An underwater structure ultrasonic detection method and equipment integrating visual and acoustic technology

Abstract

The underwater structure ultrasonic detection method and equipment integrating visual and acoustic technology disclosed in the present invention include: establishing a local coordinate system with a ship as an underwater positioning base point; an underwater remote control robot uses a multi-beam imaging sonar to judge the surface state of the underwater structure in a blind spot and hydroacoustic detection damage, if there is a blind spot, the blind spot coordinate Q1 is recorded, if there is damage, a large-scale damage is output, otherwise an underwater pan-tilt camera is used to judge the blind spot and visual detection damage; if there is a blind spot, the blind spot coordinate Q2 is recorded, if there is damage, the underwater structure has medium-scale damage, otherwise a longitudinal wave probe is used in the blind spot coordinate Q1 and / or Q2 area to carry out ultrasonic detection, if it is judged that there is damage, the underwater structure has small-scale damage, and the detection is terminated. The present invention uses hydroacoustic signals, visual signals, and ultrasonic signals to respectively characterize large-scale, medium-scale, and small-scale damages, so as to achieve a balance between detection efficiency and accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of underwater structure detection, and in particular to an underwater structure ultrasonic detection method and equipment integrating visual and acoustic technologies. Background Art

[0002] There are underwater engineering structures made of metal or concrete in dams, bridge piers, gates, ships, and pipelines. Due to operational requirements, these structures are difficult to separate from the water once they are in service, making it impossible to conduct inspections in a waterless environment and evaluate the safety of the structures. Therefore, it is necessary to propose underwater structure inspection methods.

[0003] The existing technology mainly adopts the following schemes for underwater structure detection:

[0004] 1) Use core sampling, cutting sampling and other methods to carry out destructive sampling, and send the samples to the laboratory to test their material properties and mechanical parameters. The disadvantage is that it is destructive to the original structure and needs to be repaired, which endangers the safety of the original structure.

[0005] 2) Multi-beam sonar detection is used to detect the appearance of underwater structures through the rebound of water sound on the surface of the structure. Since the instrument is equipped with multiple pairs of transmitters and receivers, it can simultaneously transmit, receive and resolve multiple groups of beams, thereby improving the detection efficiency and achieving large-scale rapid detection at the km level. However, its disadvantages are: low resolution, and only obvious damage to the structure (usually m level) can be detected; it is greatly disturbed by turbid water and high-speed water flow; it cannot detect damage when the surface of the structure is covered with aquatic plants and silt; it can only detect surface damage of the structure, and cannot detect material degradation, internal voids, steel bar debonding, soil debonding and other damage.

[0006] 3) Underwater video detection: Through searchlights and underwater cameras, the apparent cracks of the structure are detected. The detection speed and range are lower than those of multi-beam sonar, but higher than those of ultrasonic methods. The detection accuracy is higher than that of multi-beam sonar, but lower than that of ultrasonic methods. Underwater video detection can identify cracks and damage on the exposed surface of the structure, but its disadvantages are: when the water is turbid, the light is low at depth, or the surface of the structure is covered by aquatic plants and silt, it is unable to detect damage; it can only detect surface damage of the structure, but cannot detect material degradation, internal voids, steel bar debonding, soil debonding and other damage. Summary of the invention

[0007] Purpose of the invention: In view of the fact that the existing technology uses a single signal source to detect underwater structures, and the detection efficiency and accuracy cannot be balanced, the present invention provides an underwater structure ultrasonic detection method and equipment that integrates visual and acoustic technology. By integrating hydroacoustic, video, and ultrasonic signals, multi-scale damage detection from meters to decimeters and then to centimeters is achieved. When large-scale damage is found during the large-scale damage detection process, the detection is stopped. Therefore, on the basis of taking into account the detection efficiency, the detection accuracy can be gradually refined, and structural damage of different scales can be quickly detected.

[0008] Technical solution: The underwater structure ultrasonic detection method and equipment integrating visual and acoustic technologies described in the present invention include the following steps:

[0009] S1: deploy an underwater remote-controlled robot equipped with an underwater positioning beacon in the water, and moor a ship equipped with a GPS positioning system on the water surface. The GPS positioning system obtains the signal sent by the underwater positioning beacon through the receiver and calculates the position, and establishes a local coordinate system (x, y, z) with the ship as the underwater positioning base point;

[0010] S2: The underwater remote-controlled robot uses a multi-beam imaging sonar to scan the surface state of the underwater structure, generate a pixel map P1 (x, y, z) of the surface morphology of the underwater structure, and process the pixel map P1 (x, y, z) to obtain an underwater acoustic detection damage map P2 (x, y, z); based on P2 (x, y, z), damage judgment and blind area judgment are performed respectively. If it is judged that there is a blind area, the blind area coordinate Q1 is recorded and S4 is executed; if it is judged that there is damage, it is output that there is large-scale damage to the underwater structure and the detection is ended. If it is judged that there is no damage, S3 is executed;

[0011] S3: The underwater remote control robot uses an underwater pan-tilt camera to shoot a video of the underwater structure, generates a pixel map P3(x, y, z) through the grayscale value of the video, and processes the pixel map P3(x, y, z) to obtain a visual detection damage map P4(x, y, z); based on P4(x, y, z), damage judgment and blind spot judgment are performed respectively. If it is judged that there is a blind spot, the blind spot coordinate Q2 is recorded and S4 is executed; if it is judged that there is damage, the output is that the underwater structure has medium-scale damage, and the detection ends;

[0012] S4: In the blind area coordinates Q1 and / or Q2, grinding is performed by a mechanical grinding wheel carried by the underwater remote control robot to form an exposed structural surface, and then S5 is executed;

[0013] S5: The underwater remote-controlled robot uses a longitudinal wave probe to vertically abut against the surface of the structure to carry out ultrasonic testing, wherein the probe at the center of the structure surface is used to excite the pulse wave, and the probes around the structure surface are used to receive the pulse wave. The received pulse wave is imaged to obtain a pixel map P5 (x, y, z), and the pixel map P5 (x, y, z) is processed to obtain an ultrasonic detection damage map P6 (x, y, z). Damage judgment is performed based on P6 (x, y, z). If it is judged that there is damage, it is output that there is small-scale damage to the underwater structure, and the detection is terminated, and S6 is executed;

[0014] S6: Image the structural surface at the coordinate (x, y, z) in space and superimpose the pixels P2, P4, and P6 at the coordinate (x, y, z) to generate a damage cloud map P7. The pixel spectrum of the damage cloud map P7 = P2+P4+P6.

[0015] To further improve the above technical solution, the underwater remote-controlled robot adopts blue ROV, the underwater positioning beacon adopts USBL XL150, the ship feeds signals between the underwater positioning beacon USBL X150 and the ultra-short baseline receiver UGPS100 by dragging, and the GPS positioning system calculates the relative position of the underwater remote-controlled robot blue ROV and the ship.

[0016] Furthermore, in S2, a normalization operation is performed on P1: P2(x,y,z)=P1(x,y,z) / max(P1(x,y,z)), where max(P1(x,y,z)) refers to the maximum value of all pixels in the P1 matrix. After normalization, the value of P2 is between 0 and 1. P2(x,y,z) represents the intensity of water acoustic reflection at the coordinates (x,y,z). The closer the value is to 0, the lower the water acoustic intensity and the greater the degree of damage to the structural surface.

[0017] Furthermore, P3 is normalized: P4(x,y,z)=P3(x,y,z) / max(P3(x,y,z)), where max(P3(x,y,z)) refers to the maximum value of all pixels in the P3 matrix. After normalization, the value of P4 is between 0 and 1. P4(x,y,z) represents the optical reflection intensity at the coordinates (x,y,z). The closer the value is to 0, the lower the optical intensity and the greater the degree of damage to the structural surface.

[0018] Furthermore, in the S5, 9 longitudinal wave probes are used vertically against the surface of the structure, one of which is located at the center of the structure surface for exciting pulse waves, and the other eight are evenly arranged in a circle with a radius of 90 mm on the surface of the structure; the excitation pulse is generated by a Puyuan DG822 signal generator, the generated signal is amplified by a PINTEK HA205 amplifier, connected to the longitudinal wave probe located in the center for excitation, the received signal is amplified by PXPA3, and then received by the 8 surrounding longitudinal wave probes, and the received signal is stored by a Puyuan DS1102Z-E oscilloscope.

[0019] Furthermore, a delayed superposition algorithm is used to image the 8 received echo waveforms to obtain a pixel map P5(x,y,z) based on the grayscale value of the ultrasound. P5 is normalized: P6(x,y,z)=1-P5(x,y,z) / max(P5(x,y,z)). After normalization, the value of P6 is between 0 and 1. P6(x,y,z) represents the ultrasonic reflection intensity at the coordinates (x,y,z). The closer the value is to 0, the lower the intensity of the ultrasonic echo and the greater the degree of damage to the structural surface.

[0020] Furthermore, the underwater structure types are metal and concrete; the damage types are cracks, cavities, and voids in fill.

[0021] The underwater structure ultrasonic detection equipment integrating visual and acoustic technology used to realize the above-mentioned underwater structure ultrasonic detection includes an underwater remote-controlled robot equipped with an underwater positioning beacon and a ship equipped with a GPS positioning system. The ship establishes a local coordinate system (x, y, z) centered on the ship by towing an ultra-short baseline receiver to feed signals between the underwater positioning beacon; the underwater remote-controlled robot is equipped with a multi-beam imaging sonar, an underwater pan-tilt camera, a mechanical grinding wheel, and a longitudinal wave probe group; the multi-beam imaging sonar is used to scan the apparent state of the underwater structure. A pixel map P1 (x, y, z) of the surface morphology of the structure is generated, and P1 is normalized to obtain P2 (x, y, z); the underwater gimbal camera shoots a video of the underwater structure to generate a pixel map P3 (x, y, z), and P3 is normalized to obtain P4 (x, y, z); the mechanical grinding wheel is used to grind the surface of the structure, and the longitudinal wave probe performs laser detection on the polished surface of the structure to obtain a pixel map P5 (x, y, z), and P5 is normalized to obtain P6 (x, y, z).

[0022] Beneficial effects: Compared with the prior art, the advantages of the present invention are: the present invention can quickly detect damaged structures through sonar, and after detecting damage on a large scale, the detection can be terminated without continuing visual and ultrasonic detection, thereby improving the detection efficiency; visual detection can provide smaller-scale damage judgment based on sonar detection. For the situation where sonar and visual detection are affected by aquatic plants and silt, ultrasonic detection is used to cover the blind spots of sonar and visual detection and can detect the internal cavities of the structure, so that on the basis of taking into account the detection efficiency, the detection accuracy can be gradually refined, and structural damage of different scales can be quickly detected.

[0023] The present invention uses hydroacoustic signals to characterize large-scale damage, uses visual signals to characterize medium-scale damage, and uses ultrasonic signals to characterize small-scale damage. When damage that endangers the safety of the structure is detected at a large scale, there is no need to carry out medium-scale damage detection. When damage that endangers the safety of the structure is detected at a medium scale, there is no need to carry out small-scale damage detection. In the process of hydroacoustic signal and visual signal detection, only surface imaging is needed, and pixel values ​​do not need to be superimposed. If the pixel value of a certain point is less than 20% of the surrounding pixel values, it is considered that there is damage at the point, thereby improving the detection efficiency. On this basis, the signals of the three physical properties are respectively converted into pixel values, and a conversion relationship between hydroacoustic, visual, and ultrasonic signals and pixel values ​​in three-dimensional space is established, and they are integrated into the same damage cloud map. At different positions, pixel values ​​generated by different signals are used to characterize the degree of structural damage. For areas where only hydroacoustic detection is carried out, the resolution of the damage cloud map is determined by the resolution of the sonar equipment, which is usually at the meter level. For areas where video detection is carried out, the resolution is usually at the cm level. For areas where ultrasonic detection is carried out, the resolution is usually at the mm level, achieving a balance between detection efficiency and accuracy, and providing key technologies for non-destructive detection and safety assessment of underwater structures. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 is a flow chart of the method of the present invention;

[0025] Figure 2 It is a schematic diagram of the underwater ultra-short line positioning system of the present invention;

[0026] Figure 3 It is a pixel map based on multi-beam sonar;

[0027] Figure 4 It is based on the pixel map of the captured video;

[0028] Figure 5 It is the arrangement of the ultrasonic probe;

[0029] Figure 6 It is a pixel map based on ultrasound.

[0030] Figure 2In the middle, 1 is the blue ROV2 underwater remote control robot, 2 is the USBL X150 underwater positioning beacon, and 3 is a surface moored vessel. DETAILED DESCRIPTION

[0031] The technical solution of the present invention is described in detail below with reference to the accompanying drawings, but the protection scope of the present invention is not limited to the embodiments.

[0032] like Figure 1 The underwater structure ultrasonic detection method that integrates visual and acoustic technology shown in the figure realizes multi-scale damage detection from meters to decimeters and then to centimeters through the integration of hydroacoustic, visual and ultrasonic signals. When large-scale damage is found during large-scale damage detection, the detection is stopped. Therefore, on the basis of taking into account the detection efficiency, the detection accuracy can be gradually refined to quickly identify structural damage of different scales. In addition, when there are aquatic plants or silt covering the surface of the underwater structure, the hydroacoustic and video methods cannot effectively detect the damage. The ultrasonic method can be used to cover the detection blind spots of the above two methods. By converting hydroacoustic, video and ultrasonic signals into pixel maps and superimposing them in three-dimensional space, the damage map of the underwater structure is obtained, which provides key technologies for non-destructive detection and safety assessment of underwater structures.

[0033] Step 1: Build an underwater positioning system

[0034] like Figure 2 As shown in the figure, a blue ROV underwater remote-controlled robot is placed in the water, and the underwater remote-controlled robot is equipped with a USBLX150 underwater positioning beacon; a ship moored on the surface is equipped with a GPS positioning system as the base point for underwater positioning, and the ship tows an ultra-short baseline receiver UGPS100, which feeds signals with the USBL X150 underwater positioning beacon to calculate the relative position of the blue ROV and the ship, thereby establishing a local coordinate system (x, y, z) centered on the ship; the local coordinates are within a range of about 100 m square, and the positioning accuracy reaches 1 cm in an ideal state, that is, (x, y, z) changes in centimeters.

[0035] Step 2: Blue ROV is equipped with M750D multi-beam imaging sonar to scan the surface morphology of the underwater structure and generate a pixel map P1 (x, y, z) of the surface morphology of the structure, such as Figure 3 As shown, normalization operation is performed on P1:

[0036] P2(x,y,z)=P1(x,y,z) / max(P1(x,y,z)),

[0037] max(P1(x,y,z)) represents the maximum value of all pixels in the P1 matrix. At this time, the value of P2 is between 0 and 1, representing the water acoustic reflection intensity at the (x,y,z) coordinate. The closer the value is to 0, the lower the water acoustic intensity. Since cracks have low reflection intensity to water acoustics, low pixel values ​​represent cracks.

[0038] When the structure in the P2 diagram shows obvious damage such as fracture or overturning, it is considered that large-scale damage has occurred in the structure and the detection process is stopped; when the P2 diagram is covered by aquatic plants, silt or other obstructions, the coordinates Q1 (x, y, z) of the obscured area are recorded.

[0039] Step 3: Use the blue ROV equipped with a 1080P underwater gimbal camera to carry out visual inspection and generate a pixel map P3 (x, y, z) based on the grayscale value of the captured video, as shown in Figure 4 As shown, the normalization operation is performed on P3:

[0040] P4(x,y,z)=P3(x,y,z) / max(P3(x,y,z))

[0041] max(P3(x,y,z)) represents the maximum value of all pixels in the P3 matrix. At this time, the value of P4 is between 0 and 1, representing the optical reflection intensity at the (x,y,z) coordinate. The closer the value is to 0, the lower the optical intensity. Since cracks have low optical reflection intensity, low pixel values ​​represent cracks.

[0042] When the structure in the P4 diagram shows obvious damage such as fracture or overturning, it is considered that medium-scale damage has occurred in the structure and the detection process is stopped; when the P4 diagram is covered by aquatic plants, silt or other obstructions, the coordinates Q2 (x, y, z) of the obscured area are recorded.

[0043] Step 4: Conduct ultrasonic testing in the Q1 and Q2 areas. First, the mechanical grinding wheel carried by the blue ROV is used to grind the exposed structural surface; then the DYW-28-G 28 kHz longitudinal wave probe is placed vertically against the structural surface. Figure 5 9 probes are pressed separately, and the radius of the outer ring is 90 mm. The probe in the center excites a 28 kHz rectangular pulse wave, and the probes in the periphery receive the echo of the rectangular pulse wave; the excitation pulse is generated by a Puyuan DG822 signal generator, and the generated signal is amplified by a PINTEK HA205 amplifier and connected to the probe in the center for excitation. The received signal is amplified by a PXPA3 and then received by the surrounding probes. The received signal is stored by a Puyuan DS1102Z-E oscilloscope.

[0044] Step 5: Use the delay superposition algorithm to image the 8 received echo waveforms and obtain Figure 6The pixel map P5 (x, y, z) is generated based on the gray value of ultrasound, and the normalization operation is performed on P5:

[0045] P6(x,y,z)=1-P5(x,y,z) / max(P5(x,y,z))

[0046] At this time, the value of P6 is between 0 and 1, representing the ultrasonic reflection intensity at the (x, y, z) coordinate. The closer the value is to 0, the lower the intensity of the ultrasonic echo. Since cracks reflect ultrasonic waves, low pixel values ​​represent cracks.

[0047] Step 6: Image the surface of the structure in space according to the (x, y, z) coordinates. The pixel map P = P2 + P4 + P6. The closer the pixel value is to 0, the higher the degree of damage. The pixel map P has a high resolution in the Q1 and Q2 areas, reaching the centimeter level. In areas outside Q1 and Q2, the resolution is at the decimeter or meter level.

[0048] Since the range of a single underwater ultrasonic inspection is limited, it takes a long time to conduct full coverage inspection of large underwater structures. Therefore, the damaged structure is first quickly detected by sonar. After the damage is detected on a large scale, the inspection is terminated immediately, which improves the inspection efficiency. Then, visual hydroacoustic technology is used for more detailed inspection. At the same time, for the situation where sonar and visual inspections are affected by aquatic plants and silt, the area where ultrasonic inspection is required is determined, thereby improving the efficiency of ultrasonic inspection. Ultrasonic inspection is used to cover the blind spots of visual inspection, and the conversion relationship between hydroacoustic, visual, and ultrasonic signals and pixel values ​​in three-dimensional space is established. By integrating the inspection results of vision, hydroacoustic, and ultrasonic into the same pixel map, damage imaging of underwater structures is achieved, and pixel values ​​generated by different signals at different locations are used to characterize the degree of structural damage. The final damage pixel map has variable resolution, providing key technologies for non-destructive inspection and safety assessment of underwater structures.

[0049] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes in form and details may be made without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A method for ultrasonic detection of underwater structures integrating visual and acoustic technologies, characterized in that: The steps include: S1: deploy an underwater remote-controlled robot equipped with an underwater positioning beacon in the water, and moor a ship equipped with a GPS positioning system on the water surface. The GPS positioning system obtains the signal sent by the underwater positioning beacon through the receiver and calculates the position, and establishes a local coordinate system (x, y, z) with the ship as the underwater positioning base point; S2: The underwater remote-controlled robot uses a multi-beam imaging sonar to scan the surface state of the underwater structure, generate a pixel map P1 (x, y, z) of the surface morphology of the underwater structure, and process the pixel map P1 (x, y, z) to obtain an underwater acoustic detection damage map P2 (x, y, z); based on P2 (x, y, z), damage judgment and blind area judgment are performed respectively. If it is judged that there is a blind area, the blind area coordinate Q1 is recorded and S4 is executed; if it is judged that there is damage, it is output that there is large-scale damage to the underwater structure and the detection is ended. If it is judged that there is no damage, S3 is executed; S3: The underwater remote control robot uses an underwater pan-tilt camera to shoot a video of the underwater structure, generates a pixel map P3(x, y, z) through the grayscale value of the video, and processes the pixel map P3(x, y, z) to obtain a visual detection damage map P4(x, y, z); based on P4(x, y, z), damage judgment and blind spot judgment are performed respectively. If it is judged that there is a blind spot, the blind spot coordinate Q2 is recorded and S4 is executed; if it is judged that there is damage, the output is that the underwater structure has medium-scale damage, and the detection ends; S4: In the blind area coordinates Q1 and / or Q2, grinding is performed by a mechanical grinding wheel carried by the underwater remote control robot to form an exposed structural surface, and then S5 is executed; S5: The underwater remote-controlled robot uses a longitudinal wave probe to vertically abut against the surface of the structure to carry out ultrasonic testing, wherein the probe at the center of the structure surface is used to excite the pulse wave, and the probes around the structure surface are used to receive the pulse wave. The received pulse wave is imaged to obtain a pixel map P5 (x, y, z), and the pixel map P5 (x, y, z) is processed to obtain an ultrasonic detection damage map P6 (x, y, z). Damage judgment is performed based on P6 (x, y, z). If it is judged that there is damage, it is output that there is small-scale damage to the underwater structure, and the detection is terminated, and S6 is executed; S6: Image the structural surface at the coordinate (x, y, z) in space and superimpose the pixels P2, P4, and P6 at the coordinate (x, y, z) to generate a damage cloud map P7. The pixel spectrum of the damage cloud map P7 = P2+P4+P6.

2. The underwater structure ultrasonic detection method integrating visual and acoustic technologies according to claim 1 is characterized in that: The underwater remote control robot adopts blue ROV, the underwater positioning beacon adopts USBL XL150, the ship feeds signals to each other with the underwater positioning beacon USBL X150 by dragging the ultra-short baseline receiver UGPS100, and the GPS positioning system calculates the relative position of the underwater remote control robot blue ROV and the ship.

3. The underwater structure ultrasonic detection method integrating visual and acoustic technologies according to claim 1 is characterized in that: In S2, P1 is normalized: P2(x,y,z)=P1(x,y,z) / max(P1(x,y,z)), where max(P1(x,y,z)) refers to the maximum value of all pixels in the P1 matrix. After normalization, the value of P2 is between 0 and 1. P2(x,y,z) represents the water acoustic reflection intensity at the coordinates (x,y,z). The closer the value is to 0, the lower the water acoustic intensity and the greater the degree of damage to the structure surface.

4. The underwater structure ultrasonic detection method integrating visual and acoustic technologies according to claim 1 is characterized in that: Normalize P3: P4(x,y,z)=P3(x,y,z) / max(P3(x,y,z)), max(P3(x,y,z)) refers to the maximum value of all pixels in the P3 matrix. After normalization, the value of P4 is between 0 and 1. P4(x,y,z) represents the optical reflection intensity at the coordinates (x,y,z). The closer the value is to 0, the lower the optical intensity and the greater the degree of damage to the structure surface.

5. The underwater structure ultrasonic detection method integrating visual and acoustic technologies according to claim 1 is characterized in that: In the S5, 9 longitudinal wave probes are vertically pressed against the surface of the structure, one of which is located at the center of the structure surface for exciting pulse waves, and the other eight are evenly arranged in a circle with a radius of 90 mm on the structure surface; the excitation pulse is generated by a Puyuan DG822 signal generator, the generated signal is amplified by a PINTEK HA205 amplifier, connected to the longitudinal wave probe located in the center for excitation, the received signal is amplified by PXPA3, and then received by the 8 surrounding longitudinal wave probes, and the received signal is stored by a Puyuan DS1102Z-E oscilloscope.

6. The underwater structure ultrasonic detection method integrating visual and acoustic technologies according to claim 5 is characterized in that: The delayed superposition algorithm is used to image the 8 received echo waveforms to obtain a pixel map P5(x,y,z) based on the grayscale value of the ultrasound. P5 is normalized: P6(x,y,z)=1-P5(x,y,z) / max(P5(x,y,z)). After normalization, the value of P6 is between 0 and 1. P6(x,y,z) represents the ultrasonic reflection intensity at the coordinates (x,y,z). The closer the value is to 0, the lower the intensity of the ultrasonic echo and the greater the degree of damage to the structure surface.

7. The underwater structure ultrasonic detection method integrating visual and acoustic technologies according to claim 1 is characterized in that: The underwater structure types are metal and concrete; the damage types are cracks, cavities, and voids in the fill.

8. An underwater structure ultrasonic detection equipment for implementing the underwater structure ultrasonic detection method integrating visual and acoustic technology as claimed in claim 1, characterized in that: It includes an underwater remote-controlled robot equipped with an underwater positioning beacon and a ship equipped with a GPS positioning system. The ship feeds signals to the underwater positioning beacon by dragging an ultra-short baseline receiver. The GPS positioning system obtains the signal sent by the underwater positioning beacon through the receiver and calculates the position, and establishes a local coordinate system (x, y, z) with the ship as the underwater positioning base point; The underwater remote-controlled robot is equipped with a multi-beam imaging sonar, an underwater pan-tilt camera, a mechanical grinding wheel, and a longitudinal wave probe group; The multi-beam imaging sonar is used to scan the surface state of the underwater structure to generate a pixel map P1 (x, y, z) of the surface morphology of the structure, perform a normalization operation on P1 to obtain P2 (x, y, z), and perform damage judgment and blind area judgment based on P2 (x, y, z): if it is judged that there is a blind area, the blind area coordinate Q1 is recorded; if it is judged that there is damage, it is output that there is large-scale damage to the underwater structure; if it is judged that there is no damage, the underwater gimbal camera shoots a video of the underwater structure to generate a pixel map P3 (x, y, z), and perform a normalization operation on P3 to obtain P4 (x, y, z); based on P4 (x, y, z), damage judgment and blind area judgment are performed: if it is judged that there is a blind area, the blind area coordinate Q2 is recorded; if it is judged that there is damage, it is output that there is medium-scale damage to the underwater structure; The mechanical grinding wheel is used to grind the blind area coordinates Q1 and / or Q2 to form an exposed structural surface. The longitudinal wave probe performs laser detection on the polished structural surface to obtain a pixel map P5 (x, y, z), performs a normalization operation on P5 to obtain P6 (x, y, z), and performs damage judgment based on P6 (x, y, z): if it is judged that damage exists, it is output that there is small-scale damage to the underwater structure, and the detection is terminated. The structural surface at the coordinates (x, y, z) is imaged in space and the pixels of P2, P4, and P6 at the coordinates (x, y, z) are superimposed to generate a damage cloud map P7. The pixel map of the damage cloud map P7 = P2+P4+P6.

Images