JP2022031300A - Electronic apparatus and notification method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an electronic apparatus capable of easily obtaining depth information by one camera, and a notification method.

SOLUTION: An electronic apparatus according to an embodiment includes: input means inputting an image including a first color component image based on a first wavelength region transmitting a first region and a second color component image based on a second wavelength region transmitting a second region, the image being a captured image by a camera including a filter that has the first region transmitting the first wavelength region and the second region transmitting the second wavelength region; and processing means outputting the captured image in a state where an effective region for the calculation of depth information based on deviation of color information in the first color component image and the second color component image is identified. The processing means acquires or calculates the effective area, on the basis of a distance between the filter and an aperture of a lens of the camera, the distance being at least one imaging parameter among a plurality of imaging parameters.

SELECTED DRAWING: Figure 18

COPYRIGHT: (C)2022,JPO&INPIT

Description

Embodiments of the present invention relate to electronic devices and notification methods.

A method using a stereo camera is known as a method of simultaneously acquiring a captured image and depth information. However, since two cameras are required and the distance between the two cameras needs to be long in order to secure the parallax, the device for realizing this method becomes large. Therefore, a method of obtaining depth information with one camera is desired. As an example, a technique for obtaining depth information based on the difference in the shape of blur of an image for each of a plurality of color components captured by an image sensor has been proposed. In order to cause blurring of each color component in the image, the image is taken using a filter consisting of at least two regions that pass through different color components. A mechanism for obtaining depth information using a filter having transmission characteristics of different color components is called a color aperture.

Japanese Unexamined Patent Publication No. 2016-102733

In order to cause blurring of each color component in the image in any part such as the center and the periphery of the image, it is preferable to arrange a filter having transmission characteristics of different color components in the lens aperture. However, for this purpose, it is necessary to custom-order the camera lens or modify a commercially available camera lens. Therefore, it was not easy to obtain depth information with one camera.

An object of the present invention is to provide an electronic device and a notification method that can easily obtain depth information with one camera.

The electronic device according to the embodiment is an image taken by a camera including a filter having a first region transmitted through the first wavelength region and a second region transmitted through the second wavelength region, and is a first image transmitted through the first region. An input means for inputting a captured image including a first color component image based on one wavelength region and a second color component image based on a second wavelength region transmitted through the second region, and the captured image being the first color component. It is an electronic device including a processing means for outputting in a state where an effective region for calculation of depth information based on a bias of color information in the image and the second color component image is identified. The processing means acquires or calculates the effective region based on the distance between the aperture of the lens of the camera and the filter, which is at least one imaging parameter among the plurality of imaging parameters.

It is a figure which shows an example of the change of the light ray with respect to the distance of a subject when a filter is added to the aperture of a lens. It is a figure which shows an example of a filter. It is a figure which shows an example of the transmittance characteristic of the 1st filter area and the 2nd filter area of a filter. It is a figure which shows an example of the change of a light ray when a filter is arranged in a lens opening. It is a figure which shows an example of the change of a light ray when a filter is arranged outside the tip of a lens. It is a figure which shows an example of the appearance of the electronic device by 1st Embodiment. It is a block diagram which shows an example of the electric structure of the electronic device by 1st Embodiment. It is a figure which shows an example of the structure of an image sensor. It is a figure which shows an example of the color filter of an image sensor. It is a figure which shows an example of the relationship between the light ray change by a filter, and the shape of a blur. It is a block diagram which shows an example of the functional structure of the electronic device by 1st Embodiment. It is a figure which shows an example of the blurring function of a reference image in 1st Embodiment. It is a figure which shows an example of the blurring function of the target image in 1st Embodiment. It is a figure which shows an example of the blur correction filter in 1st Embodiment. It is a figure which shows an example of the effective domain determination part in 1st Embodiment. It is a figure which shows an example of determination of an effective domain in 1st Embodiment. It is a figure which shows an example of the effective domain table in 1st Embodiment. It is a figure which shows an example of the guideline which shows the effective domain in 1st Embodiment. It is a flowchart which shows an example of effective domain notification in 1st Embodiment. It is a figure which shows an example of the appearance of the electronic device by 2nd Embodiment. It is a figure which shows an example of the system including the electronic device by 2nd Embodiment. It is a flowchart which shows an example of the effective domain notification by the system including the electronic device in 2nd Embodiment.

Hereinafter, embodiments will be described with reference to the drawings. The disclosure is merely an example, and the invention is not limited by the contents described in the following embodiments. Modifications that can be easily conceived by those skilled in the art are naturally included in the scope of disclosure. In order to clarify the explanation, in the drawings, the size, shape, etc. of each part may be changed with respect to the actual embodiment and represented schematically. In a plurality of drawings, the corresponding elements may be given the same reference numerals and detailed description may be omitted.

[Principle of acquisition of depth information]
FIG. 1 shows an example of a change in light rays depending on the distance of a subject when a filter 10 having transmission characteristics in different wavelength ranges is added to the aperture of the lens 12. FIG. 2 is a diagram showing an example of the filter 10. The filter 10 includes a plurality, for example, two first filter regions 10a and a second filter region 10b. The x direction is the left-right direction, the y direction is the up-down direction, and the left-right direction is the left-right direction when viewed from the subject. The center of the filter 10 coincides with the optical center 16 of the lens. The first filter region 10a and the second filter region 10b are arranged so as not to overlap each other. In the example shown in FIG. 2, the first filter region 10a and the second filter region 10b each have a semicircular shape in which the circular filter 10 is divided into left and right by a vertical line segment passing through the optical center 16. ing. When the image sensor captures a color component image of three colors of red (R), green (G), and blue (B), the first filter region 10a is a combination of the first two colors in R, G, and B. The second filter region 10b transmits the wavelength range corresponding to the light of the first two colors, and the second filter region 10b transmits the wavelength range corresponding to the light of the combination of the second two colors different from the combination of the first two colors.

FIG. 3 shows an example of the transmittance characteristics of the first filter region 10a and the second filter region 10b. For example, the first filter region 10a is composed of a yellow (Y) filter that transmits red and green light wavelength regions, and the second filter region 10b is a cyan (C) filter that transmits green and blue light wavelength regions. Consists of. The transmittance characteristic 18a of the first filter (yellow filter) region 10a has a high transmittance for light in the wavelength range corresponding to the R image and the G image, and hardly transmits light in the wavelength range corresponding to the B image. The transmittance characteristic 18b of the second filter (cyan filter) region 10b has a high transmittance for light in the wavelength range corresponding to the B image and the G image, and hardly transmits light in the wavelength range corresponding to the R image.

Therefore, as shown in FIG. 1, the light rays passing through the aperture change for each of the R, G, and B color component images. The light in the wavelength range of the G image passes through both the first filter (yellow filter) region 10a and the second filter (cyan filter) region 10b, but the light in the wavelength range of the R image passes through the first filter (yellow filter) region. Only 10a is passed, and light in the wavelength range of the B image passes only through the second filter (cyan filter) region 10b. Therefore, the blurred shapes of the R image and the B image of the captured image change according to the distance d to the subject. Further, since the opening is divided into two, the bias direction of the blur of the R image and the B image is reversed before and after the focal plane df. Therefore, the front blur and the back blur can be distinguished, and the depth information corresponding to the distance d from the shape of the blur to the subject can be obtained.

Instead of arranging the filter 10 in the lens aperture, the lens 12 may be arranged between the filter 10 and the image sensor 14 on the optical path of the light incident on the image sensor 14, or the lens 12 and the image sensor may be arranged. The filter 10 may be arranged between 14 and 14. Further, when the lens 12 is composed of a plurality of lenses, the filter 10 may be arranged between any of the lenses 12. Further, instead of the Y filter or the C filter, a magenta (M) fill that transmits the wavelength range of red and blue light may be used. Further, the number of regions of the filter 10 is not limited to 2, and may be 3 or more. The shape of the region is not limited to the configuration in which the entire region is divided by a straight line passing through the center of the circle, and a configuration in which a plurality of regions are arranged in a mosaic pattern may be used.

[Unnecessary color shift for depth estimation]
It is preferable to arrange the filter in the lens opening in order to cause blurring of each of a plurality of color components in the image in any part of the image such as the center and the periphery. However, for this purpose, it is necessary to custom-order the camera lens or modify a commercially available camera lens. Therefore, it is considered to arrange the color file on the outside of the lens. 4 and 5 show an example of the difference in the change of light rays when the filter is arranged at the lens aperture and when the filter is arranged outside the lens tip (subject side). 4 and 5 are views viewed from the plus side (upper side) of the y-axis. As shown in FIGS. 4A and 4B, when the filter 10 is arranged at the opening of the lens 12, the light rays from the central portion and the peripheral portion of the subject also cover both the first filter region 10a and the second filter region 10b. It passes evenly and is incident on the image sensor 14. However, the filter 10 is fixed by screwing the filter 10 into the subject side from the tip of the lens 12, for example, a screw for attaching the filter at the tip of the lens, or fixing the filter 10 to the front surface of the lens 12 by a clip or the like like a wide-angle lens attachment of a smartphone. In this case, as shown in FIG. 5A, the light beam from the central portion of the subject passes through both the first filter region 10a and the second filter region 10b evenly and is incident on the image sensor 14. However, as shown in FIG. 5B, the light rays from the peripheral portion of the subject pass through one of the first filter region 10a and the second filter region 10b more and pass less than the other, and the image sensor. It is incident on 14. In this way, when the filter is divided into the first and second filter areas on the left and right, the image sensor 14 cannot evenly capture the R image and the B image at both the left and right ends of the subject, and the R image and the image sensor 14 The balance with the B image is lost, and undesired color shift that is unnecessary for depth estimation occurs.

However, when the filter 10 is arranged outside the lens aperture, the depth information cannot be measured at all, and the depth information of the subject in the central region of the screen can be obtained, but the depth information of the subject in the peripheral region can be obtained accurately. There are only cases where it cannot be done. Since the area of the subject on which the image sensor 14 can shoot the R image and the B image evenly is determined by various parameters related to the optical characteristics of the camera at the time of image shooting (hereinafter referred to as shooting parameters), the measurement error of the depth information is a practical problem. It is possible to identify a region that is not so small, that is, a region where the reliability of the depth information is higher than the reliability of a certain standard. Further, it is not always necessary to obtain the depth information of the entire screen, and it may be sufficient to obtain the depth information of the part of the subject located in the central region.

In the embodiment, a filter 10 that causes a plurality of color components to have different blurs depending on the distance is arranged other than the lens aperture, and the reliability of the depth information is notified to the user. If the reliability is low, it is possible to obtain highly reliable depth information by changing the composition of the camera and taking a picture again. This provides an electronic device and a notification method that can easily obtain depth information with a single camera without the need to custom-order a camera lens or modify a commercially available camera lens.

[Appearance of the first embodiment]
FIG. 6 shows the appearance of the electronic device of the first embodiment. 6 (a) is a perspective view, and FIG. 6 (b) is a side view showing a cross-sectional structure. As the electronic device of the first embodiment, the smartphone 40 can be used. The smartphone 40 is an example of an electronic device having a function of capturing an image and processing the captured image. The embodiment is not limited to the smartphone 40, but a camera, a mobile phone having a camera function, a personal digital assistant such as a PDA (Personal Digital Assistant, Personal Data Assistant), a personal computer having a camera function, or the like can also be used.

A clip-type attachment 44 is attached to the smartphone 40 so as to cover the front surface of the camera lens 42 of the smartphone 40. The attachment 44 includes a first lens barrel 46 having a diameter larger than that of the camera lens 42, and the second lens barrel 48 is inserted inside the first lens barrel 46. A filter 52 is arranged at the tip of the second lens barrel 48. Similar to the filter 10 shown in FIG. 2, the filter 52 includes a yellow first filter region 52a and a cyan second filter region 52b. When the user applies a rotational force to the second lens barrel 48, the second lens barrel 48 is rotated with respect to the first lens barrel 46. Thereby, the orientation of the filter 52 (the orientation of the region dividing line by the first filter region 52a and the second filter region 52b) can be adjusted. When the user does not apply a rotational force, the second lens barrel 48 is stationary with respect to the first lens barrel 46. A set screw (not shown) or the like may be provided on the first lens barrel 46 to suppress the rotation of the second lens barrel 48 after the orientation of the filter 52 is adjusted.

As a condition of the subject for which depth information is required, there is a direction in which the edge extends (edge direction). The edge direction of a vertical object such as a pillar is horizontal, and the edge direction of a horizontal object such as a desk is vertical. Depth information can be obtained when the area division direction of the filter and the edge direction of the subject are not parallel. In the filter shown in FIG. 2, since the two regions are arranged side by side, the division direction is the vertical direction. Therefore, the subject from which the depth information can be obtained by using the filter shown in FIG. 2 is a vertical object having a horizontal edge. Here, it is assumed that the edge direction of the subject for which depth information is to be obtained is the horizontal direction. The attachment 44 is attached so that the area division direction of the filter intersects the edge direction of the subject so as to be suitable for such a subject. Therefore, the marker 46a for determining the mounting position of the attachment 44 is marked on the surface of the first lens barrel 46, and the marker 48a is marked on the surface of the second lens barrel 48. The attachment 44 is attached to the smartphone 40 so that the marker 46a matches the LED 40a of the smartphone 40, and the second lens barrel 48 is positioned with respect to the first lens barrel 46 so that the positions of the marker 48a and the marker 46a are aligned. In this case, the depth information of the subject whose edge direction is the horizontal direction is required. If the edge direction of the subject is not horizontal, then the user rotates the second lens barrel 48 so that the direction of the area dividing line of the filter 52 intersects the edge direction, and the direction of the filter 52 is adjusted. ..

Although the second lens barrel 48 is provided inside the first lens barrel 46, it may be provided outside the first lens barrel 46.
When the edge direction of the subject is parallel to the area dividing line of the filter 52, the smartphone 40 itself may be rotated so that the direction of the area dividing line intersects the edge direction. In this case, since the filter 52 does not have to be rotatable, the second lens barrel 48 is unnecessary, and the filter 52 may be provided on the first lens barrel 46.

[Electrical configuration of the first embodiment]
FIG. 7 is a block diagram showing an example of the electrical configuration of the smartphone 40. The smartphone 40 includes an image sensor 54, a CPU 56, a RAM 58, a display 60, a communication unit 62, a non-volatile memory 64, a memory card slot 66, and an audio output unit 68 for photographing a subject image incident through the filter 52 and the lens 42. Be prepared. The image sensor 54, CPU 56, RAM 58, display 60, communication unit 62, non-volatile memory 64, memory card slot 66, and audio output unit 68 are connected to each other via a bus line 69.

The CPU 56 is a processor that controls the operation of various components in the smartphone 40. The CPU 56 executes various programs loaded into the RAM 58 from the non-volatile memory 64, which is a storage device. In addition to various application programs, the non-volatile memory 64 also stores an OS (Operating System) 64a. Examples of various application programs are the depth calculation program 64b and the effective domain notification program 64c. The non-volatile memory 64 may also store an image taken by the image sensor 14 and processing results of the image by the programs 64b and 64c.

The communication unit 62 is an interface device configured to execute wired communication or wireless communication. The communication unit 62 includes a transmission unit that transmits a signal by wire or wirelessly, and a reception unit that receives the signal by wire or wirelessly. For example, the communication unit 62 can connect the smartphone 40 to the external server 94 via the Internet 92.

The display 60 is, for example, an LCD (Liquid Crystal Display). The display 60 displays a screen image based on a display signal generated by the CPU 56 or the like. For example, the display 60 has a function of displaying an captured image and depth information, and a function of notifying the user of the reliability by displaying the reliability itself or information according to the reliability as text, an icon, a mark, or the like. The display 60 may be a touch screen display. In that case, for example, the touch panel is arranged on the upper surface of the LCD. The touch panel is a capacitive pointing device for inputting on the screen of the LCD. The touch panel detects the contact position on the screen to which the finger is touched and the movement of the contact position.

The audio output unit 68 is a speaker, an earphone jack, or the like, and notifies the user of the reliability itself or information according to the reliability as voice.
A memory card, which is various portable storage media such as an SD (registered trademark) memory card and an SDHC (registered trademark) memory card, can be inserted into the memory card slot 66. When a storage medium is inserted into the memory card slot 66, data can be written to and read from the storage medium. The data is, for example, image data or distance data.

The application program does not have to be built in the smartphone 40. For example, the depth calculation program 64b may be provided in the server 94 instead of the smartphone 40. In this case, the image captured by the image sensor 54 may be transmitted to the server 94 by the communication unit 62, the depth information may be obtained by the depth calculation program provided in the server 94, and the depth information may be returned to the smartphone 40. Further, the depth calculation program 64b and the effective domain notification program 64c may be provided in the server 94 instead of the smartphone 40. In this case, the image captured by the image sensor 54 is transmitted to the server 94 by the communication unit 62, the depth information is obtained by the depth calculation program provided in the server 94, and the effective area is determined by the effective area notification program provided in the server 94. , Depth information and notification information of the effective area may be returned to the smartphone 40.

[Image sensor]
The image sensor 54, for example, a CCD type image sensor, consists of a photodiode as a light receiving element arranged in a two-dimensional matrix and a CCD that transfers a signal charge generated by the photodiode converting incident light by photoelectric conversion. Become. FIG. 8 shows an example of the cross-sectional structure of the photodiode portion. A large number of n-type semiconductor regions 72 are formed on the surface region of the p-type silicon substrate 70, and a large number of photodiodes are formed by pn junctions between the p-type substrate 42 and the n-type semiconductor region 72. Here, two photodiodes constitute one pixel. Therefore, each photodiode is also referred to as a sub-pixel. A light-shielding film 74 for suppressing crosstalk is formed between the photodiodes. A multilayer wiring layer 76 in which transistors, various wirings, and the like are provided is formed on the p-type silicon substrate 70.

The color filter 80 is formed on the wiring layer 76. The color filter 80 is composed of a two-dimensional array of a large number of filter elements that transmit, for example, red (R), green (G), or blue (B) light for each pixel. Therefore, each pixel generates only image information of any color component of R, G, and B. The image information of the color components of the other two colors that are not generated in the pixel is obtained by interpolation from the color component image information of the surrounding pixels. Moire and false colors may occur during interpolation in the shooting of periodic repeating patterns. In order to prevent this, although not shown, an optical low-pass filter made of quartz or the like and slightly blurring the repeating pattern may be arranged between the lens 42 and the image sensor 54. Instead of providing an optical low-pass filter, the same effect may be obtained in the signal processing of the image signal.

A microlens array 82 is formed on the color filter 80. The microlens array 82 consists of a pair of microlenses 82a and 82b corresponding to one pixel in a two-dimensional array-like arrangement. Although FIG. 8 shows the front surface incident type image sensor 54, the back surface incident type may be used. The two photodiodes constituting one pixel are configured to receive light that has passed through different regions of the exit pupil of the lens 42 via a pair of microlenses 82a and 82b, and so-called pupil division is realized. Has been done.

FIG. 9A is a plan view showing an example of the relationship between the photodiodes 84a and 84b constituting each pixel and the microlenses 82a and 82b. As shown in FIG. 9A, light that has passed through the first filter region 10a of the filter 52 is incident on the photodiode 84a as the first sub-pixel. Light that has passed through the second filter region 10b of the filter 52 is incident on the photodiode 84b as the second sub-pixel.

FIG. 9B shows an example of the color filter 80. The color filter 80 is, for example, a Bayer array primary color filter. As a result, the image sensor 54 includes an R pixel that senses red light, a G pixel that senses green light, and a B pixel that senses blue light. The color filter 80 may be a complementary color filter. Further, when it is not necessary to capture a color image and only the depth information needs to be obtained, the image sensor 54 does not have to be a color sensor, may be a monochrome sensor, and may not have a color filter 80.

The image sensor 14 is not limited to the CCD type sensor, and a CMOS (Complementary Metal Oxide Semiconductor) type sensor may be used.
[Blur shape due to filter]
With reference to FIG. 10, the light ray change and the shape of the blur due to the filter 52 will be described. The filter 52 includes a yellow first filter region 52a and a cyan second filter region 52b. When the subject 88 is deeper than the focusing distance df (d> df), the image captured by the image sensor 54 is blurred. The blur function (PSF: Point Spread Function) indicating the shape of the blur on the image is different between the R image, the G image, and the B image. The blur function 101R of the R image shows the shape of the blur biased to the left, the blur function 101G of the G image shows the shape of the blur without bias, and the blur function 101B of the B image shows the shape of the blur biased to the right. There is.

When the subject 88 is at the focusing distance df (d = df), almost no blur occurs in the image captured by the image sensor 54. The blur function indicating the shape of the blur on this image is almost the same for the R image, the G image, and the B image. That is, the blur function 102R of the R image, the blur function 102G of the G image, and the blur function 102B of the B image show an unbiased shape of the blur.

When the subject 88 is in front of the focusing distance df (d <df), blurring occurs in the image captured by the image sensor 54. The blur function indicating the shape of the blur on the image is different between the R image, the G image, and the B image. The blur function 103R of the R image shows the shape of the blur biased to the right, the blur function 103G of the G image shows the shape of the blur without bias, and the blur function 103B of the B image shows the shape of the blur biased to the left. There is.

In the present embodiment, the depth information up to the subject is calculated for each pixel of the image by utilizing such a characteristic.
[Functional configuration of the embodiment]
FIG. 11 is a block diagram showing an example of the functional configuration of the smartphone 40. The arrow from the color filter 10 to the image sensor 54 indicates the path of light. The arrow ahead of the image sensor 54 indicates the path of the electrical signal.

The red light that has passed through the first filter (Y) region 52a of the filter 52 is incident on the R pixel (referred to as the first sensor) 54a of the image sensor 54 via the lens 42, and the first filter (Y) of the filter 52. The green light that has passed through the region 52a is incident on the G pixel (referred to as the second sensor) 54b of the image sensor 54 via the lens 42. The green light that has passed through the second filter (C) region 52b of the filter 52 is incident on the second sensor 54b of the image sensor 54 via the lens 42, and the blue light that has passed through the second filter (C) region 52b of the filter 52. Light is incident on the B pixel (referred to as the third sensor) 54c of the image sensor 54 via the lens 42.

The output signals of the first sensor 54a, the second sensor 54b, and the third sensor 54c are input to the image input unit 102. The image input unit 102 acquires a G image whose point spread function (PSF) shows an unbiased shape as a reference image, and one of an R image and a B image whose blur function shows a biased shape. Or both are acquired as the target image. The target image and the reference image are images taken at the same time by the same camera.

The output signal of the image input unit 102 is supplied to the display 60, and is also supplied to the depth calculation unit 104 and the effective domain determination unit 106. The outputs of the depth calculation unit 104 and the effective domain determination unit 106 are also supplied to the display 60.
[Depth calculation]
The depth calculation unit 104 calculates the depth information up to the subject reflected in the image for each pixel by using the blur correction filter. Instead of calculating the depth information for each pixel, the depth information may be calculated for each pixel block of a set of a plurality of pixels. The blur correction filter is a filter that, when calculated on the target image, makes the correlation between the target image after the calculation and the reference image higher than the correlation between the target image before the calculation and the reference image. The method of calculating the correlation between the corrected image and the reference image will be described later. When calculated as a blur correction filter, the shape of the blur of the target image approaches the shape of the blur of the reference image.

In the embodiment, a plurality of blur correction filters for each of a plurality of distances created by assuming the distance to the subject in the image as a plurality of distances are prepared. A corrected image in which the blur shape of the target image is corrected is generated, and a blur correction filter having the highest correlation between the corrected image and the reference image is searched for from a plurality of blur correction filters. The distance corresponding to the blur correction filter having the highest correlation is determined as the depth information to the subject. However, instead of selecting one blur correction filter from a plurality of blur correction filters prepared in advance, a blur correction filter satisfying the above conditions may be obtained by calculation.

The depth calculation unit 104 may output a distance image in which each pixel of the image is colored according to the distance from the further calculated depth.
The blur function of the captured image is determined by the shape of the filter 52 (a mode of region division) and the distance between the position of the subject and the focus position. FIG. 12 is a diagram showing an example of a blur function of the reference image according to the embodiment. As shown in FIG. 12, since the aperture shape through which the wavelength region of green light corresponding to the second sensor 54b is transmitted is a circular shape that is point-symmetrical, the shape of the blur shown by the blur function is the focus position. There is no change in the front and back, and the width of the blur changes depending on the size of the distance between the subject and the focus position. The blur function showing the shape of such a blur can be expressed as a Gaussian function in which the width of the blur changes depending on the size of the distance between the position of the subject and the focus position. The blur function may be expressed as a pillbox function in which the width of the blur changes depending on the distance between the position of the subject and the focus position.

On the other hand, FIG. 13 is a diagram showing an example of a blur function of the target image according to the embodiment. The center of each figure is (x, y) = (0,0). As shown in FIG. 13, the blur function of the target image (for example, R image) can be expressed as a Gaussian function in which the width of the blur is attenuated at x> 0 when the subject is d> df farther than the focus position. Further, the blur function of the target image (for example, R image) can be expressed as a Gaussian function in which the width of the blur is attenuated at x <0 when the subject is closer to the focus position than d <df. The blur function of the other target image (for example, the B image) can be expressed as a Gaussian function in which the width of the blur is attenuated at x <0 when the subject is d> df farther than the focus position. Further, the blur function of the B image can be expressed as a Gaussian function in which the width of the blur is attenuated at x> 0 when the subject is closer to the focus position than d <df.

By analyzing the blur function of the reference image and the blur function of the target image, it is possible to obtain a plurality of blur correction filters for correcting the blur shape of the target image to the blur shape of the reference image.
FIG. 14 is a diagram showing an example of a blur correction filter according to an embodiment. The blur correction filter passes through the center point of the line segment (vertical direction) at the boundary between the first filter region 52a and the second filter region 52b, and is distributed on a straight line (horizontal direction) orthogonal to this line segment. The distribution is a mountain-like distribution as shown in the figure, in which the peak point (position on a straight line, height) and the way of spreading from the peak point are different for each assumed distance. The blur shape of the target image can be corrected to various blur shapes assuming an arbitrary distance by using a blur correction filter. That is, it is possible to generate a corrected image assuming an arbitrary distance.

The depth calculation unit 104 obtains the distance at which the blurred shape of the generated corrected image and the reference image most closely approximates or matches with each pixel of the captured image. The degree of matching of the blurred shape may be calculated by calculating the correlation between the corrected image and the reference image in a rectangular area of an arbitrary size centered on each pixel. The existing similarity evaluation method may be used to calculate the degree of matching of the blurred shapes. The depth calculation unit 104 obtains the corresponding distance of the correction filter having the highest correlation between the corrected image and the reference image, and calculates the depth information to the subject reflected in each pixel. For example, existing similarity evaluation methods include SSD (Sum of Squared Difference), SAD (Sum of Absolute Difference), NCC (Normalized Cross-Correlation), ZNCC (Zero-mean Normalized Cross-Correlation), Color Alignment Measure, etc. You can use it. In the present embodiment, the Color Alignment Measure is used, which utilizes the fact that the color components of the natural image have a characteristic of having a locally linear relationship. In the Color Alignment Measure, an index showing the correlation is calculated from the dispersion of the color distribution in the local boundary region centered on the target pixel of the captured image.

[Depth information reliability notification]
The embodiment has a function of obtaining the reliability of the depth information calculated by the depth calculation unit 104 and notifying the user of the information according to the reliability. As an example of information notification according to the reliability, an effective area composed of pixels having a reliability higher than the reference reliability may be obtained and the effective area may be notified to the user. 15 (a), (b), and (c) show some examples of the effective domain determination unit 106 shown in FIG. 11 regarding this notification.

The effective domain determination unit 106a shown in FIG. 15A includes an effective domain calculation unit 112 and an imaging parameter input unit 114. Since the effective area is determined by the shooting parameters, the effective area calculation unit 112 calculates the effective area according to the shooting parameters input from the shooting parameter input unit 114.

The shooting parameters include the brightness (F value) of the lens, the focal length, the distance between the filter and the lens aperture, the size of the image sensor, and the like. The shooting parameter input unit 114 may be configured to display a parameter input / selection screen on the display 60 so that the user can input each shooting parameter using the touch panel provided on the display 60. Since the brightness of the lens, the focal length, and the size of the image sensor are values unique to the smartphone 40, even if the effective domain notification program 64c is configured to be automatically input when the smartphone 40 is installed. good. Further, when the lens 42 has an optical zoom function, the focal length is variable, so that the focal length information may be input from the zoom mechanism. Since the distance between the filter and the lens aperture is a value peculiar to the attachment 44, it is necessary to input it for each attachment 44. However, the shooting parameter input unit 114 may be provided with a table that stores this distance for each type of attachment 44, and the user may be configured to input the type of attachment 44 instead of the distance itself.

When the filter 52 having the same vertical division configuration as the filter 10 as shown in FIG. 2 is used, the effective area is a vertically long area having a constant width centered on the center of the screen, so that the effective area calculation unit 112 has a shooting parameter. Based on, the coordinates of the left and right edges of the effective area (for example, the number of pixels from the center in the left-right direction of the screen) are calculated.

An example of calculation of the effective domain will be described. When a subject whose pixel values are known for all pixels of the screen, for example, a pure white background, is photographed using the filter 52 as shown in FIG. 2, the captured image is near the center in the left-right direction as shown in FIG. 16A. Is white, bluish toward the right end, and reddish toward the left end. FIG. 16 (b) shows a cross-sectional profile of the pixel value (luminance) of the captured image in the broken line portion of FIG. 16 (a). Once the imaging parameters are known, the cross-section profile can be calculated by an optical simulation program. Therefore, the effective domain calculation unit 112 calculates the cross-sectional profile as shown in FIG. 16B by optical simulation based on the shooting parameters input from the shooting parameter input unit 114. Then, the effective domain calculation unit 112 calculates the position coordinates of the left and right end portions of the effective domain whose R pixel value and B pixel value are equal to or higher than the predetermined luminance. The predetermined brightness is related to the reference reliability. The reference reliability for determining the effective region may be different depending on the allowable value of the measurement error of the depth information (referred to as the depth requirement level). That is, when the depth demand level is high, the predetermined luminance is high and the effective region is narrow, and when the depth demand level is low, the predetermined luminance is low and the effective region is wide. The depth request level is also input from the shooting parameter input unit 114. Information about the determined effective area is displayed on the display 60 or output as audio via the audio output unit 68.

The effective domain determination unit 106b shown in FIG. 15B includes a shooting parameter input unit 114, an effective domain acquisition unit 116, and an effective domain memory 118. The effective domain calculation unit 112 shown in FIG. 15 (a) pre-calculates the cross-sectional profile as shown in FIG. 16 (b) by optical simulation for various imaging parameters and depth required levels, and various imaging parameters and depth required levels are obtained. The effective domain table showing the effective domain of is stored in the effective domain memory 118. The depth request level is not essential, and when the effective area is held in advance as a table, the information indicating the pre-calculated effective area corresponding only to the shooting parameters is held, and the effective area is variable according to the depth request level. It does not have to be. It should be noted that the R pixel value and the B pixel value are equal to or higher than the predetermined brightness by analyzing a photographed image in which a subject whose pixel value is known for all pixels of the screen, for example, a pure white background, is analyzed without calculation by an optical simulation program. The position coordinates of the left and right ends of the effective area may be obtained and stored in the memory 118.

FIG. 17 shows an example of the effective area table in the effective area memory 118. The effective domain acquisition unit 116 reads the information of the effective range corresponding to the imaging parameter and the depth request level input from the imaging parameter input unit 114 from the effective domain memory 118. When the lens has an optical zoom function, the focal length is variable, but the effective area is the same for a certain focal length range, so the effective area is obtained for each of several focal length ranges within the variable focal length range. And register it in the table. The information about the effective area read from the effective area memory 118 is displayed on the display 60 or output as audio via the audio output unit 68.
The effective domain can also be calculated statistically by machine learning. For a scene and / or a pixel pattern for which correct depth information is known, the estimation error can be obtained by estimating the depth with a certain shooting parameter. Obtaining the estimation error for a sufficient number of scenes and / or pixel patterns gives a statistic of the estimation error between the shooting parameters and the scenes and / or image patterns. Therefore, when the shooting parameters and the scene and / or the image pattern are specified, the effective domain calculation unit 112 of FIG. 15A can obtain the statistics of the estimation error.
For example, when the variance is used as a statistic, it can be determined that the region where the variance of the estimation error is large is the region where the reliability is low. A statistical model that expresses the relationship between statistics and reliability can be obtained by machine learning such as a neural network if training data with known correct depth information is prepared. The acquired statistical model is held in the shooting parameter input unit 114, and is input to the effective area calculation unit 112 when the effective area is calculated. As a result, the effective domain calculation unit 112 can obtain the reliability based on the shooting parameters and the scene and / or the image pattern and the statistical model by machine learning, and can obtain the effective domain whose reliability is higher than the reference reliability.

The effective domain calculation unit 112 shown in FIG. 15 (a) and the effective domain memory 118 shown in FIG. 15 (b) do not need to be provided in the smartphone 40 of the embodiment, and may be provided in the server 94. FIG. 15C shows an example of the effective domain determination unit 106c in this case. The effective domain determination unit 106c includes a shooting parameter input unit 114 and an effective domain acquisition unit 120. The shooting parameters input from the shooting parameter input unit 114 to the effective domain acquisition unit 120 are transmitted to the server 94 via the communication unit 62. The effective area is calculated by the server 94 or read from the effective area memory, the information about the effective area is returned to the communication unit 62, and the effective area acquisition unit 120 acquires the information about the effective area. Information about the effective area received by the communication unit 62 is displayed on the display 60 or output as voice via the audio output unit 68.

FIG. 18 shows an example of information about the effective area displayed on the display 60. The image acquired by the image input unit 102 is displayed on the display 60. A pair of guidelines 124 showing the left and right edges of the effective area are displayed on this image. The user can see the image on which the guideline 124 is displayed and confirm whether or not the measurement target for which the depth is to be obtained is located within the effective region. When the measurement target is located outside the effective area, the measurement target can be located inside the effective area by changing the composition or the like. Therefore, even if color shift occurs due to a filter located other than the lens aperture, the measurement error of the depth information of the measurement target can be set to the allowable value or less. Since the arrangement direction of the filter 10 (the direction of the boundary line between the first filter area and the second filter area) is set to the vertical direction, the effective area is a vertically long area having a certain width centered on the center of the screen, but the filter. The effective area differs depending on the arrangement direction of 10. For example, since the arrangement direction of the filter 10 is set to the horizontal direction, the effective area is a horizontally long area having a certain width centered on the center of the screen.

As a modification of FIG. 18, instead of displaying the guideline 124, an image in which only the effective area is clipped may be displayed. That is, instead of the images on the right side of the guideline on the right side of FIG. 18 and the image on the left side of the guideline on the left side, a constant brightness, for example, a black background may be displayed. Further, the image may be displayed by changing the color, brightness or contrast between the effective area and the non-effective area. When the image is displayed by changing the color, brightness, or contrast of the image, the image is not limited to the binary change between the effective area and the non-effective area, and may be continuously changed according to the reliability. The reliability itself can be obtained based on the cross-sectional profile of the pixel value (luminance) of the captured image, as described with reference to FIG. For example, it can be determined that the cross-sectional position where the difference between the R pixel value and the B pixel value is large has color shift and the reliability is low, and the cross-sectional position where the difference between the R pixel value and the B pixel value is small has no color shift and the reliability is high. .. Further, a reliability map in which each pixel of the captured image is colored according to the reliability may be displayed and saved, or the reliability (numerical value) of each pixel may be displayed and saved in the form of a table corresponding to the pixel. Is also good. The reliability map and the reliability table may be displayed together with the captured image.
Further, the effective area may be presented to the user by displaying it as an image on the display 60, or if the position of the effective area or the guideline 124 is determined in advance, the image other than the effective area cannot be seen by the user on the display 60. The effective area may be presented to the user by physical means such as attaching a sticker to the area other than the effective area.

FIG. 19 is a flowchart showing an example of notification of information according to the reliability according to the first embodiment. At block 152, the user activates the depth calculation program 64b and the effective domain notification program 64c. At block 154, the user attaches the attachment 44 to the front of the lens 42 of the smartphone 40. As shown in FIG. 18, when the measurement target for which depth information is obtained is an object whose edge direction is vertical, the user determines the orientation of the attachment 44 so that the marker 46a of the first lens barrel 46 matches the LED 40a of the smartphone 40. The orientation of the second lens barrel 48 is determined so that the marker 48a of the second lens barrel 48 is aligned with the marker 46a. When the edge direction of the measurement target is horizontal, the user determines the direction of the attachment 44 so that the marker 46a of the first lens barrel 46 matches the LED 40a of the smartphone 40, and then the marker of the second lens barrel 48. The orientation of the second lens barrel 48 is determined so that the 48a deviates from the marker 46a by 90 degrees.

At block 156, the user launches a camera application program (not shown, stored in non-volatile memory 64). When the camera application program is activated, at block 158, the display 60 displays a through image output from the image sensor 54.

At block 162, the user inputs a shooting parameter (type of attachment 44) using the touch pad. It should be noted that the shooting parameters may be input not after the execution of the block 158 but before the execution of the block 164. For example, when the depth calculation program 64b and the effective domain notification program 64c are started in the block 152, a setting screen may be displayed and shooting parameters may be input at that stage. The lens brightness, focal length, and image sensor size, which are shooting parameters unique to the smartphone 40, are values unique to the smartphone 40 and do not need to be input by the user, but these may also be input by the user. ..

At block 164, the effective domain determination unit 106 calculates the effective domain based on the shooting parameters (and in some cases also the depth request level) (in the case of FIG. 15A) and reads it from the effective domain memory 118 (FIG. 15 (FIG. 15). In the case of b)) or acquired from the server 94 via the communication unit 62 (in the case of FIG. 15 (c)). In block 166, the coordinate information of the left and right ends of the effective information is supplied from the effective area determination unit 106 to the display 60, and as shown in FIG. 18, a pair of guidelines 124 indicating the left and right ends of the effective area are displayed on the through image. Will be done.

At block 168, the effective domain determination unit 106 performs image analysis to identify the background and the object in the through image. The background is an area with a constant luminance value without contrast, and the object is an area with contrast and a difference in luminance value. When the effective domain determination unit 106 recognizes an object, the effective domain determination unit 106 sets the object as a depth measurement target. Further, an object-like region may be extracted by an image recognition technique or an object detection technique. The depth measurement target is not limited to being automatically recognized by using image analysis, and may be selected by the user on the touch panel.

At block 172, the effective domain determination unit 106 determines whether or not the depth measurement target is located within the effective domain. When the depth measurement target is not located in the effective region (no determination in block 172), in block 174, the effective region determination unit 106 notifies the user of shooting guidance for positioning the depth measurement target in the effective region. .. Examples of guidance include displaying text that changes the composition such as "more right" and "more left" on the display 60, and voice guidance that changes the composition such as "more right" and "more left". It includes outputting from the audio output unit 68.

When analyzing the image of the block 168, the edge direction of the subject is also recognized from the contrast, and in the block 172, the edge direction of the subject is the horizontal direction, that is, the edge direction of the subject and the direction of the area dividing line of the filter 52 are set. If it is determined that they intersect, the guidance of the block 174 may be such that the direction of the filter 52 is rotated by 90 degrees.
By notifying the user of such a guide, the user can surely position the depth measurement target in the effective area, and after the depth measurement target is located in the effective area, the user performs a shutter operation. Therefore, the block 176 determines whether or not the shutter has been pressed. If the shutter is not pressed, the process from block 164 is repeated. When the shutter is pressed, it can be determined that the depth measurement target is located in the effective area, the depth information is calculated for each pixel in the block 178, and the depth information is displayed or stored. The depth information may be displayed and saved as a distance image in which each pixel of the image is colored according to the distance, or the depth information (numerical value) for each pixel may be displayed and saved in the form of a table corresponding to the pixel. good.

Further, the captured image corresponding to the distance image may be saved, but the captured image may have color shift and may be unsightly if displayed as it is. Since the range of color shift corresponds to the effective region, it is possible to compensate for the color shift based on the shooting parameters for determining the effective region to obtain a shot image without color shift. If the reliability of each pixel is required, the reliability can be compensated for the color shift. Therefore, when saving the captured image corresponding to the distance image, the color shift may be compensated and then saved.

Note that the calculation of the depth information may be performed for each frame of the through image by the block 164 or the like in parallel with the acquisition of the effective region regardless of the shutter operation. In this case, the depth measurement target of the block 168 may be recognized based on the depth information instead of being recognized based on the contrast by image analysis. This is because the depth information can only be obtained up to a certain finite distance, and the depth information of an object at infinity cannot be obtained.

According to the first embodiment, a filter 52 that causes a plurality of color components to have different blurs depending on a distance is attached to the front surface of the lens 42 of the smartphone 40. The filter 52 has at least two filter regions. Depending on the installation method of the filter 52 and the shooting parameters of the camera optical system, the light of a plurality of color components may not pass evenly at the end of the lens opening, and the light of one of the color components may be larger than the light of the other color components. There are cases where the depth information cannot be obtained accurately on the entire screen due to undesired color shift (color information bias) that is unnecessary for depth estimation. In the first embodiment, the user is notified of the effective area from which the depth information can be obtained with high accuracy. In addition, if the measurement target is not located in the effective area, guidance for locating the measurement target in the effective area is also notified. Therefore, it is expected that shooting will be performed so that the depth information of the measurement target can be obtained with high accuracy. According to the first embodiment, there is provided an electronic device and a notification method that can easily obtain depth information with one camera without the need to custom-order a camera lens or modify a commercially available camera lens.

Hereinafter, other embodiments will be described. In the following description, the components corresponding to the first embodiment are designated by the same reference numerals, and detailed description thereof will be omitted.
[Second Embodiment]
FIG. 20 shows the appearance of the electronic device of the second embodiment. 20 (a) is a perspective view, and FIG. 20 (b) is a side view showing a cross-sectional structure. The camera 202 can be used as the electronic device of the second embodiment.

The first lens barrel 206 is screwed into the filter mounting screw at the tip of the lens 204 of the camera 202. Alternatively, the first lens barrel 206 may be attached to the tip of the lens 204 by a bayonet-type fixing means. The second lens barrel 208 is inserted inside the first lens barrel 206. The filter 212 is arranged at the tip of the second lens barrel 208. The filter 212 is composed of a yellow first filter region 212a and a cyan second filter region 212b, similarly to the filter 52 of the first embodiment. When the user applies a rotational force to the second lens barrel 208, the second lens barrel 208 is rotated with respect to the first lens barrel 206. Thereby, the orientation of the filter 212 (the orientation of the region dividing line by the first filter region 212a and the second filter region 212b) can be adjusted. When the user does not apply a rotational force, the second lens barrel 208 is stationary with respect to the first lens barrel 206. A set screw (not shown) or the like may be provided on the first lens barrel 206 to suppress the rotation of the second lens barrel 208 after the orientation of the filter 212 is adjusted.

Although the second lens barrel 208 is provided inside the first lens barrel 206, it may be provided outside the first lens barrel 206.
When the edge direction of the subject is different from the area dividing line of the filter 212, the camera 202 itself may be rotated so that the direction of the area dividing line is aligned with the edge direction. In this case, since the filter 212 does not have to be rotatable, the second lens barrel 208 is unnecessary, and the filter 212 may be provided on the first lens barrel 206.

The electrical configuration of the camera 202 is almost the same as that of the example of the electrical configuration of the smartphone 40 of the first embodiment, but is different from FIG. 7 in that the depth calculation program 64b and the effective domain notification program 64c are not installed. The depth calculation program and the effective domain notification program are provided in the second device.

The second device may be a server 94 or a personal computer owned by the owner of the camera 202. FIG. 21 shows an example of cooperation between a personal computer 230 equipped with a depth calculation program and an effective domain notification program and a camera 202. The captured image (still image) of the camera 202 is stored in the memory card 222 in the memory card slot 220. The memory card 222 taken out from the memory card slot 220 of the camera 202 is inserted into the memory card slot 232 of the personal computer 230. The personal computer 230 reads an image from the memory card 222 in the memory card slot 232. As a result, the image captured by the image sensor 214 of the camera 202 is sent to the personal computer 230. When the camera 202 has a communication function, it is not necessary to go through the memory card 222, and the image captured by the image sensor 214 of the camera 202 is communicated to the personal computer 230.

The personal computer 230 performs depth calculation and effective domain notification processing in the same manner as the smartphone 40 of the first embodiment.
FIG. 22 is a flowchart showing an example of information notification according to the reliability according to the 21st embodiment. It is assumed that the flowchart of FIG. 22 starts with the memory card 222 in which the captured image of the camera 202 is stored inserted in the memory card slot 232 of the personal computer 230.

At block 302, the user activates the depth calculation program and the effective domain notification program on the personal computer 230. At block 304, the personal computer 230 reads an image from the memory card 222.
At block 306, the user inputs a shooting parameter and a depth request level. In some cases, it is not necessary to input the depth request level. Further, when the camera 202 has a function of recording the shooting parameters together with the image, the shooting parameters are read into the personal computer 230 together with the images, so that no user input is required.

At block 308, personal computer 230 displays an image. A list of images of several frames may be displayed, or the images may be switched and displayed frame by frame. At block 312, the user selects an image frame for depth calculation.
At block 314, the personal computer 230 determines the effective domain by the effective domain notification program. Similar to the first embodiment, the personal computer 230 may calculate the effective area based on the shooting parameters, or may read the effective area information from the effective area memory included in the personal computer 230 based on the shooting parameters. However, the effective domain information may be acquired from another device, for example, the server 94 via the communication unit.

At block 316, as in the first embodiment, the personal computer 230 displays the selected frame image on the display 236 and displays a pair of guidelines 124 indicating the left and right edges of the effective area, as shown in FIG. do.
At block 318, the personal computer 230 identifies an object in a region having contrast and a difference in luminance value from a background which is a region having a constant luminance value without contrast by image analysis, and recognizes the object as a depth measurement target. .. The depth measurement target may be selected by the user on the touch panel instead of being recognized based on the contrast by image analysis.

At block 322, the personal computer 230 determines whether or not the depth measurement target is located within the effective region. If the depth measurement target is not located within the effective region (no determination in block 312), block 324 alerts that the frame cannot measure depth information. The alert may be a prompt for re-shooting. The alert may be displayed on the display 236 or may be output as audio from an audio output unit (not shown). Further, the alert may be saved in association with the frame image. By doing so, when selecting the frame for which the depth information is calculated in the block 212, it is possible to eliminate the waste of reselecting the frame once determined that the measurement target is located outside the effective area. Also, if the depth measurement target is saved with the image as alert information on which side of the effective area, the frame in which the alert is saved is displayed on the camera 202 and the alert is also output when reshooting. By doing so, it can also be used as a guide when deciding the composition.

Then, at block 326, the personal computer 230 determines whether another frame has been selected. If another frame is not selected, the operation ends. If another frame is selected, the personal computer 230 re-executes the processing of the blocks after the block 208.

When the depth measurement target is located in the effective area (yes judgment of block 312), in block 328, the personal computer 230 calculates the depth information for each pixel by the depth calculation program, and displays or saves the depth information. .. Similar to the first embodiment, the depth information may be a distance image in which each pixel of the image is colored according to the distance, or the depth information (numerical value) for each pixel may be in the form of a table corresponding to the pixel.

According to the second embodiment, in addition to the effect of the first embodiment, there is an effect that it is not necessary to install the depth calculation program and the effective domain notification program in the camera 202 which is an electronic device. These programs may be provided in a second electronic device to which the image of the camera is transferred, for example, a server or a personal computer, and can perform depth calculation and effective domain notification at a higher speed than processing by individual cameras. ..

[Application example]
Next, an application example of an electronic device for obtaining depth information will be described.
The electronic device of the embodiment can be applied to a monitoring system that detects an intruder in a space photographed by a camera and issues an alarm. In the monitoring system, for example, the flow of people and cars in a store, a parking lot, etc. is grasped, and the intrusion of a person into a specific area is monitored. Further, when it is detected that a person moves in front of the automatic door, the door may be opened.

The electronic device of the embodiment can also be applied to a mobile control system. In this system, moving objects such as mobile robots (Automated Guided Vehicles), cleaning robots, communication robots and other autonomously moving robots, vehicles including automobiles, flying objects such as drones and airplanes, and subjects in the direction of travel such as ships Is imaged, and the operation of safety devices such as acceleration, deceleration, stop, collision avoidance, direction change, and air bag are controlled. The system also includes a so-called front camera that captures the front of the vehicle and a so-called rear camera that captures the rear when backing. Further, a camera may be installed at the tip of a stationary robot arm or the like to control object gripping of the robot arm.

When applied to drones, when inspecting cracks from the sky, broken wires, etc., the inspection target is photographed by the camera and the distance to the inspection target is detected, or is the inspection target in front of the reference distance, or is it? It is determined whether it is in the back. Based on this detection result or judgment result, the thrust of the drone is controlled so that the distance to the inspection target becomes constant. This allows the drone to fly in parallel with the inspection target.

Also, when the drone is flying, the camera captures the direction of the ground and detects the height from the ground to the drone, or determines whether the height from the ground is before or behind the reference distance. Will be done. Based on this detection result or determination result, the thrust of the drone is controlled so that the height from the ground becomes the specified height. This allows the drone to fly at a specified height.

It should be noted that the present invention is not limited to the above embodiment as it is, and at the implementation stage, the components can be modified and embodied within a range that does not deviate from the gist thereof. In addition, various inventions can be formed by an appropriate combination of the plurality of components disclosed in the above-described embodiment. For example, some components may be removed from all the components shown in the embodiments. In addition, components from different embodiments may be combined as appropriate.

42 ... lens, 52 ... filter, 54 ... image sensor, 60 ... display, 62 ... communication unit, 64b ... depth calculation program, 64c ... effective domain notification program.

Claims (15)

An image taken by a camera including a filter having a first region transmitted through the first wavelength region and a second region transmitted through the second wavelength region, based on the first wavelength region transmitted through the first region. An input means for inputting a captured image including a one-color component image and a second color component image based on the second wavelength region transmitted through the second region.
The photographed image is provided with a processing means for outputting the photographed image in a state where an effective area for calculation of depth information based on the bias of the color information in the first color component image and the second color component image is identified.
The processing means is an electronic device that acquires or calculates the effective region based on the distance between the aperture of the lens of the camera and the filter, which is at least one imaging parameter among a plurality of imaging parameters. The electronic device according to claim 1, wherein the depth information is obtained based on the shape of the blur of the first color component image and the shape of the blur of the second color component image. The processing means transmits an effective domain acquisition request including a distance between the aperture of the lens of the camera and the filter, which is at least one imaging parameter among the plurality of imaging parameters, to an external device, and the at least one imaging. The electronic device according to claim 1, wherein the effective area corresponding to a parameter is received from the external device. The effective region obtained from the result of simulating the optical characteristic distribution of the camera based on the distance between the aperture of the lens of the camera and the filter, which is at least one of the plurality of shooting parameters, is stored. The electronic device according to claim 1, further comprising a memory to be used. The electronic device according to any one of claims 1 to 4, wherein the plurality of photographing parameters further include the sensor size of the camera, the focal length of the lens, or the brightness of the lens. The one according to any one of claims 1 to 5, wherein the filter is arranged on the subject side of the front end or the tip of the lens of the camera, or on the image sensor side of the rear end or the rear end of the lens. Electronic equipment. 6. The processing means detects the direction of an edge component of the object when an object for which depth information is desired in the captured image is designated, and outputs information according to the direction of the detected edge component. The listed electronic device. The electronic device according to claim 1, wherein the processing means extracts an image in the effective region from the captured image input by the input means and outputs the extracted image in the effective region. The processing means according to claim 1, wherein the processing means outputs guidance for shooting so that the measurement target for which depth information is obtained in the captured image is located within the effective region, by at least one of voice, text, and video. Electronics. Further, a specific means for specifying a measurement target for obtaining depth information in the captured image displayed on the display unit by any of user operation, distribution of depth information in the effective region, and object recognition from the captured image is further provided. The electronic device according to claim 1. The electronic device according to claim 1, wherein the processing means outputs a guidance recommending re-shooting when it is determined that the measurement target for which depth information is obtained in the captured image is not within the effective region. The electronic device according to claim 1, wherein the processing means expresses a guideline indicating the effective region on the captured image displayed on the display unit by at least one of color, brightness, and contrast. Further, a display unit for displaying the captured image output by the processing means is provided.
The electronic device according to claim 1, wherein the processing means compensates for a color shift of the captured image according to the effective region, and displays the compensated captured image on the display unit. Further, a display unit for displaying the captured image output by the processing means is provided.
The electronic device according to claim 1, wherein the processing means displays at least one of the reliability of the effective region and the depth estimation used for obtaining the effective region together with the captured image on the display unit. An image taken by a camera including a filter having a first region transmitted through the first wavelength region and a second region transmitted through the second wavelength region, and the first image is based on the first wavelength region transmitted through the first wavelength region. Input of a captured image including a color component image and a second color component image based on the second wavelength region transmitted through the second region, and
The color of the captured image in the first color component image and the second color component image based on the distance between the aperture of the lens of the camera and the filter, which is at least one imaging parameter among the plurality of imaging parameters. Output in a state where the effective area for the calculation of depth information based on information bias is identified.
A notification method that comprises.

Images