JP2010101828A - Time measuring device and time measuring method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a time measuring device and a time measuring method capable of measuring a time inexpensively with sufficient flexibility of a measuring section, and without installing a machine such as a sensor on a moving body. <P>SOLUTION: A boat detection means 110 detects a boat 320 from a moving image wherein a start spot and a goal spot are imaged, and a start/goal detection means 130 detects that the boat 320 has passed the start spot and the goal spot. A time calculation means 140 calculates from a frame wherein the start spot is passed up to a frame wherein the goal spot is passed in images constituting the moving image, and calculates a time of the moving body required from the start spot to the goal spot from the calculated number of frames. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a time measuring device and a time measuring method, and more particularly to a time measuring device and a time measuring method for measuring a time from a start of a moving object to be measured to a goal.

As a method of measuring the time from the start of a moving object to the goal, there is a method in which a time measurer manually measures using a stopwatch.
In the case of such manual time measurement, there are problems that the measurement error is larger than that of mechanical measurement and that the time measurement person's skill and tone are affected.

On the other hand, as a method of mechanical measurement, a method using a slit camera (for example, refer to Patent Document 1) and a method for installing a sensor on a moving body (for example, refer to Patent Document 2) have been proposed.
According to such mechanical measurement, time can be measured with higher accuracy than manual time measurement.
Japanese Patent Laid-Open No. 9-121344 JP 2006-259967 A

However, when performing mechanical measurement using a slit camera as disclosed in Patent Document 1, there is a problem that the slit camera itself is very expensive and cannot be easily installed.
In addition, it is difficult to freely move the installation location of the slit camera, and only the time in the specific section where the slit camera is installed can be measured, and there is a problem that flexibility is poor.

  Moreover, even if it is going to install a sensor in the moving body which is a measuring object like patent document 2, and is going to measure a machine, when a big restriction | limiting is imposed on providing a certain machine in a moving body like a public competition etc., for example However, there is a problem that time cannot often be measured by this method.

  The present invention has been made in view of the above points, and is inexpensive, rich in flexibility of a measurement section, and capable of measuring time without installing a machine such as a sensor on a moving body, and a time An object is to provide a measurement method.

  In the present invention, in order to solve the above problem, in the time measuring device for measuring the time from the start to the goal of the moving object to be measured, the moving object detecting means for detecting the moving object from the moving image obtained by imaging the start point and the goal point And passage detection means for detecting that the moving body has passed the start point and the goal point, and from a frame that has passed the start point to a frame that has passed the goal point in an image constituting the moving image. And a time calculating means for calculating the time of the moving object required from the start point to the goal point from the calculated number of frames.

  Thereby, the moving object detecting means detects the moving object from the moving image obtained by imaging the start point and the goal point, and the passage detecting means detects that the moving object has passed the start point and the goal point, and the time The calculating means calculates from the frame that has passed through the start point to the frame that has passed through the goal point in the image constituting the moving image, and required from the calculated start point to the goal point from the calculated number of frames. The time of the moving object is calculated.

  Further, in the present invention, in the time measuring method for measuring the time from the start to the goal of the moving object to be measured, the moving object detecting means detects the moving object from a moving image obtained by imaging the start point and the goal point; A step of detecting that the moving object has passed the start point and the goal point; and a time calculating unit from the frame that has passed the start point in the image constituting the moving image. And calculating the time required for the moving object from the start point to the goal point based on the calculated number of frames. .

  Thereby, the moving object detecting means detects the moving object from the moving image obtained by imaging the start point and the goal point, and the passage detecting means detects that the moving object has passed the start point and the goal point, and the time The calculating means calculates from the frame that has passed through the start point to the frame that has passed through the goal point in the image constituting the moving image, and required from the calculated start point to the goal point from the calculated number of frames. The time of the moving object is calculated.

  According to the time measuring device of the present invention, a moving object is detected from a moving image obtained by imaging the start point and the goal point, and the detected moving object is determined from the number of frames up to the frame that passes the start point and the frame that passes the goal point. Since the time is measured, the time can be measured as long as there is a moving image of the moving object that is the target of time passing through the start point and the goal point.

  Therefore, it is not necessary to use an expensive camera such as a special slit camera only for measuring time, and the measurement section can be changed any number of times by changing the angle of the imaging device that captures moving images. Therefore, it is possible to measure time without mounting a special device such as a sensor for measuring time on a moving object that is rich in measurement.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
In the present embodiment, description will be made using an example in which the display time of a trial run performed before a race of a boat race is measured.

FIG. 1 is a schematic diagram of a time measuring device according to the present embodiment.
As shown in FIG. 1, a camera 210 and a moving image storage device 220 are connected to the time measuring device 100.

  The time measuring apparatus 100 includes a boat detecting unit 110 that detects a boat 320 that is a moving object of time measurement, a noise removing unit 120 that removes noise such as white waves generated by the boat 320 in a moving image, and the boat 320 is a starting point. And a start / goal detecting means 130 for detecting whether or not the goal point has been passed, and a time calculating means 140 for calculating the time from the start to the goal.

  The camera 210 images the imaging region 310 including the start point and the goal point in the boat racing course 300.

The imaging area 310 can be changed in any way by changing the angle or zoom of the camera 210.
Specifically, the imaging area 310 is set to capture a straight course, but if you want to measure the speed of the curve, you can pass through the curve by changing the angle and zoom so that the curve is captured. It is also possible to measure the time to perform.

  The moving image data captured by the camera 210 is directly input to the time measuring device 100 or stored in the moving image storage device 220.

  When a moving image captured by the camera 210 is directly input, or when a moving image file stored in the moving image storage device 220 is input, the boat detection unit 110 of the time measuring device 100 detects a time measurement target from the input moving image. A boat 320 is detected.

The noise removing unit 120 removes noise from the moving image in which the boat 320 is detected.
The start / goal detection means 130 detects whether or not the moving object has passed a preset start point and goal point from a moving image from which noise has been removed.

  The start / goal detection means 130 stores the number of frames of the moving image when it is detected that the boat 320 has passed the start point and the goal point.

The time calculation means 140 passes the goal point after the boat 320 passes the start point from the number of frames stored by the start / goal detection means 130 when the boat 320 detects that the start point and the goal point have passed. Calculate the time until
Here, it demonstrates as an example which posts the exhibition time performed in order to see the condition of each boat which goes into a race before a race starts.

  Since the exhibition time from when the boat 320 passes the start point to when it passes the goal point is calculated based on the number of frames of the moving image, the frame number correction process is performed to calculate the frame number more accurately. I do.

Next, the calculation method and processing will be specifically described using a flowchart.
FIG. 2 is a flowchart showing a time measurement process performed by the time measurement device according to the present embodiment.

In the following, the process illustrated in FIG. 2 will be described in order of step number.
[Step S11] When the boat detection means 110 starts the time measurement process, it resets the frame counter of the moving image.
The time measurement process is started by detecting, for example, the user pressing the process start button.

[Step S12] The boat detection means 110 determines whether or not there is an instruction to end the time measurement process.
If there is an instruction to end the time measurement process, the process ends. If there is no instruction to end the time measurement process, the process proceeds to step S13.

  The determination as to whether or not there has been an instruction to end the time measurement process is made, for example, by detecting that the user has pressed the process end button.

[Step S13] The boat detection means 110 adds 1 to the frame counter.
[Step S14] The boat detection unit 110 reads the value of the frame counter of the moving image, and cuts out the frame image of the counter value (hereinafter, the counter value is N) as a bitmap.
Further, the clipped image is converted to gray scale and stored in a ring buffer.

[Step S15] The boat detection unit 110 determines whether or not the value of the frame counter exceeds 3.
If it exceeds 3, the process proceeds to step S16, and if it is 3 or less, the process returns to step S12.

[Step S16] The boat detection means 110 takes the difference between the Nth frame image and the (N-3) th frame image.
Here, how to take the difference will be specifically described with reference to the drawings.

FIG. 3 is a diagram showing an outline of processing for obtaining a difference between images by the boat detection means.
As shown in FIG. 3, when there is a moving image composed of images from the (N-4) th frame to the Nth frame, the boat detecting means 110 extracts the (N-3) th and Nth frame images. The difference between the two images is taken.
Specifically, the difference is obtained by subtracting the common part of both images from the Nth frame image.

The difference between the images of the Nth frame and the (N-3) th frame is the boat that is the moving object of the time measurement so that if the frames are too close together, the difference is too small and is confused with noise. This is because the detection area to be detected as 320 may be too small.
In addition, if the frame is too far away, a wave whose moving speed is slower than that of the boat 320 is detected without being canceled when a difference is taken between the frames, so that it may be erroneously detected as a moving object. It is.

In the present embodiment, a difference is obtained with images separated by three frames, but it can be arbitrarily determined how many frames are separated from each other.
Next, returning to the time measurement process, the description starts from step S17.

  [Step S17] The boat detection means 110 binarizes the difference image based on a certain threshold value (hereinafter, this threshold value is referred to as a threshold value A), and pixels that exceed the threshold value A are moving object regions, that is, imaged. The area 310 is determined to be a portion where the boat 320 exists.

[Step S18] The noise removing unit 120 performs thinning / thickening processing on the binarized image, and extracts a common portion between the thinned / thickening image and the original image.
Here, the thinning / thickening process will be specifically described with reference to the drawings.

FIG. 4 is a diagram for explaining the concept of the thinning process.
4A is a diagram showing an image before thinning processing, FIG. 4B is a diagram showing a first stage of thinning processing, and FIG. 4C is thinning. It is a figure which shows the 2nd step of a process.

As shown in FIG. 4, in the first stage of the thinning process, a process for deleting the surrounding one pixel from the image before the thinning process is performed.
In FIG. 4, the deleted portion is indicated by a slanting line with a lower right.

  Further, in the second stage of the thinning process, a process for further deleting the surrounding one pixel from the image subjected to the first stage process is performed.

  As described above, the thinning process refers to a process of making a binarized image area small and thin (thinning).

  In general, thinning processing imposes a restriction that only makes a binarized image smaller or thinner and does not completely erase what is already small and thin. Therefore, the restriction is not performed.

FIG. 5 is a diagram for explaining the concept of the thickening process.
FIG. 5A is a diagram illustrating an image before the thickening process, FIG. 5B is a diagram illustrating a first stage of the thickening process, and FIG. 5C is a thickening process. It is a figure which shows the 2nd step of a process.

As shown in FIG. 5, in the first stage of the thickening process, a process for adding one pixel around the image before the thickening process is performed.
In FIG. 5, the added portion is indicated by a diagonal line rising to the right.

  Further, in the second stage of the thickening process, a process of adding one pixel further from the image subjected to the first stage process is performed.

  As described above, the thickening process refers to a process of making the binarized image area larger or thicker (thickening).

By performing such thinning / thickening processing, noise in the image can be removed.
Next, a procedure for removing noise by performing thinning / thickening processing on an image will be specifically described with reference to the drawings.

FIG. 6 is a diagram illustrating a procedure for removing noise by thinning / thickening processing.
FIG. 6A shows an image before the noise removal process, and FIG. 6B shows an image after the thinning process is performed on the image before the noise removal process. FIG. 6C is a diagram illustrating an image after the thickening process is performed on the image after the thinning process, and FIG. 6D is a noise removal process. FIG.

As shown in FIG. 6, when the thinning process is performed on the original image before the noise removal process, all small points such as noise are deleted.
On the other hand, a relatively large object such as the boat 320 is not completely deleted even if the thinning process is performed, and the central portion of the boat 320 remains without being deleted.

  When the thinning process is performed on the thinned image, the completely deleted noise portion is not thickened, but only the central portion of the remaining boat 320 is thickened. Is made.

When thinning / thickening processing is performed by such a procedure, noise can be removed by deleting only a small range of pixel data such as noise.
Next, returning to the time measurement process, description will be made from step S19.

[Step S19] The noise removing unit 120 stores the image subjected to the noise removing process in a ring buffer.
[Step S20] The start / goal detection means 130 performs a start detection process.

Next, the start detection process performed by the start / goal detection means 130 will be specifically described with reference to a flowchart.
FIG. 7 is a flowchart of the start detection process performed by the start / goal detection means 130.
In the following, the process illustrated in FIG. 7 will be described in order of step number.

[Step S31] The start / goal detecting means 130 determines whether or not the boat 320 overlaps the start line preset in the course 300.
If it is determined that they overlap, the process proceeds to step S32. If it is determined that they do not overlap, the process ends.

  Whether or not the boat 320 and the start line overlap each other is determined when the boat 320 is reflected at a position in the image corresponding to the position of the preset start line.

[Step S32] The start / goal detecting means 130 determines whether or not the boat 320 and the start line overlap for the first time.
If it is determined that this is the first time, the process proceeds to step S33. If it is determined that this is not the first time, the process proceeds to step S35.
The number of frames for which it is determined that the boat 320 has overlapped with the start line for the first time will be described as the Mth frame.

[Step S33] The start / goal detection means 130 performs a passing frame calculation process.
Here, the time correction processing will be specifically described with reference to the drawings.

FIG. 8 is a diagram showing an outline of time correction processing performed by the start / goal detection means.
FIG. 8A is a diagram showing the positional relationship between the boat and the start line in the (M-1) th frame, and FIG. 8B is a diagram showing the positional relationship between the corrected boat and the start line. FIG. 8C is a diagram showing the positional relationship between the boat of the Mth frame and the start line.

As shown in FIG. 8, the boat 320 is positioned in front of the start line 330 in the (M−1) th frame.
On the other hand, the boat 320 has exceeded the start line 330 in the Mth frame.

  As will be described later, since the time from the start to the goal is calculated from the number of frames and the time per frame from the start to the goal, it is more accurate by correcting such an error between frames. Time can be calculated.

That is, with an integer number of frames, a measurement error occurs when the boat 320 crosses the start line 330 between frames as shown in FIG.
Therefore, a passing frame calculation process is performed based on the positional relationship between the boat 320 and the start line 330 in the (M-1) th frame, and the positional relationship between the boat 320 and the start line 330 in the Mth frame.

Next, the passing frame calculation process performed by the start / goal detecting means 130 will be described in detail with reference to a flowchart.
FIG. 9 is a flowchart showing a passing frame calculation process performed by the start / goal detecting means.
Hereinafter, the process illustrated in FIG. 9 will be described along the steps.

  [Step S41] The start / goal detection means 130 detects the distance between the boat 320 and the start line 330 in the Mth frame.

[Step S42] The start / goal detecting means 130 detects the distance between the boat 320 and the start line 330 in the (M-1) th frame.
As a method of detecting the distance between the boat 320 and the start line 330 in step S41 and step S42, it is performed by detecting the number of pixels from the foremost part of the boat 320 to the start line 330.

[Step S43] The virtual frame number is calculated from the ratio of the distance between the boat 320 and the start line 330 detected in Step S41 and Step S42.
For example, when the distance between the boat 320 and the start line 330 in the (M-1) th frame is 4, and the distance between the boat 320 and the start line 330 in the Mth frame is 6, the boat 320 is in the start line 330. Can be calculated that the boat 320 has crossed the start line 330 for the first time in the (M−1) + 0.4th frame.

Next, returning to step S34 in FIG.
[Step S34] The start / goal detection means 130 stores the virtual frame number calculated by the passing frame calculation process as a start frame.

  The start / goal detecting means 130 increments the start line counter indicating the number of times the boat 320 has overlapped with the start line 330, and resets the start timer indicating the elapsed time from the time exceeding the start line, and then starts. To finish the process.

  [Step S35] The start / goal detection means 130 reads the elapsed time since the boat 320 exceeded the previous start line 330 from the start timer, and a predetermined time or more (hereinafter, the threshold of this time is B) has elapsed. Judge whether or not.

If B or more has passed since the start line 330 was exceeded the previous time, the process proceeds to step S33, and if B has not been exceeded, the process ends.
The threshold value B is set to be shorter than the start interval when a plurality of boats 320 start, and to be set longer than the time until the pulling wave generated by the boat 320 disappears from the start line.

Boats and pulling waves are often not in one connected area, and if this pulling wave remains on the start line, it may be a pulling wave of the currently timed boat or another boat is on the start line I can't make a decision.
Therefore, by setting this threshold, a part of the pulling wave is erroneously detected as the head of the boat that is different from the boat that is currently timed and that has crossed the start line in order to measure the next time. Can be prevented.

Next, it returns to step S21 of FIG. 2, and demonstrates.
[Step S21] The start / goal detection means 130 and the time calculation means 140 perform goal detection processing.

Next, the goal detection processing performed by the start / goal detection means 130 and the time calculation means 140 will be specifically described with reference to a flowchart.
FIG. 10 is a flowchart showing a goal detection process performed by the start / goal detection means and the time calculation means.
In the following, the process illustrated in FIG. 10 will be described in order of step number.

[Step S51] The start / goal detecting means 130 determines whether or not the boat 320 overlaps the goal line preset in the course 300.
If it is determined that they have overlapped, the process proceeds to step S52. If it is determined that they have not overlapped, the process ends.

  Whether or not the boat 320 and the goal line overlap each other is determined when the boat 320 is reflected at a position in the image corresponding to a preset goal line position.

[Step S52] The start / goal detecting means 130 reads the goal counter indicating the number of times the boat 320 has overlapped the goal line, and the value of the start counter.
Then, it is determined whether or not the value of the start counter is larger than the value of the goal counter.

  If the value of the start counter is larger, the process proceeds to step S53, and if the value is the same or the value of the goal counter is larger, the process ends.

  It is usually impossible for the start counter value to be the same as the goal counter value or the start counter value to be smaller, and there is a high possibility of erroneous detection. Therefore, by making this determination, erroneous detection can be prevented.

[Step S53] The start / goal detecting means 130 determines whether or not the boat 320 and the goal line are overlapped for the first time.
If it is determined that this is the first time, the process proceeds to step S54. If it is determined that it is not the first time, the process proceeds to step S58.

[Step S54] The start / goal detection means 130 performs a passing frame calculation process.
This passing frame calculation process is the same process as calculating the virtual frame number when the boat 320 described above reaches the start line 330.

  As a result, the virtual frame number at which the boat 320 has reached the goal line can be calculated, and the measurement error is reduced.

[Step S55] The start / goal detection means 130 stores the virtual frame number calculated by the passing frame calculation process as a goal frame.
Then, the start / goal detecting means 130 increments a goal line counter representing the number of times the boat 320 has overlapped with the goal line, resets a goal timer representing the elapsed time from the time exceeding the goal line, and starts. .

  [Step S56] The time calculation means 140 reads the goal frame and the start frame, and subtracts the start frame from the goal frame, so that the moving image required until the boat 320 passes the start line 330 and then passes the goal line. The number of frames is calculated.

  [Step S57] The time calculation means 140 multiplies the number of frames of the moving image required from the time when the boat 320 passes the start line 330 to the time when the boat 320 passes the goal line by multiplying the time per frame by the boat 320. The time required from passing through the line 330 to passing through the goal line is calculated.

For example, if the video is NTFC (National
If the standard is the Television Standards Committee (30 FPS), the time per frame is 1/30 second, so multiply the number of frames by 1/30 to calculate the time. Will be able to.

  [Step S58] The start / goal detecting means 130 reads from the goal timer the elapsed time since the boat 320 exceeded the previous goal line, and a predetermined time or more (hereinafter, the threshold of this time is C) has elapsed. Determine whether or not.

If C or more has passed since the previous goal line was exceeded, the process proceeds to step S54, and if C has not been exceeded, the process ends.
The threshold value C is set to be shorter than the start interval when a plurality of boats 320 start, and to be set longer than the time until the pulling wave generated by the boat 320 disappears from the start line.

As with the start line, by setting this threshold value, the boat that is part of the pulling wave is different from the boat that is currently timing, and that has exceeded the goal line to measure the next time. Can be prevented from being erroneously detected as the head of the.
Note that although C is set for convenience because the threshold after the goal line is exceeded can be set arbitrarily, this does not preclude the use of the same threshold as B which is the threshold of the start line.

  According to the time measurement process as described above, the boat 320 is detected from the input moving image, and the boat 320 detects the boat 320 from the number of frames until the detected boat 320 passes the start line 330 and the frame that passes the goal line. Since the time is measured, the time can be measured if the boat 320 that is the time measurement target can input a moving image passing through the start line 330 and the goal line.

  Therefore, it is not necessary to use an expensive camera such as a special slit camera only for measuring time, and the measurement section can be changed any number of times by changing the angle of the camera 210 that captures a moving image. It is possible to measure the time without mounting a special device such as a sensor for measuring the time in the boat 320 that is rich in measurement.

It is a schematic diagram of the time measuring device concerning this embodiment. It is a flowchart which shows the time measurement process by the time measuring device of this Embodiment. It is a figure which shows the outline | summary of the process which takes the difference of the image by a boat detection means. It is a figure explaining the concept of a thinning process. It is a figure explaining the concept of a thickening process. It is a figure which shows the procedure which removes noise by thinning / thickening processing. 5 is a flowchart of a start detection process performed by a start / goal detection means 130. It is a figure which shows the outline | summary of the time correction process which a start / goal detection means performs. It is a flowchart which shows the procedure of the resource addition process by a management server. It is a flowchart which shows the goal detection process which a start / goal detection means and a time calculation means perform.

Explanation of symbols

DESCRIPTION OF SYMBOLS 100 Time measuring device 110 Boat detection means 120 Noise removal means 130 Start / goal detection means 140 Time calculation means 210 Camera 220 Movie storage device 300 Course 310 Imaging area 320 Boat

Claims (5)

In the time measurement device that measures the time from the start of the moving object that is the measurement target to the goal,
A moving object detecting means for detecting the moving object from a moving image obtained by imaging a start point and a goal point;
Passage detecting means for detecting that the moving body has passed the start point and the goal point;
In the image constituting the moving image, the time from the frame passing the start point to the frame passing the goal point is calculated, and the time of the moving body required from the start point to the goal point is calculated from the calculated number of frames. Time calculating means for calculating
A time measuring device comprising: The passage detection means includes
The distance between the moving object and the start point or the goal point in the first frame that passes through the start point or the goal point, and the moving object and the start point or the goal point in the frame immediately before the first frame 2. The time according to claim 1, further comprising virtual frame calculation means for calculating a virtual frame number that is a frame when the moving object is in contact with the start point or the goal point based on a ratio to the distance to the time point. Measuring device. The moving object detection means includes
The time measuring apparatus according to claim 1, further comprising a noise removing unit that removes noise of the image by performing thinning / thickening processing on the image. The noise removing means is
4. The time measuring apparatus according to claim 3, wherein noise of the image is removed by performing thinning / thickening processing after binarizing the image. In the time measurement method that measures the time from the start of the moving object to be measured to the goal,
A step of detecting a moving object from a moving image in which a moving object detecting unit images a start point and a goal point;
A passage detecting means for detecting that the moving object has passed through the start point and the goal point;
Time calculation means calculates from the frame that has passed through the start point to the frame that has passed through the goal point in the image constituting the moving image, and from the calculated number of frames to the goal point. Calculating the time of the moving object,
The time measuring method characterized by including.

Images