Methods and systems for locating a golf ball

Abstract

A method for locating a golf ball including changing a temperature of a golf ball from a first temperature to a second temperature before use or marking the ball by reflective (mirror) or fluorescent material (e.g., NIR-IR fluorescent dye). The temperature changed ball is struck. Using either a thermal imaging camera with an imaging processing unit or a near-infrared (NIR) imaging camera with an imaging processing unit to produce a digital image of a part of the golf course with a potential golf ball location. An image processing technique is applied to produce an enhanced image of the golf ball location. A thermal imaging camera and a NIR imaging camera for locating a golf ball are described. A non-transitory computer readable media is described.

Claims

1. A method for locating a golf ball, the method comprising: changing a temperature of a golf ball from a first temperature to a second temperature before use; striking the temperature changed ball on a golf course; using a thermal imaging camera with an imaging processing unit to produce a digital image of a part of the golf course with a potential golf ball location; and applying an image processing technique to produce an enhanced image of the golf ball location, wherein the image processing technique is based on image focusing using special transformation of the observed data involving focusing minimum gradient support (MGS) stabilizer, which minimizes the total area with nonzero gradients of brightness and thus generates a sharp and focused image of the ball, and by using the multinary transformation approach, wherein the brightness is described by a brightness distribution, {tilde over ()}.sub.i, using a superposition of error function: ${\stackrel{}{}}_i=E\left({}_i\right)=c{}_i+\frac{1}{2}\underset{j=1}{\overset{P}{.Math.}}\left[1+\mathrm{erf}\left(\frac{{}_i-{}^{\left(j\right)}}{\sqrt{2}{}_j}\right)\right]$ where ={.sub.i},i=1, . . . , N.sub.m, is an original vector of model parameters, {tilde over ()}={{tilde over ()}.sub.i},i=1, . . . , N.sub.m, is a new vector of the nonlinear parameters, and P is a total number of discrete (multinary) values of the brightness, .sup.(j), function E(.sub.i) is an error function, parameter .sub.j is a standard deviation of the value .sup.(j), and constant c is a small number to avoid singularities in calculation of derivatives of multinary brightness.

2. A method according to claim 1, wherein a difference between the first temperature and the second temperature is more than one degree Celsius.

3. A method according to claim 1, wherein the image processing technique is based on the multinary transformation approach, which processes the original image into the multinary image with the properties that the brightness distribution is characterized by a finite number of discrete values of the brightness with the preassigned value of 0 for the ball location to produce a bright and focused image of the ball location.

4. A method according to claim 1, wherein changing the temperature of the ball before its use by the golfer further comprises cooling the golf ball.

5. A method according to claim 1, wherein changing the temperature of the ball before its use by the golfer further comprises heating the golf ball.

6. A non-transitory computer readable medium having instructions thereon that are executable to apply the image processing technique of claim 1 to produce an enhanced image of the golf ball location.

7. A thermal imaging camera for locating a golf ball, the thermal imaging camera comprising: a processor; an image processing unit; and memory having instructions executable to: produce a digital image of a part of a golf course with a potential golf ball location; and apply the image processing technique of claim 1 to produce an enhanced image of the golf ball location.

8. A method according to claim 1, wherein the image processing minimizes the total area with nonzero gradients of the brightness using the following equation: $S\left(M_1\right)=\frac{{\left[M_1\left(x,y\right)\right]}^2}{{\left[M_1\left(x,y\right)\right]}^2+e^2}ds$ where M.sub.1 is an enhanced image and e controls the sharpness of the image.

9. A non-transitory computer readable media including instructions stored thereon that are executable to: obtain a digital image of the part of the golf course with a potential golf ball location; and apply an image processing technique to produce an enhanced image of a golf ball location, wherein the image processing technique is based on image focusing using special transformation of the observed data involving focusing minimum gradient support (MGS) stabilizer, which minimizes the total area with nonzero gradients of brightness and thus generates a sharp and focused image of the ball, and by using the multinary transformation approach, wherein the brightness is described by a brightness distribution, {tilde over ()}.sub.i, using a superposition of error function: ${\stackrel{}{}}_i=E\left({}_i\right)=c{}_i+\frac{1}{2}\underset{j=1}{\overset{P}{.Math.}}\left[1+\mathrm{erf}\left(\frac{{}_i-{}^{\left(j\right)}}{\sqrt{2}{}_j}\right)\right]$ where ={{tilde over ()}.sub.i}, i=1, . . . , N.sub.m, is an original vector of model parameters, {tilde over ()}={{tilde over ()}.sub.i}, i=1, . . . , N.sub.m, is a new vector of the nonlinear parameters, and P is a total number of discrete (multinary) values of the brightness, .sup.(j), function E(.sub.i) is an error function, parameter .sub.j is a standard deviation of the value .sup.(j), and constant c is a small number to avoid singularities in calculation of derivatives of multinary brightness.

10. A non-transitory computer readable media according to claim 9, wherein the image processing technique includes receiving thermal information.

11. A non-transitory computer readable media according to claim 9, wherein the image processing technique includes receiving NIR information.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) To further clarify the above and other advantages and features of the present invention, a more particular description of the invention will be rendered by reference to specific implementations thereof which are illustrated in the appended drawings. It is appreciated that these drawings depict only illustrated implementations of the invention and are therefore not to be considered limiting of its scope. The invention will be described and explained with additional specificity and detail through the use of the accompanying drawings in which:

(2) FIG. 1 illustrates an implementation of a method for golf ball detection using a thermal imaging camera. It presents a plot of an ambient heating (to room temperature) of the cooled golf balls.

(3) FIG. 2 illustrates an implementation of a method for golf ball detection using a thermal imaging camera with imaging processing unit. It shows a detection of a cold golf ball located in lab from 10 m distance. The blue color of the spot on cold golf ball corresponds to its relatively lower temperature (21.1 C.), which is lower than room temperature (24.5 C.).

(4) FIG. 3 illustrates an implementation of a method for golf ball detection using a thermal imaging camera with imaging processing unit. It shows a detection of a cold golf ball located in Lab from 8 m distance shielded by A4 paper. The blue color of the spot on the cold golf ball can still be seen even if it corresponds to just one-degree difference between the temperature of the ball (23.3 C.) and the room temperature (24.5 C.).

(5) FIG. 4 illustrates an implementation of a method for golf ball detection using a thermal imaging camera with imaging processing unit. It shows a detection of cold golf ball located in lab from 20 m distance.

(6) FIG. 5 illustrates the detection of the golf ball marked by a reflective (mirror) or fluorescent material and illuminated by the NIR-IR source.

DETAILED DESCRIPTION

(7) The implementations disclosed herein relate in general to location of the golf balls lost by golfer in, for example, the thick grass (e.g., rough).

(8) In one implementation of this disclosure, the golf balls may be cooled by keeping them in a portable cooler. In another implementation of the disclosure, the golf balls may be heated by keeping them in a portable ball warmer. When the golfer is ready to use the ball, he or she takes the ball from the cooler/heater and hits it. The precooled/preheated ball has a temperature well below/above the temperature of the surrounding environment, including ground and grass, thus providing a significant contrast of the temperature between the ball and the ground/grass.

(9) It takes some time for the ball to warm up (or cool down) and to reach an equilibrium with the temperature of the environment. FIG. 1 presents a plot of an ambient heating (to room temperature) of the cooled golf balls. One can see that it takes about 30 minutes for the temperature of a standard ball to raise from a typical cooler temperature (around 12 C.) up to the room temperature (around 26 C.). Thus, during almost half an hour, there is a temperature contrast of a few degrees C., strong enough to discriminate the ball from the surrounding environment.

(10) As an example, FIG. 2 shows a detection of a cold golf ball located in lab from 10 m distance using a thermal imaging camera. The blue color of the spot on cold golf ball corresponds to its relatively lower temperature (21.1 C.), which is lower than room temperature (24.5 C.).

(11) The golfer may have a situation where the ball is hidden in the long grass. This situation is imitated in the lab by shielding the ball with a standard A4 paper. FIG. 3 shows that a cold golf ball located in Lab from 8 m distance and shielded by A4 paper can still be easily detected even if the difference between the temperature of the ball (23.3 C.) and the room temperature (24.5 C.) is just about one-degree C.

(12) FIG. 4 shows a detection of cold golf ball located in lab from 20 m distance.

(13) In the implementations of this disclosure, the golfer uses a thermal imaging camera with imaging processing unit which is computer program to find the ball. The digital image generated by the camera may be processed using a specialized image processing technique, which may be optimized to enhance the ball location.

(14) In another embodiment of this disclosure, electromagnetic radiation is used to image a golf ball. The frequency must be low enough to penetrate organic matter such as leaves, but high enough to retain resolution on the size of the golf ball. The optimal solution will likely fall in the 30 GHz to 30 THz range.

(15) Depending on the frequency chosen, there may be enough ambient radiation to illuminate the golf ball. This would enable a passive sensor and simplify the instrument design. However, it may be necessary to create an active source to provide enough energy to illuminate the golf ball in the desired frequency range.

(16) To increase the detectability of the golf ball, a reflective or fluorescent material will be added. This could include a material with a unique spectral signature to aid in identification of the ball. In one implementation, NIR-IR spectroscopy may be used to detect the golf balls covered either by NIR-IR dye or NIR-IR mirror. IR light should penetrate to a certain extend through grass and soil. The general scheme of the NIR-IR golf ball location system is proposed in FIG. 5.

(17) In one implementation of this disclosure, the proposed system uses a thermal imaging camera with imaging processing unit which is computer program to find the ball.

(18) In another implementation of this disclosure, the proposed system uses a NIR imaging camera imaging camera with imaging processing unit which is computer program to find the ball.

(19) In yet another implementation of the disclosure, the digital image processing technique is based on image focusing using special transformation of the observed data which produces the enhanced image of the golf ball location.

Example 1

(20) The following is an example of at least some of the principles of the golf ball imaging reconstruction that is offered to assist in the practice of the disclosure. It is not intended thereby to limit the scope of the disclosure to any particular theory of operation or to any field of application.

(21) Supposing the image of the part of the golf course with the potential ball location has been obtained by the thermal vision (or NIR vision) camera, and it is denoted by M.sub.0.

(22) Our goal for image enhancement is to find an image M.sub.1 close enough to M.sub.0 with small variation within target area and producing the focused image of the ball. This image processing problem can be represented mathematically as the minimization of the following functional:
P(M.sub.1)=M.sub.1M.sub.0.sub.L.sub.2.sup.2+S(M.sub.1)(1)
where the first term is the Euclidean distance between the original image M.sub.0 and the enhanced image M.sub.1; and the second term imposes additional constraints such as focusing stabilizer (Zhdanov, 2015).

(23) For example, the focusing of the image of the golf ball can be achieved by using the minimum gradient support (MGS) stabilizer:

(24) $\begin{array}{cc}S\left(M_1\right)={}_{}^{}\frac{{\left[M_1\left(x,y\right)\right]}^2}{{\left[M_1\left(x,y\right)\right]}^2+e^2}\mathrm{ds}& \left(2\right)\end{array}$

(25) The above functional can minimize the total area with nonzero gradients and helps generate a sharp and focused image of the ball. The small number e controls the sharpness of the image.

(26) The MGS functional can also be expressed as pseudo-quadratic functional as follows:

(27) $\begin{array}{cc}\frac{{\left[M_1\left(x,y\right)\right]}^2}{{\left[M_1\left(x,y\right)\right]}^2+s^2}\mathrm{ds}={\left[w_e\left(x,y\right)M_1\left(x,y\right)\right]}^2\mathrm{ds}={.Math.W_e{M}_1.Math.}_{L_2}^2\mathrm{where}& \left(3\right)\\ w_e\left(x,y\right)=\frac{M_1\left(x,y\right)}{\sqrt{{\left[M_1\left(x,y\right)\right]}^2+e^2}\sqrt{{\left[M_1\left(x,y\right)\right]}^2+{}^2}}& \left(4\right)\end{array}$

(28) Note that, the MGS stabilizer functional in a general case is a nonlinear functional of M.sub.1, and it is not quadratic. By representing it in a pseudo quadratic form one can use the optimization technique developed for quadratic functional.

(29) In summary, the image enhancement technique is formulated within the general framework of the inverse problem solution, where the observed data are the original thermal image M.sub.0, and model parameters to be determined represent the enhanced image M.sub.1. By applying the Tikhonov regularization approach, one can solve the image enhancement problem (Zhdanov, 2015).

(30) In yet another implementation of the present disclosure, the digital image processing technique may be based on using special multinary transformation of the observed data which produce the enhanced image of the golf ball location.

Example 2

(31) The following is an example of at least some of the principles of the golf ball imaging reconstruction that is offered to assist in the practice of the disclosure. It is not intended thereby to limit the scope of the disclosure to any particular theory of operation or to any field of application. For example, the focusing of the image of the golf ball can be achieved by using the multinary transformation approach. In a general case, the brightness distribution of the recovered image is described by a continuous function. In ball detection problem, the desired image brightness is described by the binary function as follows:
m.sub.i={m.sub.i.sup.(1)=0,m.sub.i.sup.(2)=1}(5)
or by the ternary function:
m.sub.i={m.sub.i.sup.(1)=1,m.sub.i.sup.(2)=0,m.sub.i.sup.(3)=1}(6)

(32) Further, we can extend the description of the brightness distribution using the multinary function of order P, having discrete numbers of values:
m.sub.i={m.sub.i.sup.(1),m.sub.i.sup.(2)=0, . . . ,m.sub.i.sup.(p)}(7)

(33) In above distribution, the constant value 0 is assigned to the brightness of the image representing the ball, while all other values are assigned to the image of the surrounding environment.

(34) In yet another implementation of this disclosure, the nonlinear transformation of the multinary function into the continuous function, can be described as follows. We transform our brightness distribution, .sub.i, into a model space defined by a continuous range of multinary brightness, {tilde over ()}.sub.i, using a superposition of error function:

(35) $\begin{array}{cc}{\stackrel{~}{}}_i=E\left({}_i\right)=c{}_i+\frac{1}{2}\underset{j=1}{\overset{p}{.Math.}}\left[1+\mathrm{erf}\left(\frac{{}_i-{}^{\left(j\right)}}{\sqrt{2}{}_j}\right)\right]& \left(8\right)\end{array}$
where ={.sub.i}, i=1, . . . , N.sub.m, is the original vector of the model parameters; {tilde over ()}={{tilde over ()}.sub.i}, i=1, . . . , N.sub.m, is a new vector of the nonlinear parameters; and P is a total number of discrete (multinary) values of the model parameter (brightness), .sup.(j). The function E(.sub.i) is the error function; parameter .sub.j is a standard deviation of the value .sup.(j); and the constant c is a small number to avoid singularities in the calculation of the derivatives of the multinary brightness.

(36) Thus, using the above transformation (8), we can process the original thermal image M.sub.0, into the multinary image M.sub.1, with the properties that the brighness distribution is characterized by a finite number of discrete values of the brighness with the preassigned value of 0 for the ball location. As a result, we produce a bright and focused image of the ball location.

(37) The present disclosure may be embodied in other specific forms without departing from its spirit or essential characteristics. The described implementations are to be considered in all respects only as illustrative and not restrictive. The scope of the invention is, therefore, indicated by the appended claims rather than by the foregoing description. All changes which come within the meaning and range of equivalency of the claims are to be embraced within their scope.