JP5549109B2 - Image sonar analysis system, image sonar analysis method, image sonar analysis program - Google Patents

Description

  The present invention relates to acoustic analysis processing in an acoustic sonar system, and more particularly to image processing in an image sonar.

Image sonar is used as a technique for detecting a hull such as a surface ship at sea and an underwater submarine.
The image sonar analysis system according to the image sonar is used to detect (recognize) a surface ship, a submarine, and the like by generating image information based on a reflection signal and analyzing the image information.

In this image sonar analysis system, a transmission signal (sound wave signal) is transmitted, a reflection signal is received, and image data (sonar image) is generated based on the reflection signal.
For example, as shown in FIG. 5, the surface ship is imaged as an echo, and a shadow area where no transmission signal reaches, that is, no reflection signal is formed behind it.

However, for the transmitted signal, the sound wave signal is diffusely reflected in the sea (underwater), and the sound (reflected signal) is also returned from the sea surface. It becomes unclear whether the sound is reflected from an object.
In addition, since there are waves on the sea surface, the waves on the sea surface cause the reflection signal to become fractaled, resulting in an unclear image. In addition, there is a disadvantage that it becomes difficult to distinguish between submarines and surface ships.

  Further, for example, when a dark place on an image generated by an image sonar is detected as a shadow (shadow) and an attempt is made to identify a hull such as a surface ship using this shadow as auxiliary information, FIG. As shown, there is a disadvantage in that a dark region on the image that is not a shadow is generated and erroneous detection occurs.

  As a related technique for this, a technique for tracking a moving object by detecting and separating a shadow edge of the moving object in a sequence of a plurality of digital images is disclosed (Patent Document 1).

JP-T 8-503316

In the related technique disclosed in Patent Document 1, a moving object (target object) is detected by detecting a shadow.
However, it is difficult to apply to on-surface or below-surface acoustic sonar, and as mentioned above, images generated by image sonar utilized for on-surface or below-surface objects are skilled. In this case, the target image cannot be identified accurately by visual inspection by a sonar engineer. Therefore, the target image cannot be accurately identified.

[Object of invention]
The present invention provides an image sonar analysis system, an image sonar analysis method, and an image sonar analysis program that can improve the disadvantages of the related art and can more accurately identify a target in an image generated by an image sonar. Its purpose is to provide.

  In order to achieve the above object, an image sonar analysis system according to the present invention is a sonar that transmits a sound signal for detecting a target located on or under water and captures a reflected sound of the transmitted sound signal. Unit, image data creating means for creating two-dimensional image data based on the reflected sound, and target object detecting means for detecting the position of the target object based on a shadow area detected in the two-dimensional image data. An analysis system for image sonar, the average reverberation calculating means for calculating the average reverberation level in the two-dimensional image data, and the reflected sound level of the reflected sound level based on the distance along the traveling direction of the transmitted sound signal. Reflected sound distribution analysis means for performing distribution analysis, and an area in which the ratio of the reflected sound level obtained by the distribution analysis is greater than a preset value is detected as an edge area. And an edge detection unit that includes the detected edge region as an end, and a shadow detection unit that detects a region where the reflected sound level is lower than the average reverberation level as the shadow region. .

  Further, the acoustic analysis device according to the present invention is configured to determine the position of the target in the two-dimensional image data created based on the reflected sound of the sound signal transmitted for detection of the target located on and below the water. An acoustic analysis device for detecting based on a shadow area detected in the two-dimensional image data, the average reverberation calculating means for calculating an average reverberation level in the two-dimensional image data, and the progress of the transmitted sound signal Reflected sound distribution analysis means for performing a distribution analysis of the reflected sound level with reference to a distance along the direction, and an edge where the ratio of the reflected sound level obtained by the distribution analysis is greater than a preset value Edge detection means for detecting as an area; and an area including the detected edge area as an end, and an area where the reflected sound level is lower than the average reverberation level It is characterized in that a shadow detecting means for detecting and.

Further, the image sonar method according to the present invention is based on a sonar section for transmitting a sound signal for detecting a target located on or under water and capturing a reflected sound of the sound transmission signal, and based on the reflected sound. An image sonar analyzing method for detecting a position of the target based on a shadow area detected in the two-dimensional image data, the image sonar analyzing method comprising: two-dimensional image data; calculating the average reverberation level in the inner, along said direction of travel of the sent sound signal subjected to distribution analysis of the reflected sound level, the reflected sound of the distance along the traveling direction of the delivery sounds signal as a reference Level distribution analysis is performed, and an area where a ratio of the reflected sound level drop obtained by the distribution analysis is larger than a preset value is detected as an edge area, and the two-dimensional image is detected. A region that includes the detected edge region as an end and has a reflected sound level lower than the average reverberation level as the shadow region, and determines the position of the target based on the detected shadow region. It is characterized by detecting.

The image sonar analysis program according to the present invention includes a sonar section for transmitting a sound signal for detecting a target located above or below the water surface and capturing a reflected sound of the sound transmission signal, and for the reflected sound. Image data creation means for creating two-dimensional image data based on the image sonar analysis program for detecting the position of the target based on a shadow area detected in the two-dimensional image data, A reflected sound distribution analysis function for performing a distribution analysis of the reflected sound level with reference to a distance along the traveling direction of the transmitted sound signal, and a ratio of the reflected sound level obtained by the distribution analysis being lowered are set in advance. An edge detection function for detecting an area larger than the detected value as an edge area; and the detected edge area in the two-dimensional image data as an end, and It is characterized in that to execute a shadow detection function sound level to detect a region lower than the average reverberation level calculated based on the two-dimensional image data as the shadow area, to the computer.

  Since the present invention is configured and functions as described above, according to this, an image sonar analysis system, an image sonar analysis method, and an image sonar that can more accurately identify a target in an image generated by an image sonar. An analysis program can be provided.

It is a schematic block diagram which shows one Embodiment in the analysis system for image sonar by this invention. It is a schematic block diagram which shows an example of the internal structure of the acoustic analysis part in the analysis system for image sonar disclosed in FIG. It is explanatory drawing which showed an example of the sound pressure level graph in the analysis system for image sonar disclosed in FIG. It is a flowchart which shows the whole operation | movement process step in the analysis system for image sonar disclosed in FIG. It is explanatory drawing which showed an example of the echo (surface ship) and the shadow (shadow) detected by the analysis system for image sonar disclosed in FIG. It is explanatory drawing which showed an example of the false detection of the shadow (shadow) detected by the analysis system for image sonars.

[Embodiment]
Next, the basic configuration content of the embodiment of the present invention will be described.

  The present embodiment is an image sonar analysis system as shown in FIG. 1, which is provided in advance in the sea (underwater), a speaker that emits sound waves toward the sea, and sound waves reflected by the emitted sound waves ( A transmission / reception unit (sonar unit) 1 including a capturing unit that captures (reflected sound), and generation of a sound wave emitted from the transmission / reception unit 1 and transmission / reception that digitizes the reflected sound captured by the transmission / reception unit 1 Unit 2, an acoustic analysis unit (target object detection means) 3 that analyzes a voice signal (hereinafter referred to as “acoustic data”) digitized by the transmission / reception unit 2 and detects a target, and a detected target The detection result output unit 4 that outputs and displays the position is provided.

  The acoustic analysis unit 3 is a computer equipped with a processor, and realizes the operation functions of the following means by performing execution processing based on a preset program.

  Further, as shown in FIG. 2, the acoustic analysis unit 3 includes acoustic data receiving means 31 that receives the digitized acoustic data from the transmission / reception unit 2 as input data, and the sound pressure reflected from the acoustic data (input data). A two-dimensional image data creation means (image data creation means) 32 for creating a pixelized two-dimensional image by converting a value into a luminance value, and a maximum sound pressure level and a minimum based on the generated two-dimensional image (pixel data) Sound pressure distribution analysis means 33 for detecting the azimuth direction having a large difference in sound pressure level, edge detection means 34 for detecting an edge from the detected azimuth direction, and analysis of the sound pressure level applied to the detected edge Sound pressure level analysis means 35 to perform, shadow area analysis means 36 to calculate the area of the shadow from the analyzed sound pressure level, and shadow of the target object from the calculated data And a shadow detection means 37 for detecting c).

Here, as shown in FIG. 5, when the acoustic signal transmitted from the transmission / reception unit 1 hits the surface ship, which is the target, as shown in FIG. Assume that image data is generated that indicates that a shadow area where sound does not reach behind the signal is formed while the signal is shown as an echo.
Thus, the target can be detected quickly, and the distance to the target (range), the height and size of the target can be detected.

The two-dimensional image data creating unit 32 creates two-dimensional image data based on the acoustic data received from the acoustic data receiving unit 31.
Here, two-dimensional image data generating means 32 of the present embodiment, for example, the cross fan beam system, it is assumed to generate two-dimensional image data.
First, the two-dimensional image data creation means 32 extracts three-dimensional data of horizontal azimuth (m), vertical azimuth (n), and distance (R) based on the sensor coordinate system from the received acoustic data.
The two-dimensional image data creating unit 32 creates two-dimensional image data by extracting the maximum reflected wave (sound pressure) level in the vertical direction (n), for example.

The sound pressure distribution analyzing unit 33 is based on the two-dimensional image data (pixel data) created by the two-dimensional image data creating unit 32, and the level of the reflected wave (sound pressure) level is high and low in the same azimuth direction. Detect azimuth direction with large difference.
Note that, at this stage, it may be set to detect by rough estimation within a specific value range without setting a specific threshold value.
Thereby, the processing load concerning detection can be reduced, and further, the time until detection can be shortened.

  The edge detection means 34 has a drop in the reflected wave level set in advance along the transmission direction among the points having a high reflected wave level detected on the two-dimensional image data by the sound pressure distribution analysis means 33. A point larger than a certain value is detected as an edge (edge detection function).

  The sound pressure level analysis means 35 analyzes the reflected wave level along the range direction in a specific azimuth direction, and generates a sound pressure level graph represented by the reflected wave level and distance as shown in FIG. .

Based on this sound pressure level graph, the shadow area analysis means 36 calculates the length in the distance direction of the portion where the reflected wave (sound pressure) level is lower than the sea surface reverberation average value (average reverberation) calculated in advance.
Further, the shadow area analysis means 36 calculates the length in the distance direction along the range direction in the adjacent azimuth direction, and calculates the area of the shadow region as the sum thereof.

The shadow area determination unit 37 (shadow recognition unit) includes a target size information storage unit that stores target size information indicating information regarding the size of the target.
Further, the shadow area determination unit 37 has an area determination function for determining whether or not the area calculated by the shadow area analysis unit 36 is a shadow of a target set in advance in the target size information storage unit. .

Thereby, the shadow recognizing unit 37 compares the calculated shadow area with the candidate shadow area stored in advance, and if the difference in area between shadows is within a preset error, the area is calculated. It is determined that the shadow is a shadow of the target.
Thereby, the accuracy of shadow detection can be further improved.

  The shadow recognition unit 37 may be provided with a target information input unit that inputs information related to the size of the target.

[Description of Operation of Embodiment]
Next, an outline of the operation of this embodiment will be described.
First, a distribution analysis of the reflected sound level is performed along the traveling direction of the transmitted sound signal, and the ratio of the reflected sound level that is obtained by this distribution analysis to the distance in the traveling direction is a preset value. Detecting a larger area as an edge area (edge detection process), including an edge area detected in the two-dimensional image data as an end, and an average in which the reflected sound level is calculated based on the data in the two-dimensional image data An area lower than the reverberation level is detected as a shadow area (shadow detection step), and the position of the target is detected based on the detected shadow area.
Here, regarding the edge detection step and the shadow detection step, the execution contents may be programmed and executed by a computer.

  Next, the operation of the image sonar analysis system according to the present embodiment will be described based on the flowchart of FIG.

  First, a sound wave is transmitted from the speaker of the transmission / reception unit 1 provided in the sea (underwater), and the capturing means of the transmission / reception unit 1 captures the reflected sound (reflected wave) (step S1). Next, the transmission / reception unit 2 performs signal processing for digitizing the captured reflected sound (step S2), and generates acoustic data (reflected wave signal).

Next, the acoustic data receiving means 31 of the acoustic analysis unit 3 receives this acoustic data (step S3).
Next, the two-dimensional image data creating unit 32 generates two-dimensional image data by the cross fan beam method based on the acoustic data received from the acoustic data receiving unit 31 (step S4).

Next, the sound pressure distribution analysis means (average reverberation calculation means) 33 is based on the two-dimensional image data (pixel data) generated by the two-dimensional image data creation means 32 and the point where the reflected wave (sound pressure) level is high and low. An azimuth direction having a large level difference between points is detected (step S5).
Here, the sound pressure distribution analysis means 33 may be set to perform processing for calculating the sea surface reverberation average value in the two-dimensional image data.

  Next, the edge detection unit 34 detects, as an edge, a portion of the echo detected by the sound pressure distribution analysis unit 33 whose reflected wave level drop (drop) is greater than a certain value according to the range direction (step). S6).

  Next, the sound pressure level analysis means (reflected sound distribution analysis means) 35 analyzes the reflected wave level along the detected range direction of the azimuth direction, and is represented by the reflected wave level and the distance. As shown in FIG. 4, a sound pressure level graph indicating the reflected wave level distribution is generated (step S7).

Here, the shadow area analysis means (shadow detection means) 36 calculates the length in the distance direction of the portion where the reflected wave (sound pressure) level is lower than the pre-calculated sea surface reverberation average value based on the sound pressure level graph. To do.
Further, the shadow area analysis means 36 calculates the length in the distance direction along the range direction in the adjacent azimuth direction, and calculates the area of the shadow region as the sum (step S8).

The shadow area determination unit 37 sets the area calculated by the shadow area analysis unit 36 in advance in the target size information storage unit based on the target size information indicating the information related to the size of the target stored in advance. It is determined whether or not the target is a shadow (step S9).
Finally, the detection result output unit 4 outputs two-dimensional image data in which the detected echo and shadow are highlighted as an analysis result.

As described above, in the present embodiment, the drop (edge) of the reflected wave level that appears along the range direction after the detected echo is detected, and the reflection level behind the echo (rear) is averaged based on this edge. A portion lower than the reverberation level is detected as a shadow.
As a result, it is possible to reduce false detection that erroneously detects a dark area on a sonar image that is not originally a shadow as a shadow, and to detect a target more accurately based on the detected shadow. Can do.

  Further, by comparing the size of the target registered in advance with the area of the detected shadow, it is possible to more accurately determine whether the calculated shadow is the shadow of the target. Thereby, the detection accuracy of the shadow applied to the target can be further improved.

  The present invention can be usefully applied to a system that detects and identifies an object and a moving object in a digital image by analyzing the digital image.

DESCRIPTION OF SYMBOLS 1 Transmission / reception part 2 Transmission / reception part 3 Acoustic analysis part 4 Detection result output part 31 Acoustic data reception means 32 Two-dimensional image data creation means 33 Sound pressure distribution analysis means 34 Edge detection means 35 Sound pressure level analysis means 36 Shadow area analysis means

Claims (6)

A sonar unit for transmitting a sound signal for detecting a target located above or below water and capturing a reflected sound of the transmitted sound signal, and image data generating means for generating two-dimensional image data based on the reflected sound And an image sonar analysis system comprising: target object detection means for detecting the position of the target object based on a shadow area detected in the two-dimensional image data,
Average reverberation calculating means for calculating an average reverberation level in the two-dimensional image data;
Reflected sound distribution analyzing means for analyzing the distribution of the reflected sound level based on the distance along the traveling direction of the transmitted sound signal, and a ratio of the reflected sound level obtained by the distribution analysis being lowered Edge detecting means for detecting an area larger than the determined value as an edge area;
An image sonar analysis system comprising: a shadow detection unit that includes the detected edge region as an end portion and detects the region where the reflected sound level is lower than the average reverberation level as the shadow region. In the analysis system for image sonar according to claim 1,
A target information storage unit that stores target size information indicating the size of the target set in advance;
A shadow size for comparing the size of the shadow area detected by the shadow detection means with the stored target size information to determine whether the detected shadow is a shadow of the target An image sonar analysis system comprising a determination means. In the analysis system for image sonar according to claim 2,
An image sonar analysis system comprising size information registration means for variably registering the target size information in the target information storage unit. The shadow in which the position of the target is detected in the two-dimensional image data in the two-dimensional image data created based on the reflected sound of the sound signal sent for detecting the target located above or below the water. An acoustic analysis device that detects based on a region,
Average reverberation calculating means for calculating an average reverberation level in the two-dimensional image data;
Reflected sound distribution analysis means for performing distribution analysis of the reflected sound level with reference to the distance along the traveling direction of the transmitted sound signal;
An edge detection means for detecting an area where the ratio of the reflected sound level obtained in this distribution analysis is greater than a preset value as an edge area;
An acoustic analysis apparatus, comprising: a shadow detection unit that includes the detected edge region as an end and detects a region in which the reflected sound level is lower than the average reverberation level as the shadow region. A sonar section for transmitting a sound signal for detecting a target located above or below the water and capturing the reflected sound of the sound transmission signal; and image data generating means for generating two-dimensional image data based on the reflected sound; An image sonar analysis method for detecting a position of the target based on a shadow area detected in the two-dimensional image data,
Calculating the average reverberation level within said two-dimensional image data,
Perform a distribution analysis of the reflected sound level along the traveling direction of the transmitted sound signal,
Perform a distribution analysis of the reflected sound level with reference to the distance along the traveling direction of the transmitted sound signal,
An area where the ratio of the reflected sound level obtained by the distribution analysis is greater than a preset value is detected as an edge area, and the detected edge area is included as an end in the two-dimensional image data; An image sonar analysis method, wherein an area where the reflected sound level is lower than the average reverberation level is detected as the shadow area, and the position of the target is detected based on the detected shadow area. A sonar section for transmitting a sound signal for detecting a target located above or below the water and capturing the reflected sound of the sound transmission signal; and image data generating means for generating two-dimensional image data based on the reflected sound; An image sonar analysis program for detecting the position of the target based on a shadow area detected in the two-dimensional image data,
A reflected sound distribution analysis function for performing a distribution analysis of the reflected sound level based on a distance along the traveling direction of the transmitted sound signal;
An edge detection function for detecting, as an edge region, an area where the ratio of the reflected sound level obtained by this distribution analysis is lower than a preset value;
Shadow detection that includes the detected edge region in the two-dimensional image data as an end, and detects a region where the reflected sound level is lower than the average reverberation level calculated based on the two-dimensional image data as the shadow region Function and
An image sonar analysis program characterized by causing a computer to execute.

Images