Method and device for ball spin calculation using difference images
The method and apparatus use difference images and neural networks to accurately calculate ball spin without prior markings, addressing the limitations of existing technologies in cost and pattern dependency.
Patent Information
- Application Number
- KR1020240078850
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2024-06-18
- Publication Date
- 2026-11-11
- Estimated Expiration
- 2044-06-18
AI Technical Summary
Existing methods for measuring ball spin, such as using specialized sensors or high-speed cameras, are costly, limited to balls with predetermined patterns, or require complex calculations, and fail to accurately measure spin without prior markings.
A method and apparatus using difference images of a moving ball, processed by an artificial neural network, to calculate spin direction and speed without requiring separate sensors or prior markings, by subtracting consecutive ball images and normalizing them for size and brightness.
Accurately measures ball spin without prior markings, using difference images and neural networks, providing precise spin direction and speed without additional equipment costs.
Smart Images

Figure 112024065605967-PAT00006_ABST
Abstract
Description
Technology Field
[0001] The present disclosure relates to a method and apparatus for calculating ball spin using a difference image, and more specifically, to a method and apparatus for calculating the spin of a ball using a difference image of a moving ball. Background Technology
[0003] In sports fields that utilize a ball, such as baseball, golf, and basketball, the physical characteristics of the moving ball are accurately measured, and the measured values are used to calculate the ball's trajectory or visualize it as an image, which is then utilized in simulation devices. In particular, in the field of screen golf, a method is used in which the physical characteristics of a golf ball struck by a user are sensed, the trajectory of the ball is calculated using these characteristics, and the result is displayed on a screen.
[0004] To accurately calculate the trajectory of a ball, its physical characteristics must be precisely measured; this requires considering ball spin along with ball speed, acceleration, launch angle, and direction angle. In particular, in the field of golf, ball spin is a crucial factor in determining the ball's movement, such as its ball flight characteristics and trajectory, and therefore must be measured accurately. However, as ball spin is an inherent motion of the moving ball, it is difficult to measure using simple sensors or cameras compared to other physical characteristics, and accurate measurement requires expensive equipment or complex calculations.
[0005] Existing methods for measuring ball spin include predicting it from other physical characteristics and information of the ball, using special sensors such as Doppler radar sensors, and analyzing multiple images acquired using high-speed cameras (or ultra-high-speed cameras). For example, in the field of golf simulation, methods have been used to predict the spin of a golf ball by utilizing basic information such as ball speed, launch angle, and direction angle, as well as information about the golf club (such as the club path). As another example, the existing method of measuring ball spin using a high-speed camera involves marking a predetermined pattern or mark on the ball and analyzing changes in the marked pattern or mark from images obtained from the high-speed camera.
[0006] However, existing methods that predict ball spin based on other physical characteristics and information suffer from significant discrepancies with reality. Furthermore, methods using specialized sensors are problematic due to the requirement of additional, expensive equipment, while conventional methods utilizing high-speed cameras are limited to accurate measurements only when using balls that contain predetermined patterns or marks. Additionally, when considering cases where the ball's marks or patterns are not predetermined, a separate process is required to extract these marks or patterns beforehand to measure spin.
[0007] Ultimately, there has been a demand for a ball spin measurement method and device that can be used without using a separate sensor, even when there is no predetermined pattern or mark on the ball, and does not separately extract the ball's mark or pattern; however, there was a problem in that such a method could not be provided according to conventional technology, and the present disclosure is intended to solve this problem. The problem to be solved
[0009] The present disclosure is intended to accurately measure the spin of a ball.
[0010] Another objective of the present disclosure is to measure the spin of a ball even when there is no predetermined pattern or mark on the ball.
[0011] Another objective of the present disclosure is to measure the accurate spin of a ball without separately extracting the ball's pattern or mark.
[0012] Another objective of the present disclosure is to calculate the spin of a ball using difference images of consecutive ball images.
[0013] Another objective of the present disclosure is to calculate ball spin using an artificial neural network.
[0015] The purposes of the present disclosure are not limited to those mentioned above, and other purposes and advantages of the present disclosure not mentioned may be understood from the following description and will be more clearly understood by the embodiments of the present disclosure. Furthermore, it will be readily apparent that the purposes and advantages of the present disclosure can be realized by the means and combinations thereof set forth in the claims. means of solving the problem
[0017] One embodiment of the present disclosure aims to provide a method and apparatus for calculating ball spin using a difference image.
[0018] One embodiment of the present disclosure aims to provide a method for calculating ball spin executed on a computing device. The method may include the steps of: acquiring consecutive images of a moving ball; extracting a portion of the ball from each of the consecutive images as a target image; generating a difference image by differentiating the target images; and calculating the spin direction and spin speed of the moving ball based on the difference image.
[0019] In one embodiment, the difference image can be obtained by subtracting two consecutive target images.
[0020] In one embodiment, the difference image can be obtained by subtracting the value corresponding to each pixel of one target image from the value corresponding to each pixel of another target image.
[0021] In one embodiment, the spin direction of the moving ball is the rotation axis vector of the ball, and the spin speed may be the number of revolutions per minute of the ball.
[0022] In one embodiment, the step of calculating the spin direction and spin speed of the moving ball based on the difference image may be performed by a machine-learned artificial neural network. The difference image may include two or more difference images.
[0023] In one embodiment, the step of acquiring continuous images of the moving ball may be performed by two or more camera devices.
[0024] In one embodiment, the center coordinates of the ball image of the target image may be located at the center.
[0025] In one embodiment, the target image can be normalized so that the size of the ball image is the same.
[0026] In one embodiment, the target image can be normalized so that the brightness of the ball image is uniform.
[0027] One embodiment of the present disclosure aims to provide a ball spin calculation device. The device may include: a sensor unit for acquiring continuous images of a moving ball; a target image extraction unit for extracting a portion of the ball from each of the continuous images as a target image; an image difference unit for generating a difference image by differentiating the target images; and a spin calculation unit for calculating the spin direction and spin speed of the moving ball based on the difference image.
[0028] In one embodiment, the image difference unit can obtain the difference image by subtracting two consecutive target images.
[0029] In one embodiment, the image difference unit can obtain the difference image by subtracting the value corresponding to each pixel of one target image from the value corresponding to each pixel of another target image.
[0030] In one embodiment, the spin direction of the moving ball is the rotation axis vector of the ball, and the spin speed may be the number of revolutions per minute of the ball.
[0031] In one embodiment, the spin output unit includes a machine-learned artificial neural network, and the artificial neural network can receive the difference image as input and output the spin direction and spin speed of the moving ball.
[0032] In one embodiment, the difference image may include two or more difference images.
[0033] In one embodiment, the sensor unit may include two or more camera devices.
[0034] In one embodiment, the target image extraction unit can extract the target image such that the center coordinates of the ball image are located at the center of the target image.
[0035] In one embodiment, an image preprocessing unit is further included, and the image preprocessing unit can normalize the size of the ball image of the target image so that it is the same.
[0036] In one embodiment, an image preprocessing unit is further included, and the image preprocessing unit can normalize the brightness of the ball image of the target image so that it is uniform.
[0037] One embodiment of the present disclosure includes a program stored on a recording medium to execute a method according to one embodiment of the present disclosure on a computer.
[0038] One embodiment of the present disclosure includes a computer-readable recording medium having a program for executing a method according to one embodiment of the present disclosure on a computer.
[0039] One embodiment of the present disclosure includes a computer-readable recording medium that records a database used in one embodiment of the present disclosure. Effects of the invention
[0041] The method and apparatus according to the present disclosure have the effect of accurately measuring the spin of a ball.
[0042] In addition, according to the present disclosure, there is an effect of being able to measure the spin of a ball even when there is no predetermined pattern or mark on the ball.
[0043] In addition, according to the present disclosure, there is an effect of being able to accurately measure the spin of a ball without separately extracting the ball's pattern or mark.
[0044] In addition, according to the present disclosure, there is an effect of calculating the accurate spin of a ball by subtracting consecutive ball images.
[0045] In addition, according to the present disclosure, there is an effect of being able to calculate ball spin using a difference image.
[0046] In addition, according to the present disclosure, there is an effect of being able to calculate ball spin using an artificial neural network.
[0048] In addition to the above, the specific effects of the present disclosure are described together with the specific details for implementing the disclosure below. Brief explanation of the drawing
[0050] FIG. 1 is a conceptual diagram illustrating the configuration of a ball spin calculation device according to one embodiment of the present disclosure. Figure 2 is an example diagram illustrating the spin direction (axis of rotation direction) and spin speed of a ball. FIG. 3 is a conceptual diagram illustrating the configuration of a sensor unit according to one embodiment of the present disclosure. FIG. 4 is a conceptual diagram illustrating the configuration of an image processing unit according to one embodiment of the present disclosure. Figures 5(a) and (b) are exemplary diagrams illustrating the process of obtaining a difference image from a target image. FIG. 6 is a conceptual diagram illustrating an embodiment of a spin calculation unit that calculates the spin of a ball from a difference image. FIG. 7 is a block diagram of a ball spin calculation device according to one embodiment of the present disclosure. Specific details for implementing the invention
[0051] To clarify the technical concept of the present disclosure, embodiments of the present disclosure will be described in detail with reference to the attached drawings. In describing the present disclosure, detailed descriptions of related known functions or components will be omitted if it is determined that such detailed descriptions would unnecessarily obscure the essence of the present disclosure. Components having substantially the same functional configuration among the drawings have been assigned the same reference numerals and symbols as much as possible, even if they are shown in different drawings. For convenience of explanation, devices and methods will be described together where necessary. Each operation of the present disclosure does not necessarily have to be performed in the order described and may be performed in parallel, selectively, or individually.
[0052] The terms used in the embodiments of this disclosure have been selected to be as widely used and general as possible, taking into account the functions of this disclosure; however, these terms may vary depending on the intent of those skilled in the art, case law, the emergence of new technologies, etc. Additionally, in specific cases, terms have been selected at the applicant's discretion, and in such cases, their meanings will be described in detail in the description of the relevant embodiments. Therefore, terms used in this specification should be defined not merely by their names, but based on their meanings and the overall content of this disclosure.
[0053] Throughout this disclosure, singular expressions may include plural expressions unless the context clearly indicates otherwise. Terms such as "comprising" or "having" are intended to specify the existence of features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. That is, throughout this disclosure, when a part is described as "comprising" a certain component, it means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.
[0054] Expressions such as "at least one" modify the entire list of components and do not modify the components of the list individually. For example, "at least one of A, B, and C" and "at least one of A, B, or C" refer to only A, only B, only C, both A and B, both B and C, both A and C, all of A, B, and C, or any combination thereof.
[0055] Additionally, terms such as "...part," "...module," etc., as described in this disclosure refer to a unit that processes at least one function or operation, and may be implemented in hardware or software, or as a combination of hardware and software.
[0056] Throughout the entire disclosure, when a part is described as being "connected" to another part, this includes not only cases where they are "directly connected," but also cases where they are "electrically connected" with other elements interposed between them. Furthermore, when a part is described as "including" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.
[0057] Throughout this disclosure, the expression “configured to” may be replaced, depending on the context, with, for example, “suitable for,” “having the capacity to,” “designed to,” “adapted to,” “made to,” or “capable of.” The term “configured to” may not necessarily mean only “specifically designed to” in hardware. Instead, in some situations, the expression “system configured to” may mean that the system is “capable of” in conjunction with other devices or components. For example, the phrase “processor configured to perform A, B, and C” may mean a dedicated processor for performing the said operations (e.g., an embedded processor) or a generic-purpose processor (e.g., a CPU or an application processor) capable of performing said operations by executing one or more software programs stored in memory.
[0058] Throughout this disclosure, terms including ordinal numbers, such as “first,” “second,” etc., may be used to describe various components, but the components are not limited by the terms. The terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of this disclosure, the first component may be named the second component, and similarly, the second component may be named the first component. The term “and / or” includes a combination of multiple related items or any one of the multiple related items.
[0060] The present disclosure relates to a method and apparatus for calculating ball spin using a difference image, and more specifically, to a method and apparatus for calculating the spin of a ball using a difference image of a moving ball.
[0062] FIG. 1 is a conceptual diagram illustrating the configuration of a ball spin calculation device according to one embodiment of the present disclosure.
[0063] A ball spin calculation device according to one embodiment of the present disclosure may include a sensor unit (100), an image processing unit (200), and a spin calculation unit (300). In one embodiment, the sensor unit (100), the image processing unit (200), and the spin calculation unit (300) may be implemented as a single device (not shown). In one embodiment, the image processing unit (200) and the spin calculation unit (300) may be implemented as a single computing device (10). In another embodiment, the image processing unit (200) and the spin calculation unit (300) may be implemented by including a plurality of computing devices.
[0064] The sensor unit (100) is a means for acquiring an image of a moving ball. In one embodiment, the sensor unit (100) is a high-speed camera device capable of acquiring an image of a moving ball and may include one or more cameras. For example, the sensor unit (100) may be configured as a stereo camera that operates in a stereo manner by including two or more cameras. As another example, the sensor unit (100) may be configured as a 3D camera capable of recognizing a two-dimensional image of a ball as three-dimensional coordinates. In this disclosure, for convenience of explanation, the sensor unit (100) is described as a camera device, but it is not limited thereto, and any technical means capable of acquiring an image of a ball may be used.
[0065] The image processing unit (200) is a means for processing ball images acquired from the sensor unit (100) so as to calculate the spin of the ball. In one embodiment, the image processing unit (200) may input at least one of the difference images of consecutive ball images to the spin calculation unit (300). For example, the image processing unit (200) may acquire difference images of consecutive images of a moving ball captured by the sensor unit (100) and provide at least one of these difference images to the spin calculation unit (300).
[0066] The spin calculation unit (300) can receive at least one difference image of a moving ball from the image processing unit (200) and calculate the spin direction and spin speed of the ball. In one embodiment, the spin calculation unit (300) can receive a difference image of a moving ball and output the spin axis and spin rate of the ball. Referring to FIG. 2, the spin calculation unit (300) can calculate the spin speed of the ball as a scalar value (u) representing the number of rotations per unit time (e.g., RPM), along with a vector value (v) representing the direction of the rotation axis of the ball spin.
[0067] In one embodiment, the spin calculation unit (300) is composed of a neural network that machine-learns the spin direction and spin speed of a ball according to a difference image, and when at least one difference image is input to the neural network, the spin direction and spin speed of a ball corresponding to the input image can be output. For example, the spin calculation unit (300) is implemented based on AI deep learning technology and can receive a difference image and output (v, u). Here, v is a vector value representing the direction of the rotation axis of the ball spin, and u is a scalar value (u) representing the number of rotations per unit time (e.g., RPM) of the ball spin speed.
[0069] FIG. 3 is a conceptual diagram illustrating the configuration of a sensor unit according to one embodiment of the present disclosure.
[0070] According to one embodiment of the present disclosure, the sensor unit (100) includes one or more cameras (110) to acquire a plurality of images of a moving ball. In one embodiment, using a high-speed camera (110), images of the moving ball at each shooting point can be acquired while the ball moves within the angle of view of the high-speed camera (110). For example, if the high-speed camera (110) operates at 3800 FPS (frames per second), one image is stored every 1 / 3800 second, so a plurality of consecutive images of the moving ball can be acquired while the ball moves within the angle of view of the camera.
[0071] Referring to FIG. 3, the sensor unit (100) can be implemented as a three-dimensional camera that operates in a stereo manner, including two cameras (110), and can measure the direction and speed of the ball by calculating three-dimensional coordinates for a moving ball.
[0073] FIG. 4 is a conceptual diagram illustrating the configuration of an image processing unit according to one embodiment of the present disclosure.
[0074] According to one embodiment of the present disclosure, the image processing unit (200) may include a target image extraction unit (210) and an image difference unit (230).
[0075] In one embodiment, the target image extraction unit (210) is a means for extracting a target image, which is an image of a ball, from each of the plurality of original images acquired from the sensor unit (100). For example, the target image extraction unit (210) can recognize an image of a ball from the original images and extract a target image by taking a certain area from the center coordinates of the ball as a region of interest. In one embodiment, the target image extraction unit (210) can generate a target image such that the center coordinates of the ball correspond to the center of the target image.
[0076] In one embodiment, the image difference unit (230) can generate a difference image by differentiating consecutive target images extracted from the target image extraction unit (210). In one embodiment, the image difference unit (230) can generate a difference image by calculating the difference between two consecutive target images of a moving ball. For example, the image difference unit (230) can obtain the difference between two consecutive target images as a difference image. Here, the difference image of the two target images can be obtained by calculating the difference between digital numeric values corresponding to the pixels constituting the two target images.
[0077] In one embodiment, the difference image generated by the image difference unit (230) can display changes in patterns, markers, etc. located on the surface of the ball. In one embodiment, the pattern or marker of the ball may not be predetermined or may be irregular. However, it is not limited thereto, and the embodiment of the present disclosure may also be applied when a predetermined or rule-based pattern or marker is used. In one embodiment, if there is no special pattern or marker on the surface of the ball, the shape of the ball itself may be used. For example, if no special pattern or marker is displayed on the golf ball, changes in the dimples on the surface of the golf ball can be generated as a difference image.
[0079] According to one embodiment of the present disclosure, the image processing unit (200) may further include an image preprocessing unit (220).
[0080] In one embodiment, the image preprocessing unit (220) may be a means for preprocessing target images prior to the difference of target images. In one embodiment, the image preprocessing unit (220) may normalize the target images. For example, the size of the ball may be captured differently in the target images depending on the degree to which the moving ball approaches the camera, but since the target images are images of the same ball, the target images may be normalized so that the size of the ball included in the target images is constant prior to the difference of the target images. In addition, since the influence of ambient lighting varies depending on the position of the ball, the brightness of the target images may vary entirely or partially, so the brightness of the target images may be adjusted and normalized prior to the difference of the target images.
[0081] In one embodiment, the image preprocessing unit (220) can remove background parts or various noise parts from the target image in advance. For example, the image preprocessing unit (220) can remove the background parts in advance by comparing a base image that includes only background parts and does not include balls with the target image. For example, the image preprocessing unit (220) can recognize and remove the noise parts of the target image in advance by using a noise filter.
[0082] In the previously described embodiment, the image preprocessing unit (220) was described as preprocessing target images, but it is not limited thereto. For example, the image preprocessing unit (220) may be used as a means to preprocess a plurality of original images obtained from the sensor unit (100). Furthermore, the functions of the image preprocessing unit (220) described in the previously described embodiment may not be implemented separately as an image preprocessing unit (220), but may be implemented as a function of the target image extraction unit (210) and applied directly during the process of extracting the target image. For example, the target image extraction unit (210) may obtain the target image by undergoing normalization of the size and brightness of the ball image during the process of extracting the ball image.
[0084] Figures 5(a) and (b) are exemplary diagrams illustrating the process of obtaining a difference image from a target image.
[0085] In Figures 5(a) and (b), the images in column A are target images taken in succession of images of a moving ball, and the images in column B are difference images showing the difference between two consecutive target images.
[0086] According to one embodiment of the present disclosure, an image processing unit (200) can generate a difference image for calculating the spin direction and spin speed of a ball from a target image. Referring to FIG. 5, the image processing unit (200) calculates the difference between image (2) and image (1) and stores it as a difference image. For example, if the value corresponding to each pixel of image (2) is subtracted from the value corresponding to each pixel of image (1), the same pixel becomes '0', and only the pixel corresponding to the difference between image (2) and image (1) has a value. For example, when comparing images (1) and (2) of FIG. 5, the background and the golf ball are almost identical, and only the change in marking due to spin can be perceived as the main difference, and the image in column B, expressed as the difference between images (1) and (2), displays the change in marking. Ultimately, the difference image, which is the difference between images (2) and (1), can be stored as an image in column B.
[0087] According to one embodiment, the image processing unit (200) calculates the difference between image (3) and image (2) again and stores it as a difference image. That is, the difference between image (3) and image (2) can be calculated and stored as a difference image. For example, if the value corresponding to each pixel of image (3) is subtracted from the value corresponding to each pixel of image (2), the same pixel becomes '0', and only the pixel corresponding to the difference between image (3) and image (2) has a value, and this result can be stored as a difference image.
[0088] According to one embodiment, the image processing unit (200) can repeat this process to sequentially calculate difference images of target images (1) to (10) and generate difference images (2) to (10).
[0090] FIG. 6 is a conceptual diagram illustrating an embodiment of a spin calculation unit that calculates the spin of a ball from a difference image.
[0091] According to one embodiment of the present disclosure, the spin calculation unit (300) can receive at least one difference image from the image processing unit (200) and calculate the spin of the ball. According to one embodiment, the spin calculation unit (300) can receive at least one difference image and output the spin direction and spin speed of the moving ball detected by the sensor unit (100). For example, the spin calculation unit (300) can receive the difference image and output (v, u). Here, v is a vector value representing the direction of the rotation axis of the ball spin, and u is a scalar value (u) representing the number of rotations per unit time (e.g., RPM) of the ball spin speed.
[0092] In one embodiment, the spin calculation unit (300) may receive at least one difference image generated by differentiating target images taken at predetermined time intervals during a predetermined time period among target images of a moving ball taken continuously. For example, image (2) in column B, which is one of the difference images of target images (1) to (10) of FIG. 5, may be selected as input. As another example, image (6) in column B, which is one of the difference images of target images (1) to (10) of FIG. 5, may be selected as input.
[0093] In one embodiment, the spin calculation unit (300) can calculate the spin of the ball by receiving a plurality of difference images among the difference images of target images that have been continuously captured of the moving ball. For example, among the difference images of target images (1) to (10) of FIG. 5, image (2) of column B and image (6) of column B can be selected as inputs.
[0094] According to one embodiment, the spin calculation unit (300) may include an artificial neural network (350) capable of machine learning. The spin calculation unit (300) may be trained by training data in which the artificial neural network (350) takes a difference image as input and outputs the spin direction and spin speed of the ball. The artificial neural network (350) may be trained to take the difference image of the training data as input and output a known spin direction and spin speed of the ball. Although the learning of the artificial neural network (350) was previously described using supervised learning as an example, it is not limited thereto, and unsupervised learning, semi-supervised learning, or reinforcement learning may be used alone or in combination. Additionally, for machine learning, gradient descent, regression techniques, probability-based methods, geometry-based methods, or ensemble-based methods may be used, but are not limited thereto.
[0095] According to one embodiment, the artificial neural network (350) of the spin calculation unit (300) may be composed of a convolutional neural network, a recurrent neural network, a deep neural network, etc. However, it is not limited thereto, and any type of artificial neural network may be used as long as it can receive a difference image and infer the spin direction and spin speed of the ball.
[0097] FIG. 7 is a block diagram of a ball spin calculation device according to one embodiment of the present disclosure.
[0098] Referring to FIG. 7, the ball spin calculation device (700) may include a transceiver (710), a memory (720), and a processor (730). Additionally, the processor (730) may be implemented as a software module. However, not all components shown in FIG. 7 are essential components of the ball spin calculation device (700). The ball spin calculation device (700) may be implemented with more components than those shown in FIG. 7, or with fewer components than those shown in FIG. 7. Furthermore, the transceiver (710), the memory (720), and the processor (730) may be implemented in the form of a single chip.
[0099] In one embodiment, the transceiver (710) may communicate with a terminal, server, or other electronic device connected to the ball spin calculation device (700) via wired or wireless connection. For example, the transceiver (710) may receive images of a moving ball from a sensor and transmit the spin direction and spin speed of the moving ball to another server.
[0100] Various types of data, such as programs and files, such as applications, can be installed and stored in the memory (720). The processor (730) may access and use the data stored in the memory (720), or may store new data in the memory (720).
[0101] The processor (730) controls the overall operation of the ball spin calculation device (700) and may include at least one processor, such as a CPU, GPU, etc. The processor (730) may control other components included in the ball spin calculation device (700) to perform operations to operate the ball spin calculation device (700). For example, the processor (730) may execute a program stored in memory (720), read a stored file, or save a new file. In one embodiment, the processor (730) may perform operations to operate the ball spin calculation device (700) by executing a program stored in memory (720).
[0102] One embodiment of the present disclosure may also be implemented in the form of a recording medium comprising computer-executable instructions, such as program modules executed by a computer. A computer-readable medium may be any available medium accessible by a computer and includes both volatile and non-volatile media, and both removable and non-removable media. Additionally, a computer-readable medium may include both computer storage media and communication media. A computer storage medium includes both volatile and non-volatile, removable and non-removable media implemented by any method or technique for storing information, such as computer-readable instructions, data structures, program modules, or other data. A communication medium typically includes computer-readable instructions, data structures, or program modules and includes any information transmission medium.
[0103] The foregoing description of the present disclosure is for illustrative purposes only, and those skilled in the art will understand that other specific forms can be easily modified without altering the technical spirit or essential features of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single unit may be implemented in a distributed manner, and components described as distributed may likewise be implemented in a combined form.
[0104] The scope of the present disclosure is defined by the claims set forth below rather than by the detailed description above, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts thereof should be interpreted as being included within the scope of the present disclosure. Explanation of the symbols
[0106] 10: Computing device 20: Ball 100: Sensor section 110: High-speed camera 200: Image processing unit 210: Target Image Extraction Unit 220: Image Preprocessing Unit 230: Image difference section 300: Spin output unit 350: Artificial Neural Networks
Claims
Claim 1 A ball spin calculation method comprising: acquiring a series of images of a moving ball; extracting a portion of the ball from each of the series of images as a target image; generating a difference image by differentiating the target images; and inputting the difference image into an artificial neural network to calculate the spin direction and spin speed of the moving ball based on the output of the artificial neural network. Claim 2 A ball spin calculation method according to claim 1, wherein the difference image is obtained by subtracting two consecutive target images. Claim 3 A ball spin calculation method according to paragraph 2, wherein the difference image is obtained by subtracting the value corresponding to each pixel of one target image from the value corresponding to each pixel of another target image. Claim 4 A method for calculating ball spin according to claim 1, wherein the spin direction of the moving ball is the rotation axis vector of the ball, and the spin speed is the number of rotations per minute of the ball. Claim 5 A ball spin calculation method according to claim 1, wherein the artificial neural network is machine-learned to calculate the spin direction and spin speed of the moving ball using the difference image as input. Claim 6 A ball spin calculation method according to claim 1, wherein the difference image comprises two or more difference images. Claim 7 A ball spin calculation method according to claim 1, wherein the step of acquiring continuous images of the moving ball is performed by two or more camera devices. Claim 8 A ball spin calculation method according to claim 1, wherein the center coordinates of the ball image of the target image are located at the center. Claim 9 A ball spin calculation method according to claim 1, wherein the target image is normalized so that the size of the ball image is the same. Claim 10 A ball spin calculation method according to claim 1, wherein the target image is normalized so that the brightness of the ball image is uniform. Claim 11 A ball spin calculation device comprising: a sensor unit for acquiring continuous images of a moving ball; a target image extraction unit for extracting a portion of the ball from each of the continuous images as a target image; an image difference unit for generating a difference image by differentiating the target images; and a spin calculation unit for inputting the difference image into an artificial neural network and calculating the spin direction and spin speed of the moving ball based on the output of the artificial neural network. Claim 12 A ball spin calculation device according to claim 11, wherein the image difference unit obtains the difference image by subtracting two consecutive target images. Claim 13 A ball spin calculation device according to claim 12, wherein the image difference unit obtains the difference image by subtracting the value corresponding to each pixel of one target image from the value corresponding to each pixel of another target image. Claim 14 A ball spin calculation device according to claim 11, wherein the spin direction of the moving ball is the rotation axis vector of the ball, and the spin speed is the number of rotations per minute of the ball. Claim 15 A ball spin calculation device according to claim 11, wherein the artificial neural network is machine-learned to receive the difference image as input and output the spin direction and spin speed of the moving ball. Claim 16 In claim 11, the ball spin calculation device comprises two or more difference images, wherein the difference image comprises two or more difference images. Claim 17 In claim 11, the sensor unit comprises two or more camera devices, a ball spin calculation device. Claim 18 In claim 11, the ball spin calculation device, wherein the target image extraction unit extracts the target image such that the center coordinates of the ball image are located at the center of the target image. Claim 19 A ball spin calculation device according to claim 11, further comprising an image preprocessing unit, wherein the image preprocessing unit normalizes the size of the ball image of the target image to be the same. Claim 20 A ball spin calculation device according to claim 11, further comprising an image preprocessing unit, wherein the image preprocessing unit normalizes the brightness of the ball image of the target image so that it is uniform. Claim 21 A program stored on a computer-readable recording medium to execute the method of any one of paragraphs 1 through 10 on a computer.
Citation Information
Patent Citations
Flight parameter measurement system
JP2005529339A
Determination and analysis of motion and spin characteristics in golf shots
JP2014500078A
Device for sensing moving ball and method for the same, and image processing method of ball image for calculating spin of moving ball
KR101472275B1
Device for sensing moving ball and method for same, and method for processing ball image to calculate spin of moving ball
KR1020140090008A
