WO2023210702A1

WO2023210702A1 - Information processing program, information processing device, and information processing method

Info

Publication number: WO2023210702A1
Application number: PCT/JP2023/016503
Authority: WO
Inventors: 裕子石若; 智博吉田
Original assignee: ソフトバンク株式会社
Priority date: 2022-04-27
Filing date: 2023-04-26
Publication date: 2023-11-02
Also published as: JP2023162785A

Abstract

This information processing program causes a computer to execute: an acquisition procedure for acquiring reference information and a target image, said reference information being indicative of a relationship between a reference luminance value, which is a luminance value of an object region that is in a reference image including an underwater reference object irradiated with light of a light source and having been captured by an underwater image capture device and that is occupied by the reference object, and a reference distance, which is the distance from the image capture device to the reference object, said target image including an underwater target object irradiated with light from the light source and having been captured by the underwater image capture device; and an estimation procedure for estimating a target distance, which is the distance from the image capture device to the target object, on the basis of a comparison between the reference information and a target luminance value, which is the luminance value of a target object region that is in the target image and that is occupied by the target object.

Description

Information processing program, information processing device, and information processing method

The present invention relates to an information processing program, an information processing device, and an information processing method.

Image recognition technology for identifying objects in images (still images or moving images) is conventionally known. For example, in order to improve fish farming technology, there is a known technology that estimates the number of fish in a cage (hereinafter also referred to as the number of fish) by capturing images of fish in a cage with a camera and analyzing the captured images. ing.

International Publication No. 2019/045091

There is a need for technology that can accurately estimate information about underwater objects from images.

The present application has been made in view of the above, and aims to provide an information processing program, an information processing device, and an information processing method that can accurately estimate information regarding underwater objects from images.

The information processing program according to the present application provides a reference image including an underwater reference object irradiated with light from a light source, the brightness of an object area occupied by the reference object in the reference image captured by the underwater imaging device. reference information indicating a relationship between a reference brightness value, which is a value, and a reference distance, which is a distance from the imaging device to the reference object; and a target image including the underwater target irradiated with light from the light source. an acquisition procedure for acquiring a target image captured by the imaging device underwater; a target brightness value that is a brightness value in a target area occupied by the target object in the target image; and the reference information. an estimation procedure for estimating a target distance, which is a distance from the imaging device to the target object, based on a comparison with the target object.

According to one aspect of the embodiment, it is possible to accurately estimate information regarding an underwater object from an image.

FIG. 1 is a diagram for explaining the color attenuation rate in water. FIG. 2 is a diagram for explaining a reference information acquisition method according to the embodiment. FIG. 3 is a diagram illustrating an example of reference information according to the embodiment. FIG. 4 is a diagram for explaining the shape of a reference object used to obtain reference information according to the embodiment. FIG. 5 is a diagram illustrating a configuration example of an information processing apparatus according to an embodiment. FIG. 6 is a diagram illustrating an example of a target image according to the embodiment. FIG. 7 is a diagram showing an overview of information processing according to a modification. FIG. 8 is a diagram showing an overview of information processing according to a modification. FIG. 9 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device.

Hereinafter, embodiments (hereinafter referred to as "embodiments") for implementing the information processing program, information processing apparatus, and information processing method according to the present application will be described in detail with reference to the drawings. Note that the information processing program, information processing apparatus, and information processing method according to the present application are not limited to this embodiment. Further, in each of the embodiments below, the same parts are given the same reference numerals, and redundant explanations will be omitted.

(Embodiment)
[1. Introduction]
In recent years, fish farming has attracted attention as a means to solve global food problems. In order to supply high-quality fish through fish farming, it is important to accurately know the number of fish (number of fish) and the size of the fish, which are closely related to feeding (feeding the fish).

However, in the special environment of underwater aquaculture farms, it may be difficult to utilize terrestrial IT technology. Therefore, in the past, people scooped up some fish with a net, measured the number of fish (number of fish) and the size of the fish, and then counted the number of fish visually. This method has problems in that it puts a heavy burden on the fish and lacks accuracy.

Therefore, in recent years, methods that use computer vision to automatically detect the number of fish and the size of fish from images of schools of fish in cages have been attracting attention. A specific example is a method in which a machine learning model for image recognition is trained to estimate the number of fish and the size of fish from images. However, in order to accurately estimate the number of fish and the size of fish from images using a machine learning model, it is necessary for the machine learning model to learn a large amount of high-quality training data. For example, as training data for a machine learning model, a set of an image of underwater fish such as a school of fish in a cage and correct data indicating the position and size of each fish in the image is used. In this case, the high quality of the training data means that the accuracy of the correct data indicating the position and size of the fish in the image is high.

Here, it is generally known that light propagating in water does not reach far because it is attenuated due to scattering and absorption by the water. For example, it is known that among visible light (wavelengths of about 380 to 780 nm), there is an absorption band particularly in the red to orange wavelength range, and energy is absorbed. For this reason, it is known that almost 100% of red light is lost at a depth of 10 meters. It is also known that approximately 50% of the energy of blue light is absorbed at a depth of 20 meters. In this way, the red component is quickly lost in water, so the water appears blue.

FIG. 1 is a diagram for explaining the rate of color attenuation in water. The graph shown in Figure 1 shows the intensity of light with wavelengths corresponding to red, green, and blue (hereinafter also referred to as red light, green light, and blue light) in water, and the light source from the measuring equipment that measures the light intensity. This shows the relationship with the distance to. As shown in FIG. 1, it can be seen that the intensity of the red light, green light, and blue light decreases as the distance from the measuring instrument to the light source increases. Moreover, from the graph of FIG. 1, it can be seen that red light (light with a long wavelength) is particularly easily attenuated compared to green light and blue light.

As mentioned above, since light propagating underwater is easily attenuated, it is generally difficult to visually recognize an object in an image of an object underwater than in an image of an object in the air. Therefore, for example, compared to recognizing the position and size of an object in an image from an image taken of an object in the air, it is much easier to recognize the position and size of an object in an image from an image taken of an object underwater. It is difficult to estimate the size and size. In particular, it is very difficult to accurately estimate the position of the object in the depth direction (or the distance from the imaging device to the underwater object) from an image of the object underwater. As a result, it is difficult to provide high-quality correct data regarding the position and size of the object in the image to an image of the object underwater. Therefore, it is difficult to obtain high quality training data.

Therefore, there is a need for a technology that accurately recognizes the position in the depth direction of the fish in the water included in the image (or the distance from the imaging device to the fish in the water). For example, a method can be considered in which the position of a fish in the water in the depth direction is detected using another sensor outside the imaging device, and the position information in the depth direction of the fish in the water is complemented. However, most of the techniques useful in air cannot be used underwater. For example, the light from a LiDAR (Light Detection And Ranging) sensor hardly reaches underwater than it does in the air. Furthermore, ultrasonic sensors are suitable for finding schools of fish (also called schools of fish) in the water, but their resolution is low for counting the number of fish in schools of fish.

On the other hand, the information processing device 100 according to the embodiment is configured to detect an object area in which the reference object is imaged in an image of an object imitating the surface of a fish's body (hereinafter also referred to as a reference object) located underwater. Reference information indicating the relationship between the brightness value at (hereinafter also referred to as reference brightness value) and the distance from the imaging device to the reference object when the image was captured (hereinafter also referred to as reference distance) is acquired. The information processing device 100 also acquires an image of a fish in the water (hereinafter also referred to as a target image). The information processing device 100 then compares the brightness value (hereinafter also referred to as target brightness value) in the area where the fish is imaged (hereinafter also referred to as target object area) in the acquired target image with the acquired reference information. Based on the comparison, the distance from the imaging device to the fish when the target image was captured (hereinafter also referred to as target distance) is estimated. Specifically, the information processing device 100 specifies a reference distance corresponding to the same reference brightness value as the target brightness value based on the reference information, and estimates the specified reference distance as the target distance.

Thereby, the information processing device 100 can accurately estimate the distance from the imaging device to the fish in the water. Further, since the information processing device 100 can accurately estimate the distance from the imaging device to the fish in the water, for example, the information processing device 100 can estimate the distance from the imaging device to the fish in the water. It is possible to provide high accuracy data. That is, the information processing device 100 can obtain high-quality training data to which high-quality correct answer data is added, for example. Furthermore, since the information processing device 100 can obtain high-quality training data, for example, the high-quality training data is used to learn a machine learning model for image recognition that estimates information about fish in the water from images. Accuracy can be improved. Therefore, the information processing device 100 can accurately estimate information regarding fish in the water from the image.

In addition, although the case where the target image is an image of a fish in the water has been described above, the imaging target of the target image (hereinafter also referred to as a target object) is not limited to a fish in the water. For example, the object may be a living thing other than a fish. Here, if the target object is a living creature other than a fish, the reference object is replaced with an object that mimics the surface of the body of the other living creature, instead of an object that mimics the surface of the fish's body. . Furthermore, the target object is not limited to living things. For example, the target object may be an object other than a living thing. Here, when the target object is a non-living object, the reference object is replaced with an object that imitates the surface of the non-living object instead of an object that imitates the surface of the biological body.

[2. How to obtain reference information〕
FIG. 2 is a diagram for explaining a reference information acquisition method according to the embodiment. FIG. 2 shows a state in which a reference object O1, a camera C1 (an example of an imaging device), and a light source L1 are located in water such as a fish cage. The light source L1 irradiates light onto the underwater reference object O1. Camera C1 captures an image (reference image) of underwater reference object O1 irradiated with light from light source L1. If the distance from the camera C1 to the underwater reference object O1 when the reference image is captured is z, then the camera C1 images the underwater reference object O1 irradiated with light from the light source L1 while changing the value of z. . That is, the camera C1 captures an image (reference image) of the underwater reference object O1 irradiated with light from the light source L1 while moving in a direction away from the reference object O1.

Additionally, it is generally known that the reflectance of light on the surface of a fish varies depending on the type of fish scale. Therefore, fish scales (scale models or real scales) depending on the type of target fish are pasted on the surface of the reference object O1. Also, the bodies of many fish are flat on both sides. Therefore, the reflection of light by the surface of a fish is considered to be similar to the reflection of light by the surface of a flat object. Therefore, in the example shown in FIG. 2, the reference object O1 is planar (also referred to as plate-like), and the planar reference object O1 is shown from the side. In FIG. 2, the direction in which the camera C1 looks up at the reference object O1 coincides with the normal direction of the reference object O1. That is, the surface of the reference object O1 is located directly in front of the camera C1.

Furthermore, in FIG. 2, the light source L1 is provided at the bottom of the housing of the camera C1. Note that the light source L1 may be provided at the top of the housing of the camera C1. Moreover, the light source L1 is installed so that the direction in which the light source L1 irradiates light matches the direction in which the camera C1 looks up at the reference object O1.

Additionally, it is generally known that the color of light that fish prefers (or the color of light that they dislike) differs depending on the species of fish. Therefore, the light source L1 emits light of a wavelength corresponding to a color depending on the type of target fish. Specifically, the light source L1 may be a light with various color filters such as red, green, and blue.

FIG. 3 is a diagram showing an example of reference information according to the embodiment. In FIG. 3, a white light (white light), a light with a red filter (red light), a light with a blue filter (blue light), and a light with a violet filter (purple light) are used as the light source L1 shown in FIG. The four reference information obtained when

Furthermore, each graph shown in FIG. 3 corresponds to each reference information obtained by the method shown in FIG. 2. The horizontal axis of the graph indicates the distance (z) from the camera C1 to the underwater reference object O1 when the reference image was captured. The vertical axis of the graph indicates the brightness value of the reference image. More specifically, the solid line in the graph represents the brightness value of R (red), the broken line represents the brightness value of G (green), and the thick line represents the brightness value of B (blue). From each graph shown in FIG. 3, the information processing device 100 determines that, for example, when the brightness value of R of the target brightness value in the target object region where the fish is photographed in the target image is close to zero, the target image is photographed. The target distance from the imaging device to the fish can be estimated to be 3 m or more. Furthermore, if the target distance from the imaging device to the fish when the target image is captured is between 1 m and 2 m, the information processing device 100 can, for example, use the solid line in the graph of the light with a red filter (red light). The target distance can be estimated based on the comparison with the brightness value of R (red) shown by .

FIG. 4 is a diagram for explaining the shape of a reference object used to obtain reference information according to the embodiment. In FIG. 2, for the sake of simplicity, the case where the reference object O1 is planar has been described, but the actual reference object O1 has a shape of a truncated quadrangular pyramid without a bottom surface as shown in FIG. More specifically, the actual reference object O1 is composed of five surfaces, surface #1 to surface #5. Four surfaces, surface #2 to surface #5, are provided to surround surface #1. As in Figure 2, the surface of each side is covered with fish scales (scale models or real scales, pseudo skin (aurora bald skin), color charts, checkerboard, etc.) depending on the type of fish being targeted. is pasted. Surface #1 shown in FIG. 4 corresponds to reference object O1 shown in FIG. 2. Specifically, surface #1 is a surface located directly in front of the camera C1. The four surfaces #2 to #5 are installed at an angle with respect to surface #1. That is, the four surfaces #2 to #5 each have a different angle from the surface #1, which is located directly in front of the camera C1. Further, in general, the reflectance of light differs depending on the angle of the surface on which the light is incident. For example, when the fish's body is oriented directly sideways to the imaging device, the reflectance from the surface of the fish's body is the highest. In this way, when acquiring reference information, by using the reference object O1 having a truncated quadrangular pyramid shape without a base as shown in FIG. 4, it is possible to acquire reference information for each surface having a different angle with respect to the camera C1. I can do it. As a result, the information processing device 100 can use the reference information obtained from the four surfaces #2 to #5 even if the fish's body is oriented in a direction other than directly sideways relative to the imaging device, for example. Based on the information, the target distance from the imaging device to the fish can be estimated.

[3. Configuration of information processing device]
FIG. 5 is a diagram illustrating a configuration example of the information processing device 100 according to the embodiment. The information processing device 100 includes a communication section 110, a storage section 120, an input section 130, an output section 140, and a control section 150.

(Communication Department 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card). The communication unit 110 is connected to the network by wire or wirelessly, and transmits and receives information to and from a terminal device used by, for example, a fish manager.

(Storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. For example, the storage unit 120 stores an information processing program according to the embodiment.

(Input section 130)
The input unit 130 receives various operations from the user. For example, the input unit 130 may receive various operations from the user via a display screen (for example, the output unit 140) using a touch panel function. Further, the input unit 130 may accept various operations from buttons provided on the information processing device 100 or a keyboard or mouse connected to the information processing device 100.

(Output section 140)
The output unit 140 is a display screen realized by, for example, a liquid crystal display or an organic EL (Electro-Luminescence) display, and is a display device for displaying various information. The output unit 140 displays various information under the control of the control unit 150. Note that when a touch panel is employed in the information processing apparatus 100, the input section 130 and the output section 140 are integrated. Furthermore, in the following description, the output unit 140 may be referred to as a screen.

(Control unit 150)
The control unit 150 is a controller, and for example, executes various programs (information processing programs) stored in a storage device inside the information processing device 100 by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. (corresponding to one example) is realized by being executed using RAM as a work area. Further, the control unit 150 is a controller, and is realized by, for example, an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

The control unit 150 has an acquisition unit 151, an estimation unit 152, and an output control unit 153 as functional units, and may realize or execute the information processing operation described below. Note that the internal configuration of the control unit 150 is not limited to the configuration shown in FIG. 5, and may be any other configuration as long as it performs information processing to be described later. Further, each functional unit indicates a function of the control unit 150, and does not necessarily have to be physically distinct.

(Acquisition unit 151)
The acquisition unit 151 obtains a reference image including an underwater reference object irradiated with light from a light source, which is a reference brightness value that is a brightness value in an object area occupied by the reference object in the reference image captured by an underwater imaging device. Reference information indicating the relationship between the value and a reference distance, which is the distance from the imaging device to the reference object, is acquired. Specifically, the acquisition unit 151 indicates the relationship between a reference brightness value in a reference image including a reference object irradiated with light of a wavelength corresponding to a color corresponding to a different fish for each species and a reference distance. Get reference information. For example, the acquisition unit 151 acquires reference information shown by each graph shown in FIG. 3. For example, the acquisition unit 151 acquires reference information from a terminal device used by a user who has acquired the reference information through an experiment.

Furthermore, the acquisition unit 151 acquires a target image that includes an underwater fish irradiated with light from a light source and that is captured by an underwater imaging device. Specifically, the acquisition unit 151 acquires a target image including a fish irradiated with light of a wavelength corresponding to a color corresponding to a different fish species. For example, the acquisition unit 151 acquires a target image including a plurality of fish. FIG. 6 is a diagram illustrating an example of a target image according to the embodiment. For example, the acquisition unit 151 acquires a target image G1 including a plurality of fish as shown in FIG. For example, the acquisition unit 151 acquires the target image from the imaging device that captured the target image.

(Estimation unit 152)
The estimation unit 152 estimates the target distance, which is the distance from the imaging device to the fish, based on a comparison between the target brightness value, which is the brightness value in the target object area occupied by the fish, and the reference information in the target image. Specifically, the estimation unit 152 specifies a reference distance corresponding to the same reference brightness value as the target brightness value based on the reference information, and estimates the specified reference distance as the target distance.

Furthermore, the estimation unit 152 estimates the size of the fish according to the estimated target distance. For example, suppose that there is an image that includes two target object regions in which the size of the target object region occupied by a fish is approximately the same, and the brightness values in the target object regions are different. At this time, the estimation unit 152 calculates that the larger (smaller) the brightness value in the target object region, the closer (farther) the distance from the imaging device to the fish is. Also, it is estimated that the size of the object corresponding to the object region having a small brightness value is larger.

Furthermore, the estimation unit 152 calculates the target distance from the imaging device to each of the plurality of fish based on the comparison between the target brightness value in each of the plurality of target object areas occupied by each of the plurality of fish in the target image and the reference information. is estimated, and each of the plurality of fish is identified according to the estimated target distance from the imaging device to each of the plurality of fish. For example, assume that there is a target image in which a plurality of different fish are photographed overlapping each other and includes two target object regions in which the brightness values of the target object regions occupied by the plurality of fish are different. At this time, the estimation unit 152 estimates that the larger (smaller) the brightness value in the target object area is, the closer (farther) the distance from the imaging device to the fish is. , it can be estimated that the fish is located in front of the position of the fish corresponding to the target area with the small brightness value. In other words, the estimating unit 152 can estimate that the position of the fish corresponding to the object area with a small brightness value is located further back than the position of the fish corresponding to the object area with a large brightness value. Furthermore, in the case of fish whose shapes and sizes are known to some extent (for example, fish in a cage), it is obvious that different positions in three-dimensional space indicate different fish. Therefore, the estimation unit 152 accurately identifies an object near the imaging device (for example, a predetermined fish) and an object located far from the imaging device (for example, another fish different from the predetermined fish). be able to.

(Output control unit 153)
The output control unit 153 controls the estimation result estimated by the estimation unit 152 to be displayed on the screen. For example, the output control unit 153 may control the screen to display a numerical value indicating the target distance estimated by the estimating unit 152 in a superimposed manner on the target object area.

[4. Modified example]
The information processing device 100 according to the embodiment described above may be implemented in various different forms other than the embodiment described above. Therefore, other embodiments of the information processing device 100 will be described below. Note that the same parts as those in the embodiment are given the same reference numerals and the description thereof will be omitted.

[4-1. Machine learning model]
In the embodiment described above, a case has been described in which the estimation unit 152 estimates the distance from the imaging device to the fish based on the comparison between the brightness value in the target object area occupied by the fish in the target image and the reference information. Here, as a modified example of the embodiment, the estimation unit 152 uses a machine learning model to determine the size and size of the fish included in the target image based on the brightness value in the target area occupied by the fish in the target image. A case of estimating the number of fish (namely, the number of fish) will be explained.

FIG. 7 is a diagram showing an overview of information processing according to a modification. In FIG. 7, when an image including fish is input to the machine learning model M1, the estimation unit 152 outputs position information of each fish, information indicating the size, and each brightness value of (R, G, B). A machine learning model M1 trained to do the following is acquired. Next, the estimation unit 152 uses the machine learning model M1 to estimate the size and number of fish (ie, the number of fish) included in the target image.

FIG. 8 is a diagram showing an overview of information processing according to a modification. In FIG. 8, when an image containing a fish and a point cloud generated from the image containing a fish are input to the machine learning model M2, the estimation unit 152 calculates the position information of each fish, information indicating the size, (R, G , B), and a machine learning model M2 trained to output the position information of each point in the point cloud. Subsequently, the estimating unit 152 uses the machine learning model M2 to estimate the size of the fish and the number of fish (ie, the number of fish) included in the target image.

[4-2. Light source that turns full color LED on/off at high speed]
In the embodiment described above, a case has been described in which the light source emits light of a wavelength corresponding to a color depending on the type of target fish. Here, a case will be described in which the light source is a light source that turns on/off a full-color LED at high speed. That is, a case will be described in which a light source irradiates a fish with a plurality of lights of different wavelengths (that is, lights of different colors) at almost the same timing over a very short period of time. Specifically, the acquisition unit 151 calculates the reference brightness values and reference distances in each of a plurality of different reference images each including a reference object irradiated with light of a plurality of different wavelengths corresponding to each of a plurality of different colors. Obtain multiple different references that indicate relationships. Further, the acquisition unit 151 acquires a plurality of different target images each including a target object irradiated with light of a plurality of different wavelengths at different times within a predetermined time. The estimation unit 152 estimates the target distance based on a comparison between the target brightness value in each of a plurality of different target images and each of a plurality of different reference information.

[4-3. Estimation of transparency based on brightness value]
In the embodiment described above, a case has been described in which the estimation unit 152 estimates the distance from the imaging device to the fish based on the comparison between the brightness value in the target object area occupied by the fish in the target image and the reference information. Here, as a modified example of the embodiment, a case will be described in which the estimating unit 152 estimates the underwater transparency based on a comparison between the luminance value in the target object area occupied by the white board in the target image and reference information. .

First, transparency will be explained. Transparency in this specification refers to the distance that can be seen in the horizontal direction with respect to the bottom surface underwater. Specifically, a white planar (also referred to as a plate-shaped) object (hereinafter referred to as a white board) is submerged at a predetermined depth (for example, 1 m depth) underwater. For example, the white board may be a disc with a diameter of 30 cm. Then, the distance between the eyes of a person located at a predetermined depth underwater (for example, 1 m depth, etc.) and the white board is gradually increased underwater in the horizontal direction with respect to the bottom surface. Then, the distance between the human eye and the white board when the white board is no longer visible to the human eye is defined as the underwater visibility.

Next, transparency will be explained. Transparency as used herein refers to the distance that can be seen vertically when looking into the water from above the water surface. Specifically, the white board mentioned above is submerged in water. Then, the distance between the water surface and the white board is gradually increased in the direction perpendicular to the water surface. The distance between the water surface and the white board when the white board is no longer visible to the human eye is defined as the underwater transparency.

Specifically, in this modification, the reference object O1 shown in FIG. 2 is the above-mentioned white board. In a modified example, a measuring device is prepared in which the distance between the white board, which is the reference object O1 shown in FIG. 2, and the light source L1 and camera C1 is fixed at a measurement distance (for example, 1 m). Furthermore, the measuring device is submerged underwater at various locations in each season (that is, underwater corresponding to various degrees of visibility). Then, the camera C1 of the measuring device submerged at a predetermined depth (for example, 1 m depth) captures a transparency reference image including the underwater white board irradiated with light from the light source L1. In addition, the transparency of the water when the transparency reference image is captured is measured through experiments. In this way, perspective reference images corresponding to each depth (for example, 1 m depth, 2 m depth, 3 m depth, etc.) in the sea (that is, underwater corresponding to various transparency degrees) in each season, and Basic transparency information, which correlates the underwater visibility when the degree reference image is captured, is obtained through experiments.

Additionally, the acquisition unit 151 acquires transparency basic information. For example, the acquisition unit 151 acquires basic transparency information from a terminal device used by a user of the measuring device. Next, based on the visibility basic information, the acquisition unit 151 obtains a visibility reference brightness value, which is a brightness value in the area occupied by the white board, in the visibility reference image, and a value in the water when the visibility reference image is captured. Obtain transparency reference information that indicates the relationship with transparency.

Also, the above measuring device is submerged at a predetermined depth (for example, 1 m depth, etc.) underwater on the day on which the visibility is desired to be measured (hereinafter referred to as the measurement date). Then, an object image including an underwater white board irradiated with light from a light source L1 by a camera C1 of a measurement device submerged at a predetermined depth (for example, 1 m depth, etc.), the object imaged by the underwater camera C1. Capture an image.

Additionally, the acquisition unit 151 acquires a target image. For example, the acquisition unit 151 acquires a target image from a terminal device used by a user of the measuring device. In addition, the estimation unit 152 calculates the transparency on the measurement date based on the comparison between the target brightness value, which is the brightness value in the target object area occupied by the white board, and the transparency reference information in the target image acquired by the acquisition unit 151. Estimate. The estimation unit 152 estimates the visibility on the measurement date based on the visibility reference information corresponding to a predetermined depth (for example, a depth of 1 m, etc.). Thereby, the information processing apparatus 100 can estimate the degree of transparency based on the quantified information without using visual measurement.

Specifically, the estimation unit 152 selects at least one of the R brightness value, the G brightness value, and the B brightness value in the target object area occupied by the white board as the target brightness value in the target image. The transparency on the measurement date is estimated based on the comparison between the brightness value of the color and the transparency reference information. Here, in water with high visibility, the attenuation rate of R (red) light is the highest among R (red), G (green), and B (blue), and it is attenuated from R (red) light. It has been known. Further, it is known that in water where visibility is high, the attenuation rate of B (blue) light is the smallest, and the B (blue) light remains until the end. Therefore, if the visibility on the measurement date exceeds the first visibility, the estimation unit 152 calculates the target brightness value, which is the brightness value of R (red) in the target object area occupied by the white board, in the target image, and the visibility reference. Based on the comparison with the information, the visibility at the measurement date is estimated. For example, based on the perspective reference information, the estimation unit 152 calculates a perspective view corresponding to the perspective reference brightness value of R (red) that is the same as the target brightness value that is the brightness value of R (red) in the target object area occupied by the white board. The specified visibility is estimated as the visibility on the measurement date.

Furthermore, when the visibility on the measurement date is equal to or lower than the first visibility, the estimation unit 152 calculates the visibility on the measurement date based on the luminance values of a plurality of colors in the target object area occupied by the white board in the target image. presume. For example, based on the perspective reference information, the estimating unit 152 calculates a perspective view corresponding to a B (blue) perspective reference brightness value that is the same as the B (blue) brightness value in the target object region occupied by the white board. Identify degree. Here, one of the reasons for the low visibility is that there is a lot of chlorophyll such as algae in the water. In addition, chlorophyll such as algae in water has the property of easily absorbing blue light. Therefore, because there is a lot of chlorophyll such as algae, B (blue) light is attenuated in water where visibility is low. Therefore, the visibility estimated based only on the brightness value of B (blue) may take a value that is actually impossible (for example, it takes a value of visibility that exceeds the size of the fish tank, such as 200 m). In this way, if the transparency specified based on the luminance value of one color in the object area occupied by the white board is equal to or lower than the first transparency, the estimating unit 152 determines whether the other object in the object area occupied by the white board is The transparency on the measurement date is estimated based on the luminance value of the color. For example, based on the perspective reference information, the estimation unit 152 calculates a perspective view corresponding to the perspective reference brightness value of R (red) that is the same as the target brightness value that is the brightness value of R (red) in the target object area occupied by the white board. The specified visibility is estimated as the visibility on the measurement date.

Furthermore, the estimating unit 152 estimates the transparency on the measurement date based on the correlation between the brightness values of a plurality of colors in the target object area occupied by the white board, as the target brightness value in the target image. For example, the estimation unit 152 calculates the ratio of brightness values of a plurality of colors at each transparency based on the transparency reference information. For example, the estimation unit 152 calculates the brightness value of R (hereinafter sometimes abbreviated as R), the brightness value of G (hereinafter sometimes abbreviated as G), and the brightness value of B (hereinafter sometimes abbreviated as B). (sometimes abbreviated) is calculated for each transparency level. For example, the estimation unit 152 calculates R:G:B=1:2:3 when the visibility is 30 m based on the visibility reference information. Furthermore, the estimation unit 152 calculates R:G:B=1:2:2 when the visibility is 20 m based on the visibility reference information. Furthermore, the estimation unit 152 calculates R:G:B=1:1:1 when the visibility is 10 m based on the visibility reference information. The estimation unit 152 also compares the ratio of the brightness values of the plurality of colors at each perspective based on the transparency reference information with the ratio of the brightness values of the plurality of colors in the target object area occupied by the white board in the target image. Based on this, estimate the transparency on the measurement date. For example, if the estimation unit 152 calculates that R:G:B=1:1:1 in the target object area occupied by the white board in the target image, the estimation unit 152 calculates that R:G:B at each perspective based on the perspective reference information. Based on the comparison with the ratio of , the visibility on the day of measurement is estimated to be 10 m.

Additionally, when acquiring the above-mentioned basic visibility information through experiments, the transparency of the day is measured in addition to the visibility. Then, information that associates the underwater visibility and transparency when the visibility reference image is captured is obtained through an experiment. In addition, the transparency on the day of measurement is measured experimentally.

The acquisition unit 151 also acquires information that associates underwater visibility with transparency, and information on transparency on the measurement date. For example, the acquisition unit 151 acquires information associating underwater visibility with transparency, and information on transparency on the measurement date from a terminal device used by a user of the measuring device. Furthermore, the estimation unit 152 estimates the transparency on the measurement date from the transparency on the measurement date. More specifically, the estimating unit 152 identifies the transparency corresponding to the same transparency as the transparency on the measurement date based on information that associates underwater visibility with transparency, and sets the identified transparency on the measurement date. Estimated as the transparency at .

[4-4. Estimation of target distance based on transparency]
In the embodiment described above, a case has been described in which the estimation unit 152 estimates the distance from the imaging device to the fish based on the comparison between the brightness value in the target object area occupied by the fish in the target image and the reference information. Here, as a modification of the embodiment, a case will be described in which the estimation unit 152 estimates the distance from the imaging device to the fish based on the visibility of the water.

In general, the attenuation rate of light in water differs depending on the amount of suspended matter such as chlorophyll in the water. In other words, the attenuation rate of each color of RGB in water with a small amount of suspended matter and high visibility (for example, visibility of 20 m), and the attenuation rate of light of each RGB color in water with a large amount of suspended matter and low visibility (for example, visibility of 3 m). The attenuation rate of light of each color of RGB is different. Therefore, even if fish of the same species are located at the same distance from the imaging device, the way the fish is seen in water with a low amount of suspended matter and high visibility is different from the way the fish is seen in water with a large amount of suspended matter and low visibility. is different. Therefore, an attenuation rate map (hereinafter also referred to as reference basic information) of each color of RGB in water at each transparency level is prepared in advance.

Specifically, as reference basic information, for each depth in the water at each transparency level (for example, 1 m depth, 2 m depth, 3 m depth, etc.), a reference object with scales attached (as shown in Fig. 2 above) is obtained from the imaging device. A reference image at each distance to the reference object O1) is captured. Note that the reference image for acquiring the reference basic information is captured for a reference object to which materials such as scales or pseudo skin (aurora bald skin), color charts, checkerboards, etc., which differ for each species of fish, are pasted. In addition, the degree of transparency when capturing the reference image is measured through experiments. The acquisition unit 151 obtains, as reference basic information, the transparency when the reference image was captured, the depth when the reference image was captured, the distance from the imaging device to the reference object, and the area occupied by the reference object in the reference image. Information that associates each value of R, G, and B in the area is acquired.

In addition, data not included in the reference basic information will be supplemented by calculation. For example, the brightness values of RGB when the visibility is 1m are (R brightness value, G brightness value, B brightness value) = (5, 5, 5), and the RGB brightness values when the visibility is 2m. It is assumed that the respective brightness values of (R brightness value, G brightness value, B brightness value) = (10, 10, 10). Further, it is assumed that there is no data for each RGB luminance value with a visibility of 1.5 m. At this time, each RGB brightness value when the visibility is 1.5m is calculated based on the RGB brightness values when the visibility is 1m and the RGB brightness values when the visibility is 2m. It may be estimated that (brightness value, brightness value of G, brightness value of B) = (7.5, 7.5, 7.5). Note that although this example is linear interpolation, non-linear interpolation is also possible.

In addition, the visibility in the fish tank on the day when the distance from the imaging device to the fish is estimated (hereinafter referred to as the estimation day) is measured. Note that the estimation unit 152 performs the above [4-3. [Estimation of visibility based on brightness value]] may be used to estimate the visibility on the estimated date. In addition, the species of fish kept in the fish tank are known. Further, it is assumed that the depth of the imaging device is known. The acquisition unit 151 acquires reference basic information corresponding to the visibility on the estimated date. The acquisition unit 151 also acquires a target image in the fish tank on the estimated date. Furthermore, the estimation unit 152 calculates the target distance, which is the distance from the imaging device to the fish, based on the comparison between the target brightness value, which is the brightness value in the target object area occupied by the fish, and the reference basic information in the target image. presume. Specifically, the estimation unit 152 compares the brightness value of at least one color among the brightness value of R, the brightness value of G, and the brightness value of B in the target object region with reference basic information. Estimate the target distance based on. For example, if the visibility on the estimated date exceeds the first visibility, the estimation unit 152 calculates the target brightness value, which is the brightness value of R in the target object region of the target image, based on the comparison with the reference basic information. , estimate the target distance. For example, the estimation unit 152 identifies the distance from the imaging device to the reference object corresponding to the same R brightness value as the target brightness value, which is the R brightness value in the target object region, based on the reference basic information. Estimate the distance as the target distance. Furthermore, when the visibility on the estimation date is equal to or lower than the first visibility, the estimation unit 152 estimates the target distance based on the luminance values of the plurality of colors in the target object area. For example, the estimation unit 152 estimates the target image based on a comparison between the ratio of the brightness values of a plurality of colors at each distance based on the reference basic information and the ratio of the brightness values of a plurality of colors in the target object area of the target image. Estimate distance. In this way, the estimation unit 152 may estimate the distance from the imaging device to the fish based on the visibility of the water.

[5. effect〕
As described above, the information processing apparatus 100 according to the embodiment includes the acquisition section 151 and the estimation section 152. The acquisition unit 151 obtains a reference image including an underwater reference object irradiated with light from a light source, which is a reference brightness value that is a brightness value in an object area occupied by the reference object in the reference image captured by an underwater imaging device. reference information indicating the relationship between the value and the reference distance, which is the distance from the imaging device to the reference object, and a target image including an underwater object illuminated by light from a light source, which is a distance from the imaging device to the reference object. Obtain the captured target image. The estimation unit 152 estimates the target distance, which is the distance from the imaging device to the target object, based on the comparison between the target brightness value, which is the brightness value in the target area occupied by the target object, and reference information in the target image. do.

Thereby, the information processing device 100 can accurately estimate the distance from the imaging device to the underwater object. In addition, since the information processing device 100 can accurately estimate the distance from the imaging device to the underwater object, for example, the information processing device 100 can estimate the position of the object in the image, It is possible to provide high quality correct answer data regarding the size. That is, the information processing apparatus 100 can obtain, for example, high-quality training data to which high-quality correct answer data is added. Furthermore, since the information processing device 100 can obtain high-quality training data, for example, the high-quality training data can be used to develop a machine learning model for image recognition that estimates information about underwater objects from images. Learning accuracy can be improved. Therefore, the information processing device 100 can accurately estimate information regarding the underwater object from the image. Furthermore, since the information processing device 100 can accurately estimate information about underwater objects from images, it can help achieve Goal 9 of the Sustainable Development Goals (SDGs), “Create a foundation for industry and technological innovation.” I can contribute. In addition, the information processing device 100 can, for example, accurately estimate information about fish in the water from images, so that it can help achieve Goal 14 of the Sustainable Development Goals (SDGs), “Let's protect the richness of the oceans.” I can contribute.

Furthermore, the estimation unit 152 estimates the size of the target object according to the estimated target distance.

Thereby, the information processing device 100 can accurately estimate the size of the underwater object from the image. For example, it is assumed that there is an image that includes two object areas in which the size of the object area occupied by the object is approximately the same, but the brightness value in the object area is different. At this time, the information processing device 100 detects the target object corresponding to the object region having a large brightness value because the larger (smaller) the brightness value in the target object region is, the closer (farther) the distance from the imaging device to the target object is. It can be estimated that the size of the object corresponding to the object region with a small brightness value is larger than the size. In this way, the information processing device 100 can determine the size of a small object (e.g., a predetermined fish) near the imaging device and the size of a large object (e.g., other than the predetermined fish) located far from the imaging device. The size of each fish can be estimated with high accuracy.

Additionally, the acquisition unit 151 acquires a target image including a plurality of targets. The estimation unit 152 calculates the target distance from the imaging device to each of the plurality of objects based on the comparison between the target brightness values in each of the plurality of object regions occupied by each of the plurality of objects in the target image and reference information. is estimated, and each of the plurality of objects is identified according to the estimated object distance from the imaging device to each of the plurality of objects.

Thereby, the information processing device 100 can accurately identify multiple underwater objects from the image. For example, assume that there is an image that includes two object areas in which a plurality of different objects overlap and the object areas occupied by the plurality of objects each have different brightness values. At this time, the information processing device 100 detects the target object corresponding to the object region having a large brightness value because the larger (smaller) the brightness value in the target object region is, the closer (farther) the distance from the imaging device to the target object is. It can be estimated that the position of the target object is located closer to the viewer than the position of the target object corresponding to the target object area with the lower luminance value. In other words, the information processing device 100 estimates that the position of the object corresponding to the object area with a small brightness value is located further back than the position of the object corresponding to the object area with a large brightness value. be able to. In this way, the information processing device 100 can determine the position of an object near the imaging device (for example, a predetermined fish) and the position of an object far away from the imaging device (for example, another fish different from the predetermined fish). The positions of each can be identified with high accuracy. Furthermore, in the case of objects whose shapes and sizes are known to some extent (for example, fish in a cage), different positions in three-dimensional space mean that the objects are different. Therefore, the information processing device 100 accurately identifies an object near the imaging device (for example, a predetermined fish) and an object located far from the imaging device (for example, another fish different from the predetermined fish). can do.

In addition, the acquisition unit 151 obtains reference information indicating the relationship between a reference brightness value in a reference image including a reference object irradiated with light of a wavelength corresponding to a color corresponding to the target object, and a reference distance; A target image including a target object irradiated with light of a wavelength corresponding to a color corresponding to the color is acquired. The estimation unit 152 estimates the target distance based on a comparison between the target brightness value in the target image and reference information.

Thereby, the information processing device 100 can prevent the fish from escaping from within the imaging range while the image capturing device is capturing an image by emitting light that the fish prefers, so the information processing device 100 can emit light. It is possible to obtain an appropriate target image containing the fish. Furthermore, since the information processing apparatus 100 can acquire an appropriate target image, it can accurately estimate the target distance.

The acquisition unit 151 also obtains the relationship between the reference brightness value and the reference distance in each of a plurality of different reference images each including a reference object irradiated with light of a plurality of different wavelengths corresponding to each of a plurality of different colors. A plurality of different target images are acquired, each including a plurality of different reference information shown and a target object irradiated with light of a plurality of different wavelengths at different times within a predetermined time. The estimation unit 152 estimates the target distance based on a comparison between the target brightness value in each of a plurality of different target images and each of a plurality of different reference information.

Thereby, the information processing device 100 can irradiate a target object with a plurality of lights of different wavelengths (that is, lights of different colors) in a very short period of time. That is, even if the target object is a fish and different fish species prefer different colors, the information processing device 100 prevents the fish from escaping from within the imaging range while the imaging device captures the image. Therefore, it is possible to obtain an appropriate target image including the fish illuminated with light. Furthermore, since the information processing apparatus 100 can acquire an appropriate target image, it can accurately estimate the target distance.

Additionally, the light source is provided at the top or bottom of the casing of the imaging device.

With this, the information processing device 100 can, for example, align the direction in which the imaging device looks up at the object and the direction in which the light source irradiates the object, making it easier to capture images.

Further, the reference object is a fish scale of a specific fish species or a model of a fish scale, or an object covered with pseudo-skin, a color chart, or a checkerboard. The object is a specific species of fish.

As a result, the information processing device 100 can acquire appropriate reference information that differs for each species of fish, and therefore can accurately estimate information regarding fish in the water from the image.

The acquisition unit 151 also acquires a visibility reference image that includes an underwater reference object irradiated with light from a light source, and that is located in an object area occupied by the reference object in the visibility reference image captured by the underwater imaging device. Underwater imaging, which is a target image that includes a visibility reference brightness value, which is a brightness value, and visibility reference information that indicates the relationship between underwater visibility, and an underwater object that is irradiated with light from a light source. A target image captured by the device is acquired. The estimation unit 152 determines the target image based on the comparison between the brightness value of at least one color among the brightness values of a plurality of colors in the target area occupied by the target object in the target image and the transparency reference information. Estimate the visibility of the water when the image was captured.

As a result, the information processing device 100 can estimate the visibility of the water with high accuracy, so that, for example, the distance from the imaging device to the underwater object can be estimated with more accuracy.

Furthermore, the estimation unit 152 estimates the target distance based on the estimated underwater visibility.

Thereby, the information processing device 100 can estimate the distance from the imaging device to the underwater object with higher accuracy.

Additionally, the information processing device 100 according to the modification includes an acquisition unit 151 and an estimation unit 152. The acquisition unit 151 obtains a visibility reference image including an underwater reference object irradiated with light from a light source, which is a luminance value in an object area occupied by the reference object in the visibility reference image captured by the underwater imaging device. A target image that includes a visibility reference brightness value, visibility reference information indicating the relationship between the underwater visibility, and an underwater object irradiated with light from a light source, which is captured by an underwater imaging device. Obtain the captured target image. The estimation unit 152 determines the target image based on the comparison between the brightness value of at least one color among the brightness values of a plurality of colors in the target area occupied by the target object in the target image and the transparency reference information. Estimate the visibility of the water when the image was captured.

Furthermore, if the visibility in the water when the target image is captured exceeds the first visibility, the estimation unit 152 calculates the brightness value of one color in the target area occupied by the target object in the target image and the visibility. Based on the comparison with the reference information, the degree of transparency in the water when the target image was captured is estimated.

Thereby, the information processing device 100 can appropriately estimate the visibility of the water depending on the visibility of the water.

Furthermore, if the transparency of the water when the target image is captured is below the first perspective, the estimation unit 152 calculates the luminance values of the plurality of colors in the target area occupied by the target object in the target image and the perspective Based on the comparison with the degree reference information, the degree of visibility of the water at the time the target image was captured is estimated.

[6. Hardware configuration]
Further, the information processing apparatus 100 according to the embodiments described above is realized by, for example, a computer 1000 having a configuration as shown in FIG. FIG. 9 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device 100. The computer 1000 includes a CPU 1100, a RAM 1200, a ROM 1300, an HDD 1400, a communication interface (I/F) 1500, an input/output interface (I/F) 1600, and a media interface (I/F) 1700.

The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400 and controls each part. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started, programs depending on the hardware of the computer 1000, and the like.

The HDD 1400 stores programs executed by the CPU 1100 and data used by the programs. Communication interface 1500 receives data from other devices via a predetermined communication network and sends it to CPU 1100, and transmits data generated by CPU 1100 to other devices via a predetermined communication network.

The CPU 1100 controls output devices such as a display and printer, and input devices such as a keyboard and mouse via the input/output interface 1600. CPU 1100 obtains data from an input device via input/output interface 1600. Further, CPU 1100 outputs the generated data to an output device via input/output interface 1600.

The media interface 1700 reads programs or data stored in the recording medium 1800 and provides them to the CPU 1100 via the RAM 1200. CPU 1100 loads this program from recording medium 1800 onto RAM 1200 via media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. etc.

For example, when the computer 1000 functions as the information processing device 100 according to the embodiment, the CPU 1100 of the computer 1000 realizes the functions of the control unit 150 by executing a program loaded onto the RAM 1200. The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs may be acquired from another device via a predetermined communication network.

Some of the embodiments of the present application have been described above in detail based on the drawings, but these are merely examples, and various modifications and variations may be made based on the knowledge of those skilled in the art, including the embodiments described in the disclosure section of the invention. It is possible to carry out the invention in other forms with modifications.

[7. others〕
Furthermore, among the processes described in the above embodiments and modified examples, all or part of the processes described as being performed automatically can be performed manually, or may be described as being performed manually. All or part of this processing can also be performed automatically using known methods. In addition, information including the processing procedures, specific names, and various data and parameters shown in the above documents and drawings may be changed arbitrarily, unless otherwise specified. For example, the various information shown in each figure is not limited to the illustrated information.

Furthermore, each component of each device shown in the drawings is functionally conceptual, and does not necessarily need to be physically configured as shown in the drawings. In other words, the specific form of distributing and integrating each device is not limited to what is shown in the diagram, and all or part of the devices can be functionally or physically distributed or integrated in arbitrary units depending on various loads and usage conditions. Can be integrated and configured.

Further, the information processing device 100 described above may be realized by a plurality of server computers, and depending on the function, it may be realized by calling an external platform etc. using an API (Application Programming Interface), network computing, etc. can be changed flexibly.

Furthermore, the embodiments and modifications described above can be combined as appropriate within the scope of not contradicting the processing contents.

100 Information processing device 110 Communication unit 120 Storage unit 130 Input unit 140 Output unit 150 Control unit 151 Acquisition unit 152 Estimation unit 153 Output control unit

Claims

a reference image including an underwater reference object irradiated with light from a light source, a reference brightness value that is a brightness value in an object region occupied by the reference object among the reference images captured by the underwater imaging device; Reference information indicating a relationship with a reference distance that is a distance from the imaging device to the reference object, and a target image including the underwater object irradiated with light from the light source, an acquisition procedure for acquiring a target image captured by an imaging device;
Estimating a target distance, which is a distance from the imaging device to the target object, based on a comparison between a target brightness value, which is a brightness value in a target object area occupied by the target object, in the target image, and the reference information. an estimation procedure to
An information processing program that causes a computer to execute.
The estimation procedure is
estimating the size of the target object according to the estimated target distance;
The information processing program according to claim 1.
The acquisition procedure is as follows:
acquiring the target image including a plurality of the target objects;
The estimation procedure is
Based on the comparison between the target brightness values in each of the plurality of target object areas occupied by each of the plurality of target objects in the target image and the reference information, the distance from the imaging device to each of the plurality of target objects is estimating the object distance and identifying each of the plurality of objects according to the estimated object distance from the imaging device to each of the plurality of objects;
The information processing program according to claim 1.
The acquisition procedure is as follows:
the reference information indicating the relationship between the reference brightness value in the reference image including the reference object irradiated with the light of the wavelength corresponding to the color corresponding to the target object and the reference distance; obtaining the target image including the target object irradiated with the light having the wavelength corresponding to the color corresponding to the object;
The estimation procedure is
estimating the target distance based on a comparison between the target brightness value in the target image and the reference information;
The information processing program according to claim 1.
The acquisition procedure is as follows:
A plurality of information indicating the relationship between the reference brightness value and the reference distance in each of the plurality of different reference images each including the reference object irradiated with the light of a plurality of different wavelengths corresponding to each of a plurality of different colors. and obtaining a plurality of different target images each including the target object irradiated with each of the plurality of lights of different wavelengths at different times within a predetermined time;
The estimation procedure is
estimating the target distance based on a comparison between the target brightness value in each of the plurality of different target images and each of the plurality of different reference information;
The information processing program according to claim 1.
The light source is provided at the top or bottom of the casing of the imaging device,
The information processing program according to claim 1.
The reference object is a fish scale of a specific fish species, a model of the fish scale, or an object covered with pseudo skin, a color chart, or a checkerboard,
The target object is a fish of the specific fish species,
The information processing program according to claim 1.
The acquisition procedure is as follows:
A visibility reference image including an underwater reference object irradiated with light from the light source, which is a brightness value in an object area occupied by the reference object in the visibility reference image captured by the underwater imaging device. Transparency reference information indicating a relationship between a transparency reference brightness value and the transparency in the water, and a target image including an underwater object irradiated with light from the light source, the image being captured in the water. Obtain the target image captured by the device,
The estimation procedure is
The target image is calculated based on a comparison between the brightness value of at least one color among the brightness values of a plurality of colors in the target object area occupied by the target object in the target image and the transparency reference information. Estimate the underwater visibility when the image is taken.
The information processing program according to claim 1.
The estimation procedure is
estimating the target distance based on the estimated underwater visibility;
The information processing program according to claim 8.
A perspective reference image that includes an underwater reference object irradiated with light from a light source, which is a luminance value in an object area occupied by the reference object among the perspective reference images captured by the underwater imaging device. a target image including visibility reference information indicating a relationship between a degree reference brightness value and the underwater visibility, and an underwater object irradiated with light from the light source, the image capturing device underwater; an acquisition procedure for acquiring a target image captured by;
The target image is calculated based on a comparison between the brightness value of at least one color among the brightness values of a plurality of colors in the target object area occupied by the target object in the target image and the transparency reference information. an estimation procedure for estimating underwater visibility when imaged;
An information processing program that causes a computer to execute.
The estimation procedure is
If the visibility in the water when the target image is captured exceeds the first visibility, the brightness value of one color in the target area occupied by the target object in the target image, and the visibility reference information. estimating the underwater transparency when the target image was captured based on the comparison;
The information processing program according to claim 10.
The estimation procedure is
If the transparency of the water when the target image is captured is equal to or lower than the first transparency, brightness values of a plurality of colors in a target area occupied by the target object in the target image and the transparency reference information. estimating the underwater transparency when the target image was captured based on the comparison with the
The information processing program according to claim 10.
a reference image including an underwater reference object irradiated with light from a light source, a reference brightness value that is a brightness value in an object region occupied by the reference object among the reference images captured by the underwater imaging device; Reference information indicating a relationship with a reference distance that is a distance from the imaging device to the reference object, and a target image including the underwater object irradiated with light from the light source, an acquisition unit that acquires a target image captured by the imaging device;
Estimating a target distance, which is a distance from the imaging device to the target object, based on a comparison between a target brightness value, which is a brightness value in a target object area occupied by the target object, in the target image, and the reference information. an estimator to
An information processing device comprising:
A perspective reference image that includes an underwater reference object irradiated with light from a light source, which is a luminance value in an object area occupied by the reference object among the perspective reference images captured by the underwater imaging device. a target image including visibility reference information indicating a relationship between a degree reference brightness value and the underwater visibility, and an underwater object irradiated with light from the light source, the image capturing device underwater; an acquisition unit that acquires a target image captured by the
The target image is calculated based on a comparison between the brightness value of at least one color among the brightness values of a plurality of colors in the target object area occupied by the target object in the target image and the transparency reference information. an estimation unit that estimates underwater transparency when the image is captured;
An information processing device comprising:
An information processing method performed by a computer, the method comprising:
a reference image including an underwater reference object irradiated with light from a light source, a reference brightness value that is a brightness value in an object region occupied by the reference object among the reference images captured by the underwater imaging device; Reference information indicating a relationship with a reference distance that is a distance from the imaging device to the reference object, and a target image including the underwater object irradiated with light from the light source, an acquisition step of acquiring a target image captured by the imaging device;
Estimating a target distance, which is a distance from the imaging device to the target object, based on a comparison between a target brightness value, which is a brightness value in a target object area occupied by the target object, in the target image, and the reference information. an estimation process to
Information processing methods including.
A perspective reference image that includes an underwater reference object irradiated with light from a light source, which is a luminance value in an object area occupied by the reference object among the perspective reference images captured by the underwater imaging device. a target image including visibility reference information indicating a relationship between a degree reference brightness value and the underwater visibility, and an underwater object irradiated with light from the light source, the image capturing device underwater; an acquisition step of acquiring a target image captured by;
The target image is calculated based on a comparison between the brightness value of at least one color among the brightness values of a plurality of colors in the target object area occupied by the target object in the target image and the transparency reference information. an estimation step of estimating the underwater visibility when the image is taken;
Information processing methods including.