Image analysis device, method, and program

Abstract

The present invention provides an image analysis device, a method, and a program that automatically detect a surfer riding a wave from a video.　This image analysis device comprises: a video acquisition unit 31 that acquires a video of the surface of water in which a wave can be generated; and a detection unit 32 that detects a riding surfer who is riding a wave on the basis of the video which has been acquired by the video acquisition unit 31. The detection unit 32 detects, as the riding surfer, a subject that satisfies the following requirements r1 and r2.　Requirement r1: Movement speed changes, within a first prescribed time, from a first speed to a second speed that is faster than the first speed.　Requirement r2: The shape of the circumscribed frame of the subject at the time of the first speed is substantially a square or substantially a horizontally long rectangle, and the shape of the circumscribed frame of the subject at the time of the second speed is substantially a vertically long rectangle.

Description

Image analysis device, method, and program

[0001] The present invention relates to an image analysis device, method, and program.

[0002] There are many marine activities that utilize fracking, one of the most prominent being surfing. Surfing is greatly affected by wave conditions. Therefore, there is a need to provide data on wave conditions for surfers. Conventionally, waves have been recognized visually on-site and data on wave conditions has been collected. However, to reduce the burden on people, there is a need for an automatic determination of wave conditions from video footage.

[0003] Patent Document 1 discloses a technology for automatically measuring wave height using a buoy floating on the water surface. The technology disclosed in Patent Document 1 measures wave height by measuring the vertical displacement of a buoy that follows the water surface.

[0004] Japanese Patent Application Laid-Open No. 2018-4528

[0005] However, while Patent Document 1 can measure wave height, it does not disclose a method for measuring other wave data. Furthermore, wave conditions suitable for surfing are complexly related to various factors, so wave height data alone is not sufficient. Determining wave conditions based solely on wave height data can sometimes differ from the wave conditions experienced by actual surfers, which is a problem.

[0006] The present invention was made in consideration of such problems, and its purpose is to provide an image analysis device, method, and program for determining wave conditions that meet the needs of surfers.

[0007] Furthermore, in order to determine the wave conditions experienced by actual surfers, it is desirable to analyze the reactions of surfers who are actually riding the waves. Furthermore, to automatically analyze the reactions, it is desirable to detect surfers who are actually riding the waves from video and analyze their behavior. However, there is no technology yet that can detect only surfers riding waves from video.

[0008] The present invention has been made in consideration of such problems, and its purpose is to provide an image analysis device, method, and program that automatically detects surfers riding waves from video footage.

[0009] An image analysis device according to one aspect of the present invention includes an image acquisition unit that acquires video of water on which waves can be generated, and a detection unit that detects a ride surfer, a surfer riding a wave, based on the video acquired by the image acquisition unit, and the detection unit detects a subject that satisfies the following requirements r1 and r2 as a ride surfer: Requirement r1: The moving speed changes from a first speed to a second speed that is faster than the first speed within a first predetermined time; and Requirement r2: The shape of the circumscribing frame of the subject at the first speed is an approximately square or approximately horizontally elongated rectangle, and the shape of the circumscribing frame of the subject at the second speed is an approximately vertically elongated rectangle.

[0010] An image analysis method according to one aspect of the present invention is an image analysis method for detecting surfers riding waves, the method including: an image acquisition step of acquiring an image of a body of water where waves can be generated; and a detection step of detecting a ride surfer, a surfer riding a wave, based on the image acquired in the image acquisition step, wherein the detection step detects a subject that satisfies the following requirements r1 and r2 as a ride surfer: Requirement r1: The moving speed changes from a first speed to a second speed that is faster than the first speed within a first predetermined time; and Requirement r2: The shape of the circumscribing frame of the subject at the first speed is an approximately square or approximately horizontally elongated rectangle, and the shape of the circumscribing frame of the subject at the second speed is an approximately vertically elongated rectangle.

[0011] A program according to one aspect of the present invention is executable by a computer capable of exchanging information with an input / output device, and includes: an image acquisition step of acquiring imagery of a body of water on which waves can be generated; and a detection step of detecting a ride surfer, a surfer riding a wave, based on the image acquired in the image acquisition step, wherein the detection step detects, as a ride surfer, an object that satisfies the following requirements r1 and r2: Requirement r1: The moving speed changes from a first speed to a second speed that is faster than the first speed within a first predetermined time; and Requirement r2: The shape of the circumscribing frame of the object at the first speed is an approximately square or approximately horizontally elongated rectangle, and the shape of the circumscribing frame of the object at the second speed is an approximately vertically elongated rectangle.

[0012] According to the present invention, it is possible to detect a surfer riding a wave.

[0013] According to the present invention, the wave conditions can be determined.

[0014] Fig. 1 is a diagram showing the configuration of a wave judgment system 1 according to this embodiment. Fig. 2 is a diagram showing an example of detecting a surfer from a video. Fig. 3 is a diagram showing the posture of a surfer captured in the video. Fig. 4 is a diagram showing content created by a content provider 34. Fig. 5 is a flowchart showing the flow of wave judgment processing.

[0015] 1 is a diagram showing the configuration of a wave judgment system 1 according to this embodiment. The wave judgment system 1 includes a camera 11, a wave judgment device 12 as an image analysis device, and a user terminal.

[0016] The camera 11 is installed, for example, on a beach and captures images of a water scene where waves can be generated, including a surfer. The camera 11 supplies the captured images to the wave judgment device 12. Here, "on water" includes the ocean, rivers, lakes, swimming pools, etc.

[0017] A surfer is someone who surfs. A surfer either sits astride a surfboard and waits for a wave, or lies flat on the board and paddles with their hands as the wave moves forward (paddling), and then waits for the right timing to stand up on the board. The surfer then uses the power of the wave to move forward and ride the wave (ride). A surfer who is riding the wave will be referred to as a ride surfer below.

[0018] The wave judgment device 12 is an image analysis device such as a cloud server, a PC (personal computer), a smartphone, or a tablet. The wave judgment device 12 analyzes surfers within the image capture range based on the video captured by the camera 11 and determines the wave conditions. The wave judgment device 12 creates content based on the determination and provides it to the user device. The wave judgment device 12 may be an independent device or may be incorporated into the camera 11, etc.

[0019] The user terminal is, for example, a smartphone, and can display the wave judgment content received from the wave judgment device 12 via the network.

[0020] The video captured by the camera 11 is supplied to the wave judgment device 12 and analyzed. Note that the video captured by the camera 11 may be temporarily stored in an external server or PC and then analyzed by the wave judgment device 12. For example, video distributed by live streaming may be stored and then analyzed by the wave judgment device 12. The captured video does not necessarily have to be analyzed by the wave judgment device 12 at the same time.

[0021] (Functional Configuration of Wave Judgment Device 12) As shown in FIG. 1, the wave judgment device 12 includes a communication unit 21, a storage unit 22, and a control unit 23.

[0022] The communication unit 21 communicates with the camera 11 and the user terminal. The communication unit 21 may transmit various types of information using short-range wireless communication such as Bluetooth (registered trademark) or Wi-Fi (registered trademark). Furthermore, the communication unit 21 may exchange information with external servers and various services via a network (not shown) using wide-area wireless communication.

[0023] The storage unit 22 includes, for example, a read-only memory (ROM), a random access memory (RAM), and a non-volatile memory, and stores the control application programs described above, various data required for their execution, and information generated by processing.

[0024] The control unit 23 is composed of a CPU (Central Processing Unit), a memory section (e.g., ROM (Read Only Memory), RAM (Random Access Memory), non-volatile memory), and other elements including hardware. The control unit 23 executes a control application program (not shown) stored in the memory unit 22 to control the entire wave judgment device 12 and also functions as an image acquisition unit 31, a detection unit 32, a judgment unit 33, and a content provision unit 34.

[0025] (Video acquisition unit 31) The video acquisition unit 31 acquires video data captured by the camera 11 and stores it in the memory unit 22. The video captured by the camera 11 is captured as images that continue in time series for a predetermined period of time. Therefore, the video acquired by the video acquisition unit 31 is stored as chronologically continuous video images or as still images at predetermined time intervals. The video acquired by the video acquisition unit 31 undergoes wave judgment through real-time information processing in the detection unit 32 and judgment unit 33.

[0026] Furthermore, wave determination does not necessarily have to be performed through real-time information processing; the image can be temporarily stored in an external server or memory unit 22, and when performing wave determination processing, the image can be retrieved from the external server or memory unit 22 and the wave determination processing can be performed.

[0027] (Detection Unit 32 ) The detection unit 32 detects a ride surfer from the video captured by the video capture unit 31 .

[0028] Figure 2 shows an example of detecting multiple ride surfers from a single video. The video shown in Figure 2 shows surfers as subjects a to j. Of the subjects a to j, the detection unit 32 detects subjects a and b as ride surfers.

[0029] Specifically, the detection unit 32 detects a subject that satisfies both of the following requirements r1 and r2 as a Ride Surfer.

[0030] Requirement r1: The moving speed changes from a first speed to a second speed that is faster than the first speed within a first predetermined time. Requirement r2: The shape of the circumscribing frame of the subject at the first speed is an approximately square or an approximately horizontally elongated rectangle, and the shape of the circumscribing frame of the subject at the second speed is an approximately vertically elongated rectangle.

[0031] Requirement r1 will now be described. Requirement r1 is a requirement for detecting a surfer who has started riding after waiting for a wave or paddling as a riding surfer.

[0032] The surfer waits for a wave or paddles for a first predetermined period of time as described above, and then begins his ride.

[0033] Although it varies depending on the size and conditions of the waves, the movement speed of a surfer while waiting for waves is, for example, approximately 0 km / h, and the movement speed of a surfer while paddling is, for example, less than approximately 5 km / h. On the other hand, the movement speed of a surfer while riding is, for example, approximately 10 km / h or more, with the surfer moving faster while riding. Therefore, the first speed corresponds to approximately the movement speed while waiting for waves or paddling, and the second speed corresponds to approximately the movement speed while riding, with the second speed being faster than the first speed. The first predetermined time can be set as appropriate.

[0034] From the above, it can be said that a subject whose movement speed changes from a first speed to a second speed that is faster than the first speed within a first predetermined time is highly likely to be a ride surfer.

[0035] For example, to calculate the movement speed of the subject, the detection unit 32 may identify the position of the subject using three-dimensional coordinates, measure the number of pixels moved by the subject, and calculate the number of pixels moved by the subject per second (pixels per second) based on the number of pixels, the video resolution, and the frame rate.

[0036] The detection unit 32 can use a known method to identify the three-dimensional coordinates of the subject. For example, the detection unit 32 can use a plurality of calibrated cameras 11 to detect feature points and identify the three-dimensional coordinates of the subject through triangulation. Alternatively, the detection unit 32 can obtain the positions of ocean waves affected by tides in pixel coordinates based on forecast data, and use the coordinates of the wave positions to obtain pseudo-depth information of the subject.

[0037] Requirement r2 will now be explained. Requirement r2 is also a requirement for detecting as a ride surfer a surfer who has started riding after waiting for a wave or paddling. If requirement r1 alone was met, there was a possibility that other marine sports players would also be detected, so requirement r2 was added to accurately detect ride surfers.

[0038] 3A and 3B are diagrams showing the posture of a surfer captured in a video. As shown in Fig. 3A, the surfer is lying down on his surfboard while paddling. Therefore, the shape of the circumscribing box O of the surfer captured in the video is a substantially horizontally elongated rectangle. Also, the surfer is sitting astride the surfboard while waiting for a wave. Therefore, the shape of the circumscribing box of the surfer captured in the video is a substantially square.

[0039] 3B, when the surfer starts riding, he stands up on the surfboard, so the shape of the circumscribing box P of the surfer in the video is a substantially vertically long rectangle.

[0040] From the above, it can be said that a subject whose circumscribing frame is approximately square or approximately horizontally elongated rectangle when traveling at a first speed corresponding to waiting for waves or paddling, and whose circumscribing frame is approximately vertically elongated rectangle when traveling at a second speed corresponding to riding, is likely to be a surfer.

[0041] If requirements r1 and r2 are met, it is conceivable that in addition to ride surfers, other objects may also be detected, such as windsurfers, jet skiers, fishing boats, etc. Therefore, settings may be made to exclude objects other than ride surfers from detection targets based on their movement speed, area relative to the angle of view, and position within the video.

[0042] The detection unit 32 may detect the ride surfer by machine learning elements such as the shape of the circumscribing frame of the ride surfer and the area relative to the angle of view using AI (Artificial Intelligence).

[0043] ((Determination Unit 33)) The determination unit 33 determines the wave conditions based on the number of times that the detection unit 32 detects a ride surfer within a second predetermined time period.

[0044] (((Determining Wave Conditions Based on the Number of Rides))) First, the determination unit 33 determines the number of times that the detection unit 32 detects a ride surfer in a second predetermined time period (for example, 10 minutes or 1 hour). The number of times that the detection unit 32 detects a ride surfer in the second predetermined time period is the number of times that a surfer was able to ride. The number of times that the detection unit 32 detects a ride surfer in the second predetermined time period is hereinafter referred to as the number of rides.

[0045] The determination unit 33 determines that the more rides there are in the second predetermined time period, the more suitable the wave conditions are for surfing. For example, the determination unit 33 determines the wave conditions by calculating a ride index to quantify the determination result. For example, the ride index is a score out of 100. For example, if the number of rides is 150 or more in 10 minutes, the determination unit 33 determines that the wave is one that can be ridden many times and is suitable for surfing, and assigns a ride index of 70 points. On the other hand, if the number of rides is 50, for example, the wave is one that cannot be ridden many times and is not very suitable for surfing, and assigns a ride index of 20 points.

[0046] (((Corrected Number of Rides))) It is preferable that the number of rides used by the determination unit 33 for determination is a number of rides corrected based on the total number of surfers. Hereinafter, the corrected number of rides will be referred to as the corrected number of rides. Hereinafter, when there is no need to distinguish between the number of rides and the corrected number of rides, they will be referred to uniformly as the number of rides.

[0047] The total number of surfers captured in the video during the second predetermined time period significantly affects the number of rides. Specifically, the number of rides tends to increase as the number of surfers increases, and decreases as the number of surfers decreases. However, once the number of surfers exceeds a predetermined number, there is competition for the breaking waves needed for riding, and the tendency for the number of rides to increase as the number of surfers increases weakens.

[0048] The determination unit 33 determines the wave conditions using the number of rides. In this case, it is desirable to eliminate the influence of the number of surfers from the number of rides. Therefore, the determination unit 33 may correct the number of rides based on the number of surfers.

[0049] The process of correcting the number of rides will now be described.

[0050] First, the determination unit 33 determines the total number of surfers, including surfers not riding, based on the number of times ride surfers are detected during the second time period. For example, the detection unit 32 labels each detected ride surfer and tracks them during the second time period. The determination unit 33 then determines the number of labels assigned by the detection unit 32 and determines that the number of labels is the number of surfers.

[0051] The determination unit 33 sets a number-of-surfers ratio coefficient based on the number of surfers in the second predetermined time. The number-of-surfers ratio coefficient is a coefficient used to correct the number of rides. For example, if there are fewer than 10 ride surfers in the second predetermined time, the number-of-surfers ratio coefficient is set to a value greater than 1. On the other hand, if there are 20 or more ride surfers, the number-of-surfers ratio coefficient is set to a value less than 1, and the value gradually decreases as the number of surfers increases.

[0052] As mentioned above, once the number of surfers exceeds a certain number, the tendency for the number of rides to increase as the number of surfers increases weakens, so a lower limit may be set for the number-of-surfers ratio coefficient. For example, if there are 100 or more surfers, the number-of-surfers ratio coefficient is set to 0.6, and if there are 120 surfers, the number-of-surfers ratio coefficient remains at 0.6.

[0053] The determination unit 33 then corrects the number of rides by multiplying the determined number of rides by the number-of-surfers comparison coefficient. For example, if the number of surfers in the second predetermined time is 30 and the number of rides is 100, and the coefficient for the number of surfers entering the water is 0.8, the corrected number of rides will be 80.

[0054] This eliminates the effect that the number of surfers has on the number of rides, which is the case when the number of surfers is greater, and makes the number of rides a more fair value.

[0055] In the above description, the determination unit 33 determines the total number of surfers based on the number of times that ride surfers were detected during the second time period. However, the number of surfers may be determined using other methods. For example, the detection unit 32 may detect surfers by recognizing multiple parts, such as a human arm, a human head, or a surfboard, or the combined movement of these parts. The determination unit 33 may then determine the number of surfers detected by the detection unit 32 during the second time period.

[0056] (Content providing unit 34) The content providing unit 34 creates content to be provided to surfers based on the determination made by the determination unit 33. FIG. 4 is a diagram showing content created by the content providing unit 34. For example, as shown in box Q, the content providing unit 34 displays the ride index set by the determination unit 33.

[0057] The content providing unit 34 transmits the created content to the user terminal via the communication unit 21 and displays it on a display unit (not shown) of the user terminal.

[0058] This allows surfers to view the content and get an idea of ​​the wave conditions.

[0059] For the sake of simplicity, the ride index is displayed as is in the content. However, the ride index may be converted based on other evaluation criteria and the converted content may be displayed. The content provider 34 may also consider the ride index and factors other than the ride index to assign a new score and display the score in the content. The content provider 34 may also display comments to be notified to the user based on the ride index in the content. The number of surfers at the second predetermined time determined by the determination unit 33, congestion information, wind direction information using a weather forecast, and illustrations may also be added.

[0060] (Wave Determination Process) Next, a description will be given of the wave determination process performed by the wave determination device 12. First, the flow of the process will be described with reference to the flowchart of FIG.

[0061] The wave judgment process is started, for example, by operating a start button (not shown) of the wave judgment device 12 .

[0062] In step S1 , the video acquisition unit 31 acquires the video captured by the camera 11 from the storage unit 22 .

[0063] In step S2, the detection unit 32 detects the ride surfer from the video acquired by the video acquisition unit 31.

[0064] In step S3, the determination unit 33 determines the number of times the detection unit 32 has detected the Ride Surfer (number of rides) within a second predetermined time period.

[0065] In step S4, the determination unit 33 calculates a ride index as a wave condition based on the number of rides.

[0066] In step S5, the content providing unit 34 creates content based on the ride index calculated by the determination unit 33, and provides the content to the user terminal 13 via the communication unit 21. If the content has already been provided, the content providing unit 34 updates the content.

[0067] The process then ends.

[0068] [Other Examples]

[0069] (Wave Condition Determination Based on Ride Time) In the above, the determination unit 33 determines the number of rides, but it may also measure the time from the start to the end of the ride of the surfer (hereinafter referred to as ride time).

[0070] The start of the ride is, for example, when the detected movement speed of the Ride Surfer reaches a second speed, and the end of the ride is when the detected movement speed of the Ride Surfer slows down to a first speed.

[0071] The determination unit 33 may measure the ride time for each ride surfer and calculate the average value for a predetermined time period. For example, if the predetermined time period is 10 minutes, the determination unit 33 calculates the average ride time for the determined ride surfers over the 10-minute period.

[0072] In the above description, the determination unit 33 determines the wave condition based on the number of rides, but it may also determine the wave condition based on the ride time. Specifically, the determination unit 33 determines that the longer the ride time, the more suitable the wave condition for surfing. For example, if the ride time is 10 seconds or more, the determination unit 33 determines that the wave has a long ride time, is easy to ride, and is suitable for surfing, and assigns a ride index of 70 points. On the other hand, if the ride time is less than 3 seconds, the determination unit 33 determines that the ride time is short, the wave is difficult to ride, and is not very suitable for surfing, and assigns a ride index of 20 points.

[0073] The determination unit 33 determines the wave conditions by taking into consideration both the ride index calculated based on the number of rides and the ride index calculated based on the ride time. For example, the determination unit 33 averages the two ride indices. Furthermore, the content providing unit 34 displays the average value of the ride indices calculated by the determination unit 33.

[0074] The determination unit 33 may also add up all the riding times of each ride surfer over a given period of time and determine the wave conditions based on this total.

[0075] (Another Example of Determining Wave Conditions Based on Ride Time) The determination unit 33 may not measure the ride time in a predetermined case even if a ride surfer is detected. Alternatively, the determination unit 33 may not take the ride time of a predetermined ride surfer into account when calculating the average or total value of the ride time.

[0076] For example, the detection unit 32 detects the part of the wave that the ride surfer is riding using known technology, and the determination unit 33 does not measure the ride time of the ride surfer if the part of the wave that the ride surfer is riding is a predetermined part.

[0077] A wave is made up of a face, which is the slope of the wave, and a soup, which is the white part of the broken wave. A wave suitable for surfing is the face before the wave breaks, while the soup is not suitable for surfing. Therefore, the determination unit 33 may not measure the ride time of a ride surfer if the wave being ridden by the ride surfer is in the soup. This allows the determination unit 33 to determine wave conditions by excluding rides in the soup, which is not suitable for surfing, thereby improving the accuracy of the determination.

[0078] (Movement direction) In the above, the determination unit 33 determines the number of rides, but it may also determine the movement direction of the ride surfer. The determination unit 33 determines whether the ride surfer shown in the video moves to the right or left over time. The determination unit 33 determines the movement direction over a predetermined period of time and identifies the direction that is determined to be the most frequently.

[0079] When creating the content, the content provider 34 considers the identified moving direction of the ride surfer to be the moving direction of the wave, and adds wave moving direction information to the content.

[0080] There are right-handed and left-handed surfers, and the ease of riding a wave varies depending on the direction of the wave. That is, when the wave is in a direction suitable for right-handed surfers, many ride surfers move in the corresponding direction. On the other hand, when the wave is in a direction suitable for left-handed surfers, many ride surfers move in the corresponding direction. Therefore, the determination unit 33 determines the condition of the waves on the water based on the direction of movement of the ride surfer, and can determine whether the wave is suitable for right-handed surfers or left-handed surfers.

[0081] (Wave Condition Determination Based on the Ride Surfer's Movement) In the above, the determination unit 33 determines the wave condition based on the number of rides, but the wave condition may also be determined based on the ride surfer's movement.

[0082] For example, the detection unit 32 detects the ride surfer and his / her movements using known image analysis technology, and the determination unit 33 determines that the wave is suitable for surfing if the ride surfer is performing a predetermined movement. The predetermined movement may be, for example, the ride surfer suddenly changing direction of travel or sending up large waves. For example, if the ride surfer performs a predetermined movement, the determination unit 33 determines that the wave is easy to ride and suitable for surfing, and adds points to the ride index.

[0083] (Utilization of Information) The content providing unit described above provides information about wave conditions primarily to surfers, but information about wave conditions may also be provided to non-surfers as well as surfers.

[0084] [Additional Notes] The contents of the above-described embodiments can be understood, for example, as follows.

[0085] The image analysis device described above comprises: an image acquisition unit 31 that acquires video of water surfaces where waves can be generated; and a detection unit 32 that detects ride surfers, or surfers riding waves, based on the video acquired by the image acquisition unit 31, wherein the detection unit 32 detects as ride surfers a subject that satisfies the following requirements r1 and r2: Requirement r1: The moving speed changes from a first speed to a second speed that is faster than the first speed within a first predetermined time; and Requirement r2: The shape of the circumscribing frame of the subject at the first speed is an approximately square or approximately horizontally elongated rectangle, and the shape of the circumscribing frame of the subject at the second speed is an approximately vertically elongated rectangle.

[0086] With this configuration, the image analysis device can detect surfers riding based on changes in the speed of movement and the shape of the subject, thereby detecting only surfers riding and not surfers not riding.

[0087] The device further includes a determination unit 33 that determines the wave conditions based on the number of times the detection unit 32 detects a ride surfer within a second predetermined time period.

[0088] The number of times that the detection unit 32 detects a surfer riding a wave during the second predetermined time period corresponds to the number of times that the surfer was able to ride a wave. Typically, if the wave is suitable for surfing, the number of times that the surfer was able to ride a wave is large, and if the wave is not suitable for surfing, the number of times that the surfer was able to ride a wave is small. Therefore, the image analysis device determines the wave conditions based on the number of times that the surfer was able to ride a wave, and can determine the wave conditions based on a more realistic situation.

[0089] The determination unit 33 also determines the condition of the waves on the water based on the riding time of the ride surfer.

[0090] Typically, if the waves are suitable for surfing, the surfer will ride for a long time, and if the waves are not suitable for surfing, the surfer will ride for a short time. Therefore, the image analysis device is designed to determine wave conditions based on the surfer's riding time, so it can determine wave conditions more accurately and in line with reality.

[0091] The determination unit 33 also determines the condition of the waves on the water based on the direction of movement of the ride surfer.

[0092] There are right-handed and left-handed surfers, and the ease of riding a wave varies depending on the direction of the wave. That is, if the wave is oriented in a direction suitable for right-handed surfers, many ride surfers will move in the corresponding direction. On the other hand, if the wave is oriented in a direction suitable for left-handed surfers, many ride surfers will move in the corresponding direction. Therefore, the image analysis device determines the condition of the waves on the water based on the direction of the ride surfer's movement, so it can determine whether the wave is suitable for right-handed surfers or left-handed surfers.

[0093] The detection unit 32 further detects surfers based on the video acquired by the video acquisition unit 31, and the determination unit 33 determines the wave conditions based on the number of surfers detected by the detection unit 32 in a second predetermined time period and the number of times the detection unit 32 detects riding surfers.

[0094] The total number of surfers greatly affects the number of ride surfers detected by the detection unit 32. Specifically, the more surfers there are, the more ride surfers the detection unit 32 detects. On the other hand, the fewer surfers there are, the fewer ride surfers the detection unit 32 detects.

[0095] With the above-described configuration, for example, when the total number of surfers is large, the determination unit 33 can correct the number of detected ride surfers to be lower, and determine the wave conditions based on the corrected number of detected ride surfers. This eliminates the influence of the number of surfers on the number of detected ride surfers, which is the case when the number of surfers is larger, and allows the determination unit 33 to determine the wave conditions using a more fair value. [Supplementary explanation of the embodiment]

[0096] The above-described embodiments each show a preferred specific example of the present invention. The numerical values, components, arrangement positions and connection order of components, processing order in flowcharts, etc. shown in the following embodiments are merely examples and are not intended to limit the present invention. Furthermore, each figure is not necessarily a strict illustration.

[0097] The above-described series of processes can be executed by hardware or software. When the series of processes is executed by software, the program constituting the software is installed from a program recording medium into a computer incorporated in dedicated hardware, or into, for example, a general-purpose personal computer that can execute various functions by installing various programs.

[0098] The program executed by the computer may be a program that processes in chronological order according to the order described in this specification, or may be a program that processes in parallel or at the required timing, such as when called.

[0099] REFERENCE SIGNS LIST 1 Wave judgment system 11 Camera 12 Wave judgment device 13 User terminal 21 Communication unit 22 Storage unit 23 Control unit 31 Video acquisition unit 32 Detection unit 33 Judgment unit 34 Content provision unit

Claims

1. An image analysis device comprising: a video acquisition unit that acquires video of water surfaces where waves can be generated; and a detection unit that detects ride surfers, or surfers riding the waves, based on the video acquired by the video acquisition unit, wherein the detection unit detects as the ride surfer a subject that meets the following requirements r1 and r2: Requirement r1: The moving speed changes from a first speed to a second speed that is faster than the first speed within a first predetermined time. Requirement r2: The shape of the circumscribing frame of the subject at the first speed is an approximately square or an approximately horizontally elongated rectangle, and the shape of the circumscribing frame of the subject at the second speed is an approximately vertically elongated rectangle.

2. An image analysis device according to claim 1, further comprising a determination unit that determines the wave conditions based on the number of times the detection unit detects the ride surfer within a second predetermined period of time.

3. An image analysis device according to claim 2, wherein the determining unit determines the wave conditions based on the ride time of the ride surfer.

4. An image analysis device according to claim 2 or 3, characterized in that the determination unit further determines the wave condition based on the direction of movement of the ride surfer.

5. An image analysis device as described in claim 2, wherein the detection unit further detects surfers based on the video acquired by the video acquisition unit, and the judgment unit judges the wave conditions based on the number of surfers detected by the detection unit in the second predetermined time period and the number of times the detection unit detects riding surfers in the second predetermined time period.

6. An image analysis method for detecting surfers riding waves, comprising: an image acquisition step of acquiring video of the water surface where waves can be generated; and a detection step of detecting a ride surfer, a surfer riding the wave, based on the video acquired in the image acquisition step, wherein the detection step detects a subject that meets the following requirements r1 and r2 as the ride surfer. Requirement r1: The moving speed changes from a first speed to a second speed that is faster than the first speed within a first predetermined time. Requirement r2: The shape of the circumscribing frame of the subject at the first speed is an approximately square or an approximately horizontally elongated rectangle, and the shape of the circumscribing frame of the subject at the second speed is an approximately vertically elongated rectangle.

7. A program executable by a computer capable of exchanging information with an input / output device, comprising: an image acquisition step of acquiring an image of a body of water where waves can be generated; and a detection step of detecting a ride surfer, a surfer riding the wave, based on the image acquired in the image acquisition step, wherein the detection step detects a subject that meets the following requirements r1 and r2 as the ride surfer. Requirement r1: The moving speed changes from a first speed to a second speed that is faster than the first speed within a first predetermined time. Requirement r2: The shape of the circumscribing frame of the subject at the first speed is an approximately square or an approximately horizontally elongated rectangle, and the shape of the circumscribing frame of the subject at the second speed is an approximately vertically elongated rectangle.

Images