KR102687756B1 - Method and apparatus for generating target location information in water - Google Patents

Abstract

A method and device for generating target location information in water are disclosed. A method of generating target location information in water includes (a) acquiring target direction information for each scenario from an attitude measurement sensor attached to equipment worn by the target; (b) acquiring location information of a finger attached to equipment worn by the target at equal intervals using a receiver located in the water tank; (c) deriving speed information of the target based on the finger position information obtained at regular intervals; and (d) generating final target position information by correcting the target direction information using the speed information of the target.

Description

Method and apparatus for generating target location information in water}

The present invention relates to a method and device for generating target location information in water.

With the development of artificial intelligence, various types of artificial intelligence models are continuously emerging, but various types of learning data are required to learn these artificial intelligence models.

Due to the characteristics of the underwater environment, it is very difficult to obtain information about people, etc. underwater. Because of this, the number of underwater data is very small, and the acquisition method is also unsystematized.

The present invention is to provide a method and device for generating target location information in water.

In addition, the present invention is to provide a method and device for generating target location information in water for learning an artificial intelligence model that analyzes the location of an underwater target.

In addition, the present invention provides a method for generating target position information in water that can generate the absolute position of a target in water by matching and correcting data between sensors using data acquired by heterogeneous sensors with different synchronization periods, and the same. This is to provide a device.

According to one aspect of the present invention, a method for generating target location information in water is provided.

According to one embodiment of the present invention, (a) acquiring target direction information for each scenario from a posture measurement sensor attached to equipment worn by the target; (b) acquiring location information of a finger attached to equipment worn by the target at equal intervals using a receiver located in the water tank; (c) deriving speed information of the target based on the finger position information obtained at regular intervals; and (d) generating final target location information by correcting the target direction information using the speed information of the target.

In step (d), the final target position information may be generated by rotating and correcting the target direction information based on the vector direction of the target's speed information.

Before step (d), it further includes acquiring target image information from an image acquisition device fixed to a holding member located in the water tank, wherein the target image information and the final target location information are matched using a time stamp. It can be saved.

The target image information may be at least one of a stereo image and a sonar image for the target.

Steps (a) and (b) may be performed in synchronization with a time when the target image information can be acquired by the image acquisition device.

The image acquisition device, the posture measurement sensor, and the receiver each operate with different operation cycles, and can match and store target image information and target location information acquired at close times based on a timestamp.

According to another aspect of the present invention, an apparatus for generating target location information in water is provided.

According to one embodiment of the present invention, a posture measurement sensor is attached to equipment worn by the target and acquires target direction information for each scenario; A receiver located in the water tank and acquiring location information of a finger attached to the equipment worn by the target at equal intervals; And a target in water including a correction unit that derives target speed information based on the position information of the finger obtained at equal intervals, and generates target position information by correcting the target direction information based on the target speed information. A location information generating device may be provided.

The correction unit generates the target position information by rotating and correcting the target direction information based on the vector direction of the target speed information, but does not correct the target direction information when the target speed information is less than the reference speed. It may not be possible.

It is fixed to a holding member located in the water tank, and further includes an image acquisition device for acquiring target image information by photographing the target, wherein the target image information and the target location information can be matched and stored using a time stamp. .

By providing a method and apparatus for generating target location information in water according to an embodiment of the present invention, it is possible to generate target-related data in water for learning an artificial intelligence model that analyzes the location of an underwater target.

Additionally, the present invention can generate the absolute position of a target in water by matching and correcting data between sensors using data acquired by heterogeneous sensors with different synchronization periods.

1 is a flowchart showing a method for generating target location information in water according to an embodiment of the present invention.
Figure 2 is a diagram illustrating acquisition of underwater data according to an embodiment of the present invention.
Figure 3 is a diagram illustrating stereo image information according to an embodiment of the present invention.
Figure 4 is a diagram illustrating sonar image information according to an embodiment of the present invention.
Figure 5 is a diagram showing the results before and after noise removal according to an embodiment of the present invention.
Figure 6 is a diagram schematically showing the configuration of a target location information generating device according to an embodiment of the present invention.

As used herein, singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, terms such as “consists of” or “comprises” should not be construed as necessarily including all of the various components or steps described in the specification, and some of the components or steps may be included in the specification. It may not be included, or it should be interpreted as including additional components or steps. In addition, terms such as "... unit" and "module" used in the specification refer to a unit that processes at least one function or operation, which may be implemented as hardware or software, or as a combination of hardware and software. .

Hereinafter, embodiments of the present invention will be described in detail with reference to the attached drawings.

FIG. 1 is a flowchart showing a method for generating target location information underwater according to an embodiment of the present invention, FIG. 2 is a diagram illustrating acquisition of underwater data according to an embodiment of the present invention, and FIG. 3 is a Figure 4 is a diagram illustrating stereo image information according to an embodiment of the present invention, Figure 4 is a diagram illustrating sonar image information according to an embodiment of the present invention, and Figure 5 is a diagram illustrating noise removal according to an embodiment of the present invention. This diagram shows the before and after results.

For convenience of understanding and explanation, the data acquired by each sensor below should be understood as being acquired within the water tank. That is, as shown in FIG. 2, heterogeneous sensors are located within the water tank, and target location information according to the movement of the target within the water tank can be generated using the heterogeneous sensors.

In step 110, the target location information generating device 100 acquires target image information obtained by capturing a target from an image acquisition device.

As shown in Figure 2, the image acquisition device may be fixed to a support member located within the water tank. Additionally, there may be a plurality of image acquisition devices, but they may be heterogeneous image acquisition devices. For example, one of the first image acquisition device and the second image acquisition device may be a stereo camera, and the other may be a sonar sensor.

The first image acquisition device and the second image acquisition device may be obtained by stacking them. It is assumed that the angles of view of the first image acquisition device and the second image acquisition device are preset.

Figure 3 shows an example of an acquired stereo image, and Figure 4 illustrates sonar image information.

In step 115, the target location information generating device 100 acquires target direction information for each scenario from the posture measurement sensor. Here, the posture measurement sensor may be attached to equipment (eg, scuba equipment) worn by the target (eg, person). The attitude measurement sensor may be an Attitude Heading Reference System (AHRS) sensor. The target direction information acquired by the AHRS sensor is not an absolute direction, but (Roll, Pitch, and Yaw) can be obtained using the angular speed rotated after the AHRS sensor is operated. Therefore, Accordingly, when the AHRS sensor is turned off and then turned on, the (roll, pitch, yaw) values obtained at a different point in time are obtained, so different results may be obtained even if the same direction is measured.

Accordingly, when acquiring target direction information for each scenario, the target location information generating device 100 may acquire target direction information by turning off the AHRS sensor for each scenario and then operating it again.

In step 120, the target location information generating device 100 obtains target location information from the receiver.

As shown in Figure 2, the receiver is installed in a fixed position within the water tank, and a finger module can be mounted on the equipment equipped with the target. Because of this, the receiver can obtain finger location information as target location information from the finger module at regular intervals.

In step 125, the target location information generating device 100 derives target speed information using target location information acquired at regular intervals.

As finger location information (coordinate information) is acquired from the finger module at regular intervals, target location information can be used to derive information on the speed at which the target moves during regular intervals using the coordinate information acquired at regular intervals.

In step 130, the target location information generating device 100 corrects target direction information using target speed information and generates final target location information.

At this time, when the target speed information is less than or equal to the reference speed (for example, 0), the target location information generating device 100 does not correct the target direction information using the target speed information. When the target speed information exceeds the reference speed, the target location information generating apparatus 100 may correct the direction information of the target speed information so that the target direction information matches. That is, when the target speed information exceeds the reference speed, the target location information generating apparatus 100 may correct the target direction information by rotating it so that the direction information of the target speed information and the value of the target direction information match.

Next, the target location information generating device 100 may generate final target location information by matching the target location information and target direction information. At this time, the final target location information may be stored by matching the target image information.

Target image information, target location information, and target direction information may be acquired with different operation cycles. Therefore, the time when the target image information is acquired and the time when the target location information and target direction information are acquired may not be the same. Accordingly, the target location information generating device 100 uses the time stamp at which the target image information, target location information, and target direction information were acquired to obtain information (e.g., target image information, target location information, target direction information) acquired at a nearby time. direction information) can be matched and stored.

Additionally, the target location information generating device 100 may first perform a noise removal process before matching and storing the target image information, target location information, and target direction information. For example, the target location information generating device 100 may apply the Gaussian Rolling algorithm to remove noise that suddenly causes values to jump. Figure 5 shows an example of data before and after noise removal.

In addition, according to an embodiment of the present invention, the target location information generating device 100 is synchronized at a time when target image information for each scenario can be acquired to use it as learning data for artificial intelligence learning, and generates finger location information (target Location information) and target direction information can also be obtained.

Figure 6 is a diagram schematically showing the configuration of a target location information generating device according to an embodiment of the present invention.

Referring to FIG. 6, the target location information generating device 100 according to an embodiment of the present invention includes an image acquisition device 610, a posture measurement sensor 620, a receiver 630, a correction unit 640, and a memory ( 650) and a processor 660.

The image acquisition device 610 is fixedly installed on a support member in the water tank and acquires images (i.e., target image information) for the target for each scenario. As described above, the image acquisition device 610 may acquire a plurality of different types of target image information.

That is, one of the first image acquisition device and the second image acquisition device may acquire stereo image information, and the other may acquire sonar image information.

The posture measurement sensor 620 is attached to the equipment worn by the target and acquires target direction information for each scenario.

The posture measurement sensor 620 is turned off for each scenario and then turned back on to obtain target direction information.

The receiver 630 is installed at a specific location in the water tank and is a means for receiving location information of the finger module attached to the equipment worn by the target at equal intervals. At this time, since the finger module is attached to equipment that has a clear target, even if it has a certain error, the error is negligible and can be used as location information of the target in the water tank.

The correction unit 640 is a means for deriving target speed information based on the position information of the finger module acquired at regular intervals and correcting the target direction information based on the target speed information to generate target position information.

The memory 650 stores various commands for performing a method for generating target location information in water according to an embodiment of the present invention.

The processor 660 includes components of the target location information generating device 100 according to an embodiment of the present invention (e.g., an image acquisition device 610, an attitude measurement sensor 620, a receiver 630, and a It is a means to control the government (640), memory (650), etc.).

In addition, the processor 660 uses the image acquisition device 610, the posture measurement sensor 620, and the receiver 630 because the image acquisition device 610, the posture measurement sensor 620, and the receiver 630 do not operate in the same synchronous cycle. Using the time stamp of the data acquired in step 630), data acquired at close times can be controlled to be mapped and stored.

Additionally, the processor 660 may match and store target image information (e.g., stereo image information, sonar image information) and target location information so that the target location information can be ultimately used for learning an artificial intelligence model. At this time, the processor 660 may control the posture measurement sensor 620 and the receiver 630 so that target location information is acquired in synchronization with the time of acquisition of target image information.

Devices and methods according to embodiments of the present invention may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. Computer-readable media may include program instructions, data files, data structures, etc., singly or in combination. Program instructions recorded on a computer-readable medium may be specially designed and constructed for the present invention or may be known and usable by those skilled in the computer software field. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -Includes magneto-optical media and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, flash memory, etc. Examples of program instructions include machine language code, such as that produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.

The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

So far, the present invention has been examined focusing on its embodiments. A person skilled in the art to which the present invention pertains will understand that the present invention may be implemented in a modified form without departing from the essential characteristics of the present invention. Therefore, the disclosed embodiments should be considered from an illustrative rather than a restrictive perspective. The scope of the present invention is indicated in the claims rather than the foregoing description, and all differences within the equivalent scope should be construed as being included in the present invention.

Claims (10)

In a method of generating target location information underwater for learning an artificial intelligence model that analyzes the underwater target location,
(a) acquiring target direction information for each scenario from a posture measurement sensor attached to equipment worn by the target;
(b) acquiring location information of a finger attached to equipment worn by the target at equal intervals using a receiver located in the water tank;
(c) deriving speed information of the target based on the finger position information obtained at regular intervals; and
(d) generating final target position information by correcting the target direction information using the speed information of the target,
The posture measurement sensor is,
As it is turned off and then restarted for each scenario, it is possible to obtain different target direction information even if the same direction is measured,
In step (d),
A method of generating target position information in water, wherein the final target position information is generated by rotating and correcting the target direction information to match the vector direction of the target speed information based on the vector direction of the target speed information. delete According to claim 1,
Before step (d) above,
It further includes the step of acquiring target image information from an image acquisition device fixed to a holding member located in the water tank,
A method of generating target location information in water, characterized in that the target image information and the final target location information are matched and stored using a time stamp. According to clause 3,
The target image information is at least one of a stereo image and a sonar image for the target. According to clause 3,
A method of generating target location information in water, characterized in that steps (a) and (b) are performed in synchronization at a time when the target image information can be acquired by the image acquisition device. According to clause 5,
The image acquisition device, the posture measurement sensor, and the receiver each operate with different operation cycles, but are characterized in that the target image information and the target location information acquired at close times are matched and stored based on a timestamp. Method for generating target location information underwater. A computer-readable recording medium recording program code for performing the method according to any one of claims 1 and 3 to 6. In an underwater target location information generating device for learning an artificial intelligence model that analyzes the underwater target location,
A posture measurement sensor that is attached to the equipment worn by the target and acquires target direction information for each scenario;
A receiver located in the water tank and acquiring location information of a finger attached to the equipment worn by the target at equal intervals; and
A correction unit that derives target speed information based on the position information of the finger obtained at equal intervals, and generates target position information by correcting the target direction information based on the target speed information,
The posture measurement sensor is,
As it is turned off and then restarted for each scenario, it is possible to obtain different target direction information even if the same direction is measured,
The correction unit,
An underwater target position information generating device that generates the target position information by rotating and correcting the target direction information to match the vector direction of the target speed information based on the vector direction of the target speed information. According to clause 8,
The correction unit,
An apparatus for generating target position information in water, characterized in that the target direction information is not corrected when the target speed information is less than the reference speed. According to clause 8,
It is fixed to a holding member located in the water tank, and further includes an image acquisition device for acquiring target image information by photographing the target,
An underwater target location information generating device, wherein the target image information and the target location information are matched and stored using a time stamp.