JP7381067B2 - Imaging device and imaging device - Google Patents

Description

The present invention relates to an imaging device and an imaging device having phase difference detection pixels.

In recent years, a technique has become common in which phase difference detection pixels are arranged in place of some of the normal pixels on an image sensor, and the focus position of a subject is obtained using the phase difference detection signals outputted by these pixels. Here, focusing on the so-called Bayer type image sensor which is widely used as an image sensor, it has a feature that pixels provided with color filters of the same color are separated by one pixel in the horizontal direction.

Therefore, when arranging a pair of phase difference detection pixels, it is desirable to arrange each pixel at the same color filter position with the same sensitivity characteristics, but in the case of a Bayer type image sensor, the pixels that make up the pixel pair are There was a problem in that the accuracy of the phase difference detection signal deteriorated due to the separation.

Furthermore, a method is known that aims to obtain high phase difference resolution in the horizontal direction by arranging phase difference detection pixel pairs consecutively in the same row. However, since phase difference detection pixels are treated as defective pixels when generating a still image, if they are consecutively arranged in the same row, the number of normal pixels that can be referenced when correcting defective pixels decreases. As a result, there was a problem in that the image quality of the phase difference detection pixel row in a still image deteriorated.

Techniques for solving such problems have been proposed in the past. For example, the invention disclosed in Patent Document 1 is an imaging device in which a pixel serving as a photoelectric conversion element is arranged at each square lattice position where a plurality of horizontal lines and a plurality of vertical lines intersect. Within a predetermined area where the pixels of the element are arranged, a pair of pixels XY, which are first and second phase difference detection pixels arranged adjacently for detecting a phase difference among the pixels, are connected to one of the above lines. When arranging a plurality of paired pixels as a unit within a line (first line) and a line parallel to the first line (second line), a plurality of paired pixels are placed within the first line with at least two pixel separations. It is characterized in that the paired pixels are arranged in pairs, and the paired pixels are arranged in the second line at positions corresponding to positions spaced apart on the first line.

According to this invention, since a plurality of phase difference detection pixel pairs are not consecutively arranged on the same line, the imaging pixel signal at the phase difference detection pixel position is used as the imaging pixel signal of the normal pixel adjacent to the phase difference detection pixel. It is possible to perform pixel interpolation using image signals, which makes it possible to capture high-quality subject images.

On the other hand, in the case of phase difference detection pixels, it is particularly important for a light beam to be correctly incident on each pixel forming a pair in order to achieve high AF accuracy.

One of the reasons why the incident light ray does not correctly reach the light receiving section in the pixel is that the angle of the incident light ray to the image sensor is too large. If the incident angle of the light ray is too large, the incident light ray will not be refracted to an appropriate angle even if the microlens placed above each pixel is used, and a portion or most of the incident light will be reflected by structures inside the pixel other than the light receiving area. It will be incident on.

As a countermeasure against vignetting caused by such light rays having a large incident angle, it is widely practiced to shift the microlens according to the increase in the number of pixels. By shifting the microlens in this manner, it becomes possible to focus the light rays, which would otherwise be vignetted, onto the light receiving section.

Patent No. 5589146

However, in recent years, with the miniaturization of digital cameras, large-sized image sensors, so-called full-size, are increasingly being adopted. In digital cameras with interchangeable lenses equipped with such large image sensors, it is clear that the angle of the light rays incident on the imaging surface has a wider range than previously assumed, depending on the type of interchangeable lens attached. It has become. The wide range of incident angles to be accommodated is particularly noticeable at the periphery of the imaging surface.

Furthermore, the above-mentioned microlens shift alone has a limit in dealing with such a wide range of incident angles.

The present invention has been made in view of these circumstances, and provides an image sensor and image sensor that improves the interpolation accuracy of photographed image signals while maintaining the resolution in the phase difference detection direction, and that is capable of handling light rays with a wide range of incident angles. The purpose is to provide equipment.

In order to achieve the above object, an image sensor embodying the present invention has a plurality of phase difference detection pixels each including a plurality of normal pixels and a set of phase difference detection pixels each having a light receiving surface eccentric in opposite directions. The first line has a plurality of lines consisting of a pair and a first phase difference detection pixel pair, and is configured such that the light receiving area of each phase difference detection pixel pair as a whole is equal to that of the normal pixel, and the first line has a first phase difference between the normal pixel and the first line. The second line consists of the normal pixel and the second phase difference detection pixel pair, is parallel to the first line, and the first line and the second line are different from each other. The first phase difference detection pixel pair and the second phase difference detection pixel pair are arranged alternately with a predetermined pixel separation in the direction orthogonal to the line, and the first phase difference detection pixel pair and the second phase difference detection pixel pair are in a positional relationship that does not overlap in the direction orthogonal to the line. The first phase difference detection pixel pair and the second phase difference detection pixel pair are arranged so that the two phase difference detection pixels constituting each phase difference detection pixel pair are arranged according to their respective in-line positions. The second phase difference detection pixel pair is characterized in that the upper and lower limits of the light reception area ratio are different from those of the first phase difference detection pixel pair.

Further, in the image sensor according to the present invention, the difference in the light-receiving area ratio of the first phase difference detection pixel pair increases as the in-line position approaches the periphery of the imaging surface, and the in-line position approaches the center of the imaging surface. It is characterized in that the difference in the light-receiving area ratio becomes smaller as it approaches .

The image sensor according to the present invention is further characterized in that the phase difference detection pixel includes a light-shielding mask, and the light-receiving area is an opening area of an opening formed by the light-shielding mask.

Further, in an imaging device including a photographing lens and the above-mentioned image sensor, pixels used for phase difference detection processing are selected from the first line and the second line based on characteristic information of the photographing lens. The present invention is characterized in that it further includes means for selecting a line from which a signal is read.

According to the image pickup device and the image pickup device embodying the present invention, it is possible to improve the interpolation accuracy of the photographed image signal while maintaining the resolution in the phase difference detection direction, and to cope with light rays having a wide range of incident angles. Furthermore, the amount of data read during phase difference detection can be reduced, and processing speed can be increased.

1 is a block diagram showing the main configuration of an imaging device that is an embodiment of the present invention. FIG. 2 is a conceptual diagram of a vertical color separation type image sensor. FIG. 3 is a schematic top view of an imaging surface. FIG. 3 is a schematic cross-sectional diagram for explaining the internal structure of a pixel pair. 7 is a graph illustrating a change in aperture ratio caused by a light-shielding mask shift. FIG. 7 is a schematic top view showing an example of the shape of the light-shielding mask of the pixel pair of the second line LN2 and the opening formed by the light-shielding mask. 7 is a flowchart showing an example of line selection processing. 12 is a flowchart showing another example of line selection processing.

DESCRIPTION OF THE PREFERRED EMBODIMENTS The best mode for carrying out the present invention will be described below with reference to the accompanying drawings. Note that the present invention is not limited to this embodiment.

FIG. 1 is a block diagram showing the main configuration of an imaging device 100 that is an embodiment of the present invention. The imaging device 100 includes a camera body 200 and a photographing lens 300 that is an interchangeable lens that can be attached to and detached from the camera body 200.

The camera body 200 includes an image sensor 210, a signal processing section 230, an image processing section 240, a main CPU 250, a ROM 251, a RAM 252, a recording medium interface (I/F) 260, and a user interface (I/F). 270 and an image display section 280.

Further, the photographing lens 300 includes a zoom control section 310, a zoom lens 311, a focus control section 320, a focus lens 321, an aperture control section 330, an aperture unit 331, and a lens CPU 340.

The image sensor 210 receives the light beam focused by the photographic lens 300, performs photoelectric conversion, and outputs an image signal. The light receiving surface of the image sensor 210 is composed of a large number of pixels. As the image sensor 210 of this embodiment, a so-called three-layer vertical color separation type image sensor is used. This type of image sensor is disclosed in, for example, Japanese Patent No. 5201776.

The image sensor 210 is internally equipped with a variable gain amplifier and an A/D converter (not shown), and the image signal is output as digital data. When employing the image sensor 210 that does not include a variable gain amplifier or an A/D converter, these devices may be mounted individually.

The signal processing unit 230 performs various signal processing on the image signal read out from the image sensor 210. Examples of signal processing include correction processing to ensure linearity between the amount of light received by a pixel and an output value, and amplification processing to amplify a read image signal.

The image processing section 240 performs various image processing on the image signal sent from the signal processing section 230. Examples of image processing include white balance processing, color reproduction processing, and development processing for image data in JPEG format or TIFF format.

The main CPU 250 performs comprehensive control of the entire imaging apparatus 100 by executing programs. For example, readout control of the image sensor 210 is performed. The main CPU 250 outputs a signal that determines the drive timing of the image sensor 210, thereby controlling horizontal drive and vertical drive for each pixel, and reading RGB image signals from each pixel.

The main CPU 250 is electrically connected to the lens CPU 340 and controls the photographic lens 300 in cooperation with the main CPU 250 . During live view image acquisition or normal photography, various information is acquired from the lens CPU 340 and used to determine photography conditions and the like.

The ROM 251 stores various programs executed by the main CPU 250 and various characteristic information and setting information used by the main CPU 250 to execute processing. The characteristic information includes, for example, the spectral sensitivity characteristic of the image sensor 210.

The RAM 252 is used as a work area when the main CPU 250 executes processing according to a program.

The recording medium I/F 260 records or reads RAW data and developed image data from a recording medium (not shown). This recording medium is a removable recording medium such as a semiconductor memory.

The user I/F 270 has operating members such as a release button, a power button, a command dial, a cross key, etc., and when the user operates these operating members, the main CPU 250 issues an instruction to perform a predetermined operation. .

The image display unit 280 displays photographed images, images read from a recording medium (not shown), and the like.

A zoom lens 311 and a focus lens 321, which are part of the optical system of the photographic lens 300, are connected to a zoom control section 310 and a focus control section 320, respectively, and control such as lens driving and position detection is performed. Although the zoom lens 311 and the focus lens 321 are each depicted as one lens in the drawing for simplicity, the invention is not limited to this. Alternatively, a single focus lens without the zoom lens 311 may be used.

The aperture unit 331 is connected to the aperture control section 330. The aperture control section 330 controls the aperture value (F value) of the aperture unit 331.

The lens CPU 340 cooperates with the above-mentioned main CPU 250 to determine the control contents of the various control units and issues instructions. Further, the lens CPU 340 acquires various information such as the zoom position, focus position, and F value of the photographic lens 300 from each control section, and outputs the information to the main CPU 250 as appropriate.

Further, the lens CPU 340 is connected to a ROM (not shown), and reads the stored lens ID, principal ray angle information, correction data used in various correction processes, etc. as necessary, and outputs the readout to the main CPU 250 as appropriate. These pieces of information are output, for example, at the timing when the photographing lens 300 is attached to the camera body 200.

FIG. 2 is a conceptual diagram for explaining the vertical color separation structure of the imaging element 210 mounted in the imaging device 100 of the present embodiment described above. A so-called three-layer vertical color separation type image sensor is simply configured by three photodiodes that function as photoelectric conversion layers stacked in the depth direction. A microlens ML is provided in the upper layer of the photodiode stacked in three layers to improve the condensing efficiency of incident light.

When light enters a certain pixel, the blue (B) component in the incident light is photoelectrically converted mainly by the B photodiode PD_B located in the top layer. Similarly, the green (G) component in the incident light is mainly photoelectrically converted by the G photodiode PD_G located in the middle layer, and the red (R) component is mainly photoelectrically converted by the R photodiode PD_R located in the bottom layer. Ru.

These vertical color separations utilize the characteristics of silicon (Si) used as a material for photodiodes. As a result, the vertical color separation type image sensor does not require a color filter, which is essential to the Bayer type image sensor, and it is possible to obtain RGB color component signals with one pixel. There is also the advantage that there is no need to perform pixel interpolation because each pixel can photoelectrically convert the wavelength components of all three colors.

Note that instead of the photodiode, a plurality of photoelectric conversion films configured to have specific absorption characteristics using an organic substance, an inorganic substance, or the like may be laminated as a photoelectric conversion layer.

FIG. 3 is a schematic diagram of the image sensor 210 when viewed from the subject side, FIG. 3a is a schematic top view of the entire imaging surface of the image sensor 210, and FIG. 3b is an enlarged top view of a portion surrounded by a broken line in FIG. 3a. It is a schematic diagram.

Each rectangle in FIG. 3b represents a pixel. White rectangles are usually pixels, and are mainly used to generate signals for captured images. Dark-colored rectangles are phase difference detection pixels, which are mainly used to generate signals for phase difference detection. Each pixel has the above-mentioned vertical color separation type pixel structure.

As shown in this figure, pairs of adjacent phase difference detection pixels are distributed in a predetermined pattern among normal pixels arranged two-dimensionally.

This pair of phase difference detection pixels will be referred to as a phase difference detection pixel pair or simply a pixel pair. Further, among a pair of phase difference detection pixels constituting a pixel pair, the pixel located on the -X side of the imaging surface is also called an L pixel, and the pixel located on the +X side of the imaging surface is also called an R pixel.

The apertures of the pair of phase difference detection pixels constituting the pixel pair are limited by light-shielding masks that are eccentric in opposite directions.

On the imaging surface of the image sensor 210, a plurality of pixel pairs belong to either the first line LN1 or the second line LN2. Both of these lines are parallel to the X direction of the imaging surface of the image sensor 210, and the first line LN1 and the second line LN2 are regularly arranged at a predetermined interval in the Y direction perpendicular to the line. has been done. In this figure, two each of the first line LN1 and the second line LN2 are shown.

In addition, within each line, pixel pairs are regularly arranged at predetermined intervals, and the pixel pairs within each line are such that the adjacent direction of the two phase difference detection pixels forming the pair is the same as the direction in which each line extends. It is parallel to the X direction.

Although details will be described later, all the pixel pairs in the first line LN1 have the same light-shielding mask. Further, the pixel pairs in the second line LN2 are configured such that the size of the light-shielding mask differs depending on the position of the pixel pair in the X direction.

Pixel pairs in each line are arranged so as not to overlap each other in the Y direction orthogonal to the line. That is, the pixel pairs in the second line LN2 are arranged at positions corresponding to gaps formed by separating the pixel pairs in the first line LN1.

This has the advantage of being more resistant to image quality deterioration due to pixel interpolation than arranging phase difference detection pixels in a single row.

Hereinafter, the pixel pair belonging to the first line LN1 will also be referred to as a first pixel pair, and the pixel pair belonging to the second line LN2 will also be referred to as a second pixel pair. The first pixel pair 211 and the second pixel pair 221 each refer to a specific pixel pair located on the -X side in FIG. 3b.

As described above, the main CPU 250 performs signal value readout control for each pixel including the phase difference detection pixels arranged in this manner. When a certain row is selected by a horizontal drive signal, signal values are read out from each pixel in the selected row, for example, in the order of B photodiode PD_B, G photodiode PD_G, and R photodiode PD_R.

If the selected row is a row consisting only of regular pixels, the main CPU 250 generates an image signal using the RGB signal values read from each pixel.

On the other hand, when the selected row is a row including phase difference detection pixels, that is, either the first line LN1 or the second line LN2 described above, the main CPU 250 reads out the data from each phase difference detection pixel in the row. A phase difference detection signal is generated using the RGB signal values. Note that the phase difference detection signal may be generated using the signal value read from the photodiode of any one of the RGB layers.

Next, the main CPU 250 generates phase difference information from a pair of phase difference detection signals generated by a pair of pixels on the same line. That is, phase difference information is generated by performing a correlation calculation based on a phase difference detection signal obtained only from L pixels and a phase difference detection signal obtained only from R pixels of a pair of pixels on the same line. . Then, by using the obtained phase difference information, the focus control unit 320 performs focus adjustment processing of the focus lens 321.

In general, when the photographic lenses are the same, the angle of incidence of the incident light ray becomes steeper as the peripheral portion of the image sensor increases. Therefore, the phase difference detection pixels located in the peripheral region as shown in FIG. 3b need to be able to correctly receive light rays with a steep angle of incidence. In the following, a pixel structure of the present invention for achieving this will be explained.

FIG. 4 is a conceptual diagram for explaining the internal structure of the above-mentioned pixel pair. FIG. 4a shows a cross section of the first pixel pair 211 in FIG. 3b, and FIG. 4b shows a cross section of the second pixel pair 211 in FIG. 3b. 221 is shown.

The first pixel pair 211 shown in FIG. 4a is composed of an L pixel 211L and an R pixel 211R. Simply put, the L pixel 211L includes, in order from the incident side, a microlens ML, a light shielding mask 212L, and a photodiode PD. In addition, it includes a metal wiring layer (not shown) and the like. Similarly, the R pixel 211R includes a microlens ML, a light shielding mask 212R, a photodiode PD, and a metal wiring layer (not shown).

As can be seen from this figure, in the first pixel pair 211 belonging to the first line LN1, the light shielding masks 212 of each phase difference detection pixel are line symmetrical and have the same shape, and the right half of the L pixel 211L is The R pixel 211R has a left half opening 213L that is light-shielded and has a right half opening 213R that is light-shielded.

As described above, all the first pixel pairs belonging to the first line LN1 have an aperture ratio of 5:5 as shown in the figure. Therefore, hereinafter, this first pixel pair having an aperture ratio of 5:5 may also be referred to as a 5:5 pair.

In the L pixel 211L and the R pixel 211R, the exit pupil of the photographing lens 300 is divided by the microlens ML and the light-shielding mask 212. That is, the L pixel 211L receives only the light beam in the -X direction of the exit pupil of the photographing lens 300 through the light-shielding mask 212L, which is eccentrically formed in the +X direction, and the R pixel 211R is eccentrically formed in the -X direction. The light-shielding mask 212R formed as shown in FIG.

The second pixel pair 221 shown in FIG. 4b is similarly composed of an L pixel 221L and an R pixel 221R. The internal structure of each is almost the same as the first pixel pair 211 described above, and each has a microlens ML, a light-shielding mask 222, a photodiode PD, and a metal wiring layer (not shown). .

Compared to the first pixel pair 211, the openings 223L and 223R in the second pixel pair 221 have different sizes. That is, in the L pixel 221L, the light-shielding mask 222L is shifted in the protruding direction (-X direction), and as a result, the opening 223L becomes smaller. Further, in the R pixel 221R, the light shielding mask 222R is shifted in the retracting direction (+X direction), and as a result, the opening 223R is enlarged.

As will be described later, in the second pixel pair, the amount of shift of the light shielding mask is changed depending on the position on the second line LN2. Therefore, hereinafter, the second pixel pair may also be referred to as a shift pair.

Next, the light rays incident on each pixel pair 211, 221 in FIG. 3b will be explained using FIG. 4. As described above, the first pixel pair 211 and the second pixel pair 221 in this figure are located at the end of the imaging surface of the image sensor 210 in the -X direction, and therefore the incident angle of the light beam incident on the pixel is It is tighter than the central area of the surface.

When the light rays are incident at a severe angle, as shown in FIG. 4a, the light rays incident on the L pixel 211L of the first pixel pair 211, which is a 5:5 pair, are not correctly blocked in half, and one half of the light rays are not properly blocked. The light beams incident on the R pixel 211R are all vignetted by the light shielding mask. Therefore, in this first pixel pair 211, the amounts of light received by the left and right pixels become unbalanced, making the signal inappropriate as a phase difference detection signal. Similarly, other 5:5 pairs located in the peripheral area cannot generate appropriate phase difference detection signals, so the accuracy of focus detection using the first line LN1 is greatly reduced.

On the other hand, in the second pixel pair 221 adjacent to the first pixel pair 211, as shown in FIG. The light beam incident on the photodiode PD is correctly blocked by both the pixel 221L and the R pixel 221R, and the balance of the amount of light received by both is appropriate. Therefore, since a phase difference detection signal suitable for focus detection can be generated, the accuracy of focus detection is greatly improved compared to a 5:5 pair.

However, in a region where the angle of incidence is relatively gentle near the center of the imaging surface, if the shift amount of the light shielding mask 222 of the second pixel pair 221 shown in FIG. There is a problem that light blocking occurs. In such a region where the incident angle is relatively gentle, a configuration such as the conventional 5:5 pair can block the incident light in a better balance.

Therefore, in the image sensor 210 implementing the present invention, the shift amount of the light-shielding mask in the shift pair of the second line LN2 is changed in accordance with the horizontal position of the shift pair. This makes it possible to prevent the error case of unbalance of the incident light beam due to inappropriate vignetting, as in the case of the pixel pair in FIG. The range of angles increases.

FIG. 5 is a graph illustrating the change in aperture ratio caused by the light-shielding mask shift according to the horizontal position of the shift pair described above, and shows the relationship between the pixel position in the horizontal direction (X direction) and the aperture ratio of the light-shielding mask. There is. The solid line indicates the aperture ratio of the L pixel of the shift pair belonging to the second line LN2, and the broken line indicates the aperture ratio of the R pixel of the shift pair belonging to the second line LN2.

For reference, this figure also includes a graph showing the aperture ratio of the 5:5 pair belonging to the first line LN1. As described above, in the first line LN1, all the pixel pairs have the same aperture ratio of 5:5, so the aperture ratio is always represented as one horizontal graph showing 50% in this figure.

Note that, although the pixel pairs are actually arranged discretely on each line at predetermined intervals, they are shown by lines in this figure for convenience.

As shown in this figure, in the second line LN2 of the image sensor 210 of this embodiment, the aperture ratio of the L pixel and the R pixel changes at a predetermined rate depending on the position of the pixel pair in the X direction. Specifically, the L pixel is configured such that the aperture ratio changes linearly from 40% at the left end in the X direction to 60% at the right end in the X direction. On the other hand, the R pixel is configured to change the aperture ratio from 60% to 40%.

In this way, in the second line LN2, the light-shielding mask is shifted according to the horizontal position of the pixel pair, and in the peripheral area where the angle of incidence is tight, the aperture ratio of the inner pixel of the pixel pair, which is more affected by the angle of incidence, is changed. By increasing the angle of incidence while approaching the conventional aperture ratio of 5:5 in the central region where the angle of incidence is gentle, it becomes possible to appropriately guide light rays with a wide range of incidence angles to the photodiode PD.

FIG. 6 is a schematic top view showing an example of the shape of the light shielding mask of the pixel pair of the second line LN2 and the opening formed thereby, and FIG. 6a is a top view of the pixel pair located near the left end in the X direction. , FIG. 6b shows a top view of a pixel pair located near the right end in the X direction. As explained in FIG. 5, the aperture ratio changes from 4:6 to 6:4 in the second line LN2, and in FIG. 6a, the aperture ratio of the pixel pair is 4:6 near the left end in the X direction. FIG. 6b shows a situation where the aperture ratio of the pixel pair is 6:4 near the right end in the X direction.

As described above, the change in the aperture ratio in the shifted pair of the second line LN2 is achieved by changing the shape of the light shielding mask. That is, in the L pixel, as the light shielding mask moves from the left end in the X direction (FIG. 6a) to the right end in the The opening changes in the direction of becoming larger. Similarly, for the R pixel, the light-shielding mask is gradually shifted in the direction to cover the opening (+X direction), and as a result, the opening of the pixel changes in the direction of becoming smaller.

Note that although the opening of the phase difference detection pixel has a rectangular shape in this figure, it is not limited to this.

As described above, in order to appropriately acquire phase difference detection information from light rays with a wide range of incident angles, the image sensor 210 according to the present embodiment uses two types of lines each including a pair of phase difference detection pixels with different aperture ratios. We are prepared. Here, by appropriately selecting these lines, the amount of data read for phase difference detection processing can be reduced, and as a result, the phase difference detection processing can be speeded up.

As a specific example, the principal ray angle information of the photographing lens 300 attached to the camera body 200 is used to select a line where incident light vignetting is less likely to occur between the first line and the second line of the image sensor 210. Let's read it out. This makes it possible to reduce the amount of data read while suppressing a decrease in phase difference detection accuracy, leading to faster phase difference detection processing.

This line selection will be explained in detail below. In the following embodiments, principal ray angle information of the photographing lens 300 is used as a condition for branching the flowchart, but the information is not limited to this, and other lens characteristic information, such as the average value of the upper ray angle and the lower ray angle, may be used. is available.

<Example 1>
FIG. 7 is a flowchart showing an example of line selection processing using chief ray angle information of the photographic lens 300.

First, in step S101, the main CPU 250 acquires principal ray angle information from the lens CPU 340 of the attached photographic lens 300. The timing for acquiring the principal ray angle information may be, for example, when the photographic lens 300 is attached or when the camera body 200 is turned on, but is not limited to these timings.

Next, in step S102, the main CPU 250 determines whether the acquired chief ray angle is less than a preset threshold. Here, if the principal ray angle is less than the threshold value, the process proceeds to step S103, and if the principal ray angle is greater than or equal to the threshold value, the process proceeds to step S104.

In step S103, the main CPU 250 selects a first line having a 5:5 pair as a line used for phase difference detection processing.

On the other hand, in step S104, the main CPU 250 selects the second line having the shift pair as the line used for the phase difference detection process.

According to the above flowchart, lines of pixel pairs used in the subsequent phase difference detection process are determined. Although not described in this flowchart, after determining the readout line, it is detected that the photographer presses the release button halfway, and when this is detected, known focus detection is performed using the line selected in this flowchart. . As a result, focus detection suitable for the principal ray angle of the photographic lens can be performed, and the number of lines to be read out can be reduced, so that calculation processing related to focus detection can be speeded up. Then, when it is further detected that the photographer fully presses the release button, known focus adjustment processing is performed, and then imaging processing is performed.

<Example 2>
Next, another example regarding selection of lines used for phase difference detection processing will be described. FIG. 8 is a flowchart showing another example of line selection processing using principal ray angle information of the photographic lens 300.

First, in step S201, the main CPU 250 determines whether the distance measurement frame is set within a predetermined central area. Specifically, the main CPU 250 detects whether the distance measurement frame is moved by the photographer's operation of the user I/F 270, and if the distance measurement frame is moved, the main CPU 250 determines whether the distance measurement frame is located within a predetermined central area. Determine whether or not. Here, if the ranging frame is set within the central area, the process advances to step S202, and if the ranging frame is set outside the central area, the process advances to step S203.

Then, in step S202, the main CPU 250 selects both the first line and the second line as lines to be used in the phase difference detection process.

On the other hand, in step S203, the main CPU 250 acquires principal ray angle information from the lens CPU 340 of the attached photographic lens 300. Since each subsequent step is the same as in the first embodiment described above, a description thereof will be omitted.

From the above, in the second embodiment, a branch based on the distance measuring frame position is inserted before the branch using the chief ray angle information of the first embodiment described above, and a case where both types of lines are used is added. In other words, if the distance measurement frame is set within the central area, it can be assumed that the angle of incidence of the ray at the distance measurement frame position is gentle, so by using both the first and second lines, high accuracy can be achieved. It becomes possible to perform focus detection.

On the other hand, if the distance measurement frame is not set within the central area, branching using principal ray angle information similar to the first embodiment described above is performed to select a line to be used for phase difference detection processing. The amount of data read out for phase difference detection can be reduced, making it possible to speed up arithmetic processing, etc.

Note that, as shown in FIG. 6, an example has been shown in which the aperture area is defined and the aperture ratio is changed using a light-shielding mask in the phase difference detection pixel described above, but the present invention is not limited to this. For example, the light receiving area may be changed according to the position of the phase difference detection pixel pair without using a light-shielding mask by manufacturing the PD with a different shape within the phase difference detection pixel. In this case, the aperture ratio will be replaced by the light-receiving area ratio. Alternatively, an organic photoelectric conversion film may be substituted for the PD, and the area of the conversion film itself or the opening area of the transparent electrode may be changed.

Further, as shown in FIG. 3, although the above-described line including the phase difference detection pixels is provided in the X direction of the image sensor 210, the line is not limited to this. For example, the imaging surface of the image sensor 210 may be divided into a plurality of regions, and lines may be provided in the X direction in some regions, and lines may be provided in the Y direction in other regions. In this case, in each region, each pixel pair of the first and second lines is arranged so as not to overlap in a direction perpendicular to the lines.

Furthermore, as shown in FIG. 4, the phase difference detection pixels described above are configured such that the central axes of each pixel and the microlens ML coincide with each other. However, it is also possible to perform a known so-called microlens shift on this microlens ML, and shift the central axis of the microlens ML in accordance with the increase in height of the pixel from the center of the imaging surface. In this configuration, the principal ray angle thresholds of the branches (steps S102 and S204) in the flowcharts shown in FIGS. 7 and 8 need to be appropriately changed.

Further, in the embodiments described above, the first line LN1 including a 5:5 pair as a phase difference detection pixel pair and the second line LN2 including a shift pair are provided, but these lines have pixels with different configurations. You can also arrange pairs. For example, the light-shielding mask is also shifted for the pixel pair of the first line LN1, and the aperture ratio is changed from 3:7 to 7:3, which is different from that of the second line LN2. This makes it possible to accommodate a wider range of incident angles, which may be more suitable for cases where, for example, the same imaging element is incorporated into imaging devices with different mount systems.

As explained above, according to the image sensor and the imaging device according to the present invention, by changing the aperture ratio according to the position of the phase difference detection pixel pair, light rays with a wide range of incident angles can be correctly blocked and PD It becomes possible to lead to As a result, focus detection accuracy is greatly improved, especially for light rays with a steep angle of incidence. Furthermore, by providing multiple lines with phase difference detection pixel pairs with different rates of change in aperture ratio, it is possible to select the line to be read out during phase difference detection, reducing the amount of data and speeding up processing. .

100 imaging device, 200 camera body, 210 imaging element, PD_B B photodiode, PD_G G photodiode, PD_R R photodiode, LN1 first line, 211 first pixel pair, 211L L pixel, 212L light shielding mask, 213L aperture part, 211R R pixel, 212R light shielding mask, 213R opening, LN2 second line, 221 second pixel pair, 221L L pixel, 222L light shielding mask, 223L opening, 221R R pixel, 222R light shielding mask, 223R opening , 230 signal processing unit, 240 image processing unit, 250 main CPU, 251 ROM, 252 RAM, 260 recording medium interface (I/F), 270 user interface (I/F), 280 image display unit, 300 photographing lens, 310 zoom control section, 311 zoom lens, 320 focus control section, 321 focus lens, 330 aperture control section, 331 aperture unit, 340 lens CPU

Claims (4)

multiple normal pixels,
a plurality of phase difference detection pixel pairs each including a pair of phase difference detection pixels each having a light receiving surface eccentric in opposite directions;
In an image sensor having a plurality of lines consisting of
The light-receiving area of each of the phase difference detection pixel pairs as a whole is configured to be equal to that of the normal pixel,
The first line consists of the normal pixel and a first phase difference detection pixel pair,
The second line consists of the normal pixel and a second phase difference detection pixel pair, and is parallel to the first line,
The first line and the second line are alternately arranged with a predetermined pixel separation in a direction perpendicular to the line,
The first phase difference detection pixel pair and the second phase difference detection pixel pair are arranged in a positional relationship that does not overlap in a direction perpendicular to the line,
The first phase difference detection pixel pair and the second phase difference detection pixel pair detect light reception between the two phase difference detection pixels constituting each phase difference detection pixel pair, depending on their respective in -line positions. The area ratio changes,
The second phase difference detection pixel pair has an upper limit and a lower limit of the light receiving area ratio different from those of the first phase difference detection pixel pair.
An image sensor with this feature. In the first phase difference detection pixel pair, the difference in the light-receiving area ratio becomes larger as the in-line position approaches the periphery of the imaging surface, and the difference in the light-receiving area ratio becomes smaller as the in-line position approaches the center of the imaging surface. The image sensor according to claim 1, characterized in that: The phase difference detection pixel has a light shielding mask,
3. The image sensor according to claim 1, wherein the light-receiving area is an opening area of an opening formed by the light-shielding mask. An imaging device comprising a photographing lens and the imaging element according to claim 1 ,
An imaging apparatus further comprising means for selecting a line from which a pixel signal to be used for phase difference detection processing is to be read from the first line and the second line, based on characteristic information of the photographing lens.

Images