or blue or cyan, magenta or yellow is formed on the light receiving surface is an odd-numbered row odd-numbered column and an even-numbered row even-numbered column or an odd-numbered even-numbered column and The characteristics of the color filters arranged in the even rows and the odd columns, and formed in the two photoelectric conversion elements obliquely in the 4n (n is an integer of 0 or more) +1 row and the 4n + 2 row, the 4n + 3 row and the 4n + 4 row In an operation control method for a solid-state electronic imaging device including a C-MOS sensor having the same characteristics of color filters formed in two photoelectric conversion elements in an oblique direction in a row,
The odd-numbered signal charge read control means reads the signal charges accumulated in the photoelectric conversion elements in the odd-numbered rows one by one from the m-th (m is a positive odd number) row in the row direction, and the reading of the last row is completed. In response to the first row, reading is performed in the row direction one row at a time, or one row from the mth row is read in the direction opposite to the row direction, and the first row is read in response to the end of reading the first row. The C-MOS sensor is controlled so as to read out each row in the direction opposite to the row direction and output as an odd row video signal,
The even-numbered signal charge reading control means accumulates in the even-numbered photoelectric conversion elements when the signal charges accumulated in the odd-numbered photoelectric conversion elements are read in the row direction one by one from the mth row. The read signal charges are read one by one from the (m + 1) th row in the direction opposite to the row direction, and read from the last row one by one from the last row in the direction opposite to the row direction in response to the completion of the reading of the first row, When the signal charges accumulated in the odd-numbered photoelectric conversion elements are read one by one from the m-th row in the direction opposite to the row direction, the signal charges accumulated in the even-numbered photoelectric conversion elements are The C-MOS sensor reads out line by line from the (m + 1) th line in order in the row direction, and reads out in the forward direction line by line from the first line in response to the end of reading out the last line and outputs it as an even-numbered video signal. Control,
Operation control method of solid-state electronic imaging device.

Images