Underwater laser imaging radar recognition device based on pressure compensation sealing layer

Abstract

The invention discloses an underwater laser imaging radar recognition device based on a pressure compensation sealing layer, relates to a sealing technology of the underwater laser imaging radar recognition device, and aims to solve the problem that an existing underwater laser imaging radar shell is easy to damage in a high-pressure environment. In the device, a gap between the inner layer wall and the outer layer wall of the sealing layer serves as a pressure compensation area and is filled with working fluid, a connector is arranged on the inner layer wall, one end of the high-pressure pump is communicated with the pressure compensation area through the connector, and the other end of the high-pressure pump is communicated with the working fluid gas cylinder. The first pressure sensor and the second pressure sensor are respectively used for measuring the pressure outside the sealing layer and the pressure of the pressure compensation area; the high-pressure pump controller is used for controlling the high-pressure pump to work according to output data of the first pressure sensor and the second pressure sensor; a closed space defined by the inner layer wall serves as a working area, and the laser imaging radar recognition device, the high-pressure pump, the working liquid gas cylinder and the controller are arranged in the working area. According to the invention, the risk of shell leakage can be reduced.

Description

Technical Field

[0001] The invention relates to a sealing technology for an underwater laser imaging radar identification device. Background Art

[0002] In the field of marine technology, underwater detection technology is a key means of breaking down the barrier between humans and the ocean. Underwater LiDAR is one of the means of achieving underwater detection. Underwater LiDARs are mounted on unmanned underwater vehicles (UUVs). The housing of these radars seals and protects the internal equipment. Conventional underwater LiDARs use a single-layer housing, which can fail to withstand the high-pressure environment and can cause damage, creating leaks. Seawater can then enter the housing through these leaks, damaging the equipment inside. Summary of the Invention

[0003] The purpose of the present invention is to solve the problem that the existing underwater laser imaging radar shell is easily damaged under high-pressure environment, and to provide an underwater laser imaging radar identification device based on a pressure compensation sealing layer.

[0004] The underwater laser imaging radar recognition device based on a pressure-compensating sealing layer of the present invention includes a laser imaging radar recognition device, which is characterized in that it also includes a sealing layer, a high-pressure pump, a working fluid cylinder, a first pressure sensor, a second pressure sensor and a high-pressure pump controller; the sealing layer includes an inner wall and an outer wall, the gap between the inner wall and the outer wall serves as a pressure compensation area, the pressure compensation area is filled with working fluid, an interface is provided on the inner wall, one end of the high-pressure pump is connected to the pressure compensation area through the interface, and the other end is connected to the working fluid cylinder; the first pressure sensor and the second pressure sensor are used to measure the pressure outside the sealing layer and the pressure in the pressure compensation area, respectively; the high-pressure pump controller is used to control the operation of the high-pressure pump according to the output data of the first pressure sensor and the second pressure sensor; the enclosed space surrounded by the inner wall serves as a working area, and the laser imaging radar recognition device, the high-pressure pump, the working fluid cylinder and the controller are arranged in the working area.

[0005] Optionally, the high-pressure pump controller includes: a data acquisition unit adapted to acquire the pressure data P1 output by the first pressure sensor and the pressure data P2 output by the second pressure sensor according to a preset frequency; a comparison unit adapted to compare the pressure data P1 and the pressure data P2 acquired within the same pressure sampling period, and if P th1 <P1-P2<P th2 , start the first control unit, if P1-P2≥P th2, start the second control unit, if P1-P2≤P th1 , start the third control unit, where P th2 Less than the maximum pressure that the outer wall can withstand, and 0<P th1 <P th2 ; The first control unit is suitable for sending a control signal to the high-pressure pump to stop the high-pressure pump from working; the second control unit is suitable for sending a control signal to the high-pressure pump to make the working fluid flow from the working fluid cylinder to the pressure compensation area; the third control unit is suitable for sending a control signal to the high-pressure pump to make the working fluid flow from the pressure compensation area to the working fluid cylinder.

[0006] Optionally, the high-pressure pump controller further includes: a detection unit adapted to determine whether a leak occurs in the sealing layer based on pressure data output by the first pressure sensor and the second pressure sensor; and an alarm unit adapted to issue an alarm message when a leak occurs in the sealing layer.

[0007] Optionally, judging whether a leak occurs in the sealing layer based on the pressure data output by the first pressure sensor and the second pressure sensor includes: if the pressure data output by the first pressure sensor shows an upward trend, and the pressure data output by the second pressure sensor shows a downward trend, then judging that a leak occurs in the outer wall and / or the inner wall.

[0008] Optionally, the radar identification device includes a laser controller, a flash radar, a frequency doubling chip, a beam expander, a detector, a data acquisition card, a power supply, four wide-angle image sensors and a data processing module; the control signal output end of the laser controller is connected to the control signal output end of the flash radar, the laser output by the flash radar passes through the frequency doubling chip and the beam expander in sequence and is irradiated on the target to be measured, and the laser returned from the target to be measured is received by the detector, the signal output end of the detector is connected to the signal input end of the data acquisition card, the signal output end of the data acquisition card is connected to the data processing module, and the power supply is used to power the laser controller and the data acquisition card; the four wide-angle image sensors are arranged around the radar identification device, and the signal output ends of the four wide-angle image sensors are connected to the data processing module; windows are arranged on the sealing layer at positions corresponding to the beam expander, the detector, and the four wide-angle image sensors.

[0009] Optionally, part of the cables in the working area pass through the inner wall, the pressure compensation area and the outer wall in sequence and extend to the outside of the outer wall, and the part of the cables is sealed together with the window on the inner wall.

[0010] Optionally, the data processing module includes: a stitching unit, suitable for stitching the four images to form a wide-angle image; a first target recognition unit, suitable for identifying the target to be measured from the wide-angle image; a second target recognition unit, suitable for identifying the target to be measured based on the data collected by the data acquisition card; and a target parameter determination unit, suitable for determining the position and distance of the target to be measured based on the recognition results of the first target recognition unit and the second target recognition unit.

[0011] The present invention achieves pressure compensation on the outer wall of the sealing layer by adjusting the pressure of the working fluid in the sealing layer. While avoiding leakage points in the inner wall, the risk of leakage points in the outer wall is reduced, and the maximum diving depth of the UUV is increased. By emitting high-frequency laser pulses, the depth information of each pixel point is calculated based on the flight time interval, and combined with image stitching technology, it can not only reconstruct the three-dimensional point cloud model in a complex environment, but also eliminate the image distortion caused by dynamic water flow, thereby achieving high-precision reconstruction of the underwater environment.

[0012] In addition, the wide-angle image sensor captures images at a high frame rate to achieve fast imaging, and can achieve high-resolution imaging through the fusion of multiple technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 2 is a schematic structural diagram of an underwater laser imaging radar recognition device based on a pressure compensation sealing layer according to an embodiment of the present application;

[0014] Figure 2 is a structural diagram of a high-pressure pump controller according to an embodiment of the present application;

[0015] Figure 3 It is a structural diagram of a data processing module according to an embodiment of the present application. DETAILED DESCRIPTION

[0016] The following describes the embodiments of the present invention through specific examples. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. The details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0017] As used herein, the singular forms "a", "an", and "the" are intended to include the plural forms as well, unless the context indicates otherwise. It should be further understood that the terms "comprise", "include" indicate the presence of the described features, operations, elements, components, items, kinds, and / or groups, but do not exclude the presence, occurrence, or addition of one or more other features, operations, elements, components, items, kinds, and / or groups. The terms "or" and "and / or" used herein are interpreted as inclusive, or mean any one or any combination. Thus, "A, B, or C" or "A, B, and / or C" means "any of the following: A; B; C; A and B; A and C; B and C; A, B, and C". Exceptions to this definition occur only when the combination of elements, functions, or operations is inherently mutually exclusive in some way.

[0018] In response to the problem that the existing underwater laser imaging radar shell is easily damaged under high-pressure environment, the present invention provides an underwater laser imaging radar identification device based on a pressure-compensating sealing layer, which can reduce the pressure difference on both sides of the shell and reduce the risk of leakage points in the shell.

[0019] Figure 1 Schematic diagram of the structure of an underwater laser imaging radar recognition device based on a pressure compensation sealing layer according to an embodiment of the present application. Figure 1 As shown, the underwater laser imaging radar recognition device based on the pressure compensation sealing layer of the embodiment of the present application is mounted on the UUV, and the collected data is transmitted to the host computer arranged inside the UUV, which processes the data and gives the target recognition result.

[0020] The underwater laser imaging radar recognition device based on the pressure compensation sealing layer of the embodiment of the present application includes a laser imaging radar recognition device 1, a sealing layer 2, a high-pressure pump 3, a working fluid cylinder 4, a first pressure sensor 5, a second pressure sensor 6 and a high-pressure pump controller. The sealing layer 2 serves as the shell of the underwater laser imaging radar recognition device. The sealing layer 2 includes an inner wall 22 and an outer wall 21. The gap between the inner wall 22 and the outer wall 21 serves as a pressure compensation area 23. The pressure compensation area 23 is filled with working fluid. An interface is provided on the inner wall 22. One end of the high-pressure pump 3 is connected to the pressure compensation area 23 through the interface, and the other end is connected to the working fluid cylinder 4. The high-pressure pump 3 can pump the working fluid from the working fluid cylinder 4 into the pressure compensation area 23, and can also pump the working fluid from the pressure compensation area 23 into the working fluid cylinder 4. The working fluid can be water or other liquids. A pressure sensor 5 is arranged outside the sealing layer 2 to measure the pressure of the external seawater; the second pressure sensor 6 is arranged in the pressure compensation area 23 to measure the pressure of the pressure compensation area 23; the high-pressure pump controller can be embedded in the host computer to control the operation of the high-pressure pump 3 according to the output data of the first pressure sensor 5 and the second pressure sensor 6, so that the pressure difference between the inner wall 22 and the outer wall 21 is within their respective tolerance ranges; the enclosed space enclosed by the inner wall 22 serves as a working area, and the laser imaging radar recognition device 1, the high-pressure pump 3 and the working fluid cylinder 4 are arranged in the working area.

[0021] During the dive, the underwater laser imaging radar identification device measures the seawater pressure P1 and the pressure P2 of the working fluid in the pressure compensation zone in real time using the first and second pressure sensors 5 and 6. The pressure in the working zone is recorded as P0. The high-pressure pump controller controls the operation of the high-pressure pump 3 based on this pressure data, maintaining P2 between P0 and P1. If P2 is low, the high-pressure pump controller controls the high-pressure pump 3 to work on the working fluid, causing the working fluid in the working fluid cylinder 4 to flow toward the pressure compensation zone. This increases the density of the working fluid in the pressure compensation zone and the generated pressure P2. Conversely, if P2 is high, the high-pressure pump controller controls the high-pressure pump 3 to cause the working fluid in the pressure compensation zone to flow toward the cylinder 4. This reduces the density of the working fluid in the pressure compensation zone and the generated pressure P2.

[0022] The maximum pressure that the outer wall 21 can withstand is recorded as P' th2 It should be noted that the outer wall 21 here includes the wall material and the sealing part. The sealing part can be, for example, a window on the wall (the window is installed at the hole opened on the wall for letting light in). The maximum pressure that the outer wall 21 can withstand refers to the maximum pressure that the wall material of the outer wall 21 and its sealing part can withstand without any leakage points.

[0023] The shell of a conventional underwater laser imaging radar identification device is a single-layer shell. During the dive, the air pressure P0 inside the shell is constant. Assuming that the material, thickness of the shell and the thickness, area, and sealing process of the hole sealing material are the same as those of the outer wall 21 of the embodiment of the present application, when the UUV dives to a certain depth h1, the difference between the external seawater pressure P1 and P0, P1-P0, reaches P' th2 , a single-layer shell will have leakage points. However, the embodiment of the present application uses a double-layer shell. The working fluid in the pressure compensation zone between the inner wall 22 and the outer wall 21 can compensate for the pressure of the outer wall 21. Specifically, the high-pressure pump controller controls the operation of the high-pressure pump 3 so that the pressure value P2 of the working fluid in the pressure compensation zone is higher than P0 and lower than P1 (P0 can be considered a constant value). Then, when the UUV dives to a depth of h1, the pressure difference P1-P2 on both sides of the outer wall 21 is less than P1-P0. It can be seen that when the underwater laser imaging radar identification device of the embodiment of the present application dives to a depth of h1, no leakage points will appear in the outer wall 21. In addition, the maximum depth h2 to which the underwater laser imaging radar identification device of the embodiment of the present application can dive is higher than the maximum depth h1 to which a conventional underwater laser imaging radar identification device can dive.

[0024] It should be noted that when the high-pressure pump controller is controlling the operation of the high-pressure pump 3, the working fluid pressure P2 in the pressure compensation zone must also ensure that the difference between P2 and P0 is greater than the maximum pressure that the inner wall 22 can withstand. The inner wall 22 includes the wall material and seals, and the maximum pressure that the inner wall 22 can withstand refers to the maximum pressure that the wall material and seals of the inner wall 22 can withstand without leaking.

[0025] In one implementation, the material, thickness, thickness of the inner wall 22 and the thickness, area, and sealing process of the hole sealing material are the same as those of the outer wall 21. Therefore, it should be ensured that the pressure difference P2-P0 on both sides of the inner wall 22 is lower than P' th2 .

[0026] To reduce the risk of leaks in the outer wall 21, the pressure value P2 of the working fluid in the pressure compensation zone should be as close to P1 as possible. Simultaneously, to reduce the risk of leaks in the inner wall 22, the pressure value P1 of the working fluid in the pressure compensation zone should be close to P0. Therefore, the value range of P2 should be reasonably controlled to ensure that the pressure difference between the outer wall 21 and the inner wall 22 is not too large.

[0027] In one implementation, the high pressure pump controller includes a data acquisition unit 210, a comparison unit 220, a first control unit 230, a second control unit 240, and a third control unit 250. Figure 2 shown.

[0028] The data acquisition unit 210 is adapted to acquire the pressure data P1 output by the first pressure sensor and the pressure data P2 output by the second pressure sensor at a preset frequency.

[0029] The comparison unit 220 is adapted to compare the pressure data P1 and the pressure data P2 collected in the same pressure sampling period. th1 <P1-P2<P th2 , start the first control unit 230, if P1-P2≥P th2 , start the second control unit 240, if P1-P2≤P th1 , start the third control unit 250, where P th2 Less than the maximum pressure that the outer wall can withstand, and 0<P th1 <P th2 .

[0030] To avoid the pressure difference on both sides of the outer wall 21 being at the critical point P' for a long time th2 , increasing the risk of leakage points, the threshold P can be set th2 , P th2 is less than P' th2 During the UUV dive, the high-pressure pump controller controls the high-pressure pump 3 to work so that P2 satisfies: P1-P2<P th2 At the same time, in order to reduce the load of the high-pressure pump 3 and reduce the pressure difference on both sides of the inner wall 22, the value of P2 should be appropriately reduced, and a threshold value P greater than 0 should be set. th1 During the UUV dive, the high-pressure pump controller controls the high-pressure pump 3 to work so that P2 satisfies: P1-P2>P th1 .

[0031] The first control unit 230 is adapted to send a control signal to the high-pressure pump 3 to stop the high-pressure pump 3. A control cycle is defined as the time from when the high-pressure pump controller sends a control signal to the high-pressure pump 3 once to when the high-pressure pump controller sends a control signal to the high-pressure pump 3 next time. If the high-pressure pump 3 was in a working state (pumping the working fluid from the working fluid cylinder / pressure compensation area into the pressure compensation area / working fluid cylinder) during the previous control cycle, the control signal sent by the high-pressure pump controller to the high-pressure pump 3 in the first control unit 230 causes the high-pressure pump 3 to stop working. If the high-pressure pump 3 was in a standby state during the previous control cycle, the control signal sent by the high-pressure pump controller to the high-pressure pump 3 in the first control unit 230 may be a null signal, which is equivalent to the high-pressure pump controller not sending any signal to the high-pressure pump 3, so that the high-pressure pump 3 maintains its current state.

[0032] The second control unit 240 is adapted to send a control signal to the high-pressure pump 3 so as to enable the working fluid to flow from the working fluid cylinder to the pressure compensation area.

[0033] The third control unit 250 is adapted to send a control signal to the high-pressure pump 3 so as to enable the working fluid to flow from the pressure compensation area to the working fluid cylinder.

[0034] As a preferred embodiment of the present application, the high-pressure pump controller further includes a detection unit and an alarm unit.

[0035] The detection unit is adapted to determine whether a leak occurs in the sealing layer according to pressure data output by the first pressure sensor and the second pressure sensor.

[0036] During the dive, the pressure data P1 collected by the first pressure sensor shows an upward trend. In order to ensure that the difference between P1 and P2 is maintained at P th1 <P1-P2<P th2 Within this range, P2 should also show an upward trend. If a leak develops on the outer wall 21, because the external seawater pressure P2 is significantly higher than the working fluid pressure P2 in the pressure compensation area, seawater will quickly flow into the pressure compensation area, increasing its volume and causing P2 to change from an upward trend to a downward trend. If a leak develops on the inner wall 22, because the working fluid pressure P2 in the pressure compensation area is higher than the working fluid pressure P0, the working fluid in the pressure compensation area will flow through the leak into the working area, causing P2 to decrease. However, the high-pressure pump 3 pumps the working fluid from the working fluid cylinder 4 into the pressure compensation area, causing P2 to rise for a short period of time. After this brief rise, it will continue to decline.

[0037] Specifically, a time value T can be set. If P1 shows a continuous upward trend while P2 shows a continuous downward trend within T, it can be considered that a leak has occurred in the sealing layer 2. For example, set T to 10 seconds and record the changes in P2. If P2 changes from rising to falling at a certain moment and then continues to fall for the next 10 seconds, while P1 continues to rise within these 10 seconds, it is considered that a leak has occurred.

[0038] The alarm unit is adapted to send out an alarm message when a leakage point occurs in the sealing layer.

[0039] When the detection unit detects a leak, the alarm unit immediately issues an alarm message, which is sent via the host computer and preferably includes a sound alarm and a text prompt.

[0040] The radar identification device 1 of the present embodiment includes a laser controller 11, a flash radar 12, a frequency multiplication chip 13, a beam expander 14, a detector 15, a data acquisition card 16, a power supply 17, and four wide-angle image sensors 18. The control signal output of the laser controller 11 is connected to the control signal input of the flash radar 12. The laser light output by the flash radar 12 passes through the frequency multiplication chip 13 and the beam expander 14 in sequence before irradiating the target to be measured. The laser light returned by the target to be measured is received by the detector 15. The signal output of the detector 15 is connected to the signal input of the data acquisition card 16, and the signal output of the data acquisition card 16 is connected to the data processing module of the host computer. The power supply 17 is used to power the laser controller 11 and the data acquisition card 16. The four wide-angle image sensors 18 are arranged around the radar identification device 1, and their signal outputs are connected to the data processing module. There are several holes on the sealing layer 2, which correspond to the expander 14, the detector 15, and the four wide-angle image sensors 18 respectively. A window is embedded in each hole, and the connection between the window and the inner wall 22 and the outer wall 21 (i.e., the edge of the hole) is sealed.

[0041] Some devices in the work area need to communicate with the host computer, such as the wide-angle image sensor 18 and the data acquisition card 16. The data cables of these devices need to pass through the sealing layer 2 and extend outside the sealing layer 2 to connect to the host computer. In this embodiment of the application, these cables are packaged into a bundle and sealed together with a window in a single opening to reduce the sealing area and reduce the risk of leaks in the sealing layer 2.

[0042] In the existing technology, there are two ways to route cables inside underwater laser imaging radars: the first is for the cables to be led out of a single hole, extend to the outside of the shell, and connect to the host computer inside the UUV; the second is for the cables to be led out of the hole where the window in front of the beam expander or array detector is located, extend to the outside of the shell, and connect to the host computer inside the UUV. The first method increases the sealing area, resulting in a higher risk of leaks in the shell; the second method, because the window in front of the beam expander or array detector is directly exposed to seawater, the cables need to pass through the seawater after being led out of the hole and then return to the inside of the UUV. After long-term operation, the cables in the seawater are easily corroded, and the cables are subject to forces from all directions in the seawater, including the force of the seawater on the cables and the force of marine life, garbage, etc. on the cables, affecting the operation of the UUV.

[0043] The cable of the embodiment of the present application can be led out from a hole in the inner wall 22 and sealed together with the window in front of the beam expander 14, the detector 15 or the wide-angle image sensor 18. This not only reduces the sealing area and reduces the risk of leakage points in the shell, but also the cable passes through the inner wall 22, then passes through the pressure compensation area 23, and finally passes through the outer wall 21 connected to the UUV, enters the interior of the UUV and is connected to the host computer. The cable does not pass through seawater, will not be corroded, and will not affect the operation of the UUV.

[0044] In the radar recognition device 1, four wide-angle image sensors 18 are located at different locations and are used to capture 2D images of the surrounding environment. Before commissioning, the four wide-angle image sensors 18 must be calibrated, a mapping relationship established, distortion corrected, and transformed to a unified bird's-eye view. The overlapping areas of the fields of view of adjacent wide-angle image sensors are calculated based on the calibrated extrinsic parameters. Projection parameters are adjusted through feature point matching to eliminate stitching misalignment. Weights are assigned based on the distance of pixels from the overlapping boundary, smoothing out brightness and color differences between the images captured by the different wide-angle image sensors 18. Multi-image fusion and seam optimization are completed to restore a 360-degree panoramic image of the radar recognition device.

[0045] In one implementation, the image stitching process includes two stages: image registration and image fusion.

[0046] Image registration is the process of automatically aligning two images of the same scene taken by two cameras from different perspectives. Its essence is to calculate the affine transformation relationship between the two images. The affine transformation model is defined as follows:

[0047]

[0048] One of the two images is used as the reference image, and the other is used as the registration image. In the affine transformation model, (x, y) is the coordinate on the reference image, (x′, y′) is the coordinate on the registration image, and t x , t y Represents the amount of translation of the image in the horizontal and vertical directions respectively. For each corner point in the reference image, a normalized cross-correlation function is used to roughly match the corner point in the registration image. The correlation coefficient C is defined as follows:

[0049]

[0050] Where m and n represent the horizontal and vertical offsets of the matching window centered at (x, y) and (x′, y′) (the window ranges from [-w, w] and [-h, h] in the horizontal and vertical directions respectively); w and h represent half the width and half the height of the overlapping area of ​​the two images respectively; I1 and I2 represent the grayscale value of a point in the two images respectively; and They represent the average grayscale value of all pixels in the window centered at point (x, y) in the two images. The correlation coefficient C ranges from -1 to 1. If C is greater than a preset threshold α (usually 0.95), the corner point is considered to be matched successfully.

[0051] The above method is used to align the four images captured by the four wide-angle image sensors 18 .

[0052] Image fusion is to fuse the four aligned images to eliminate the seams between the images. x The absolute value of t y , a fusion image will be generated along the x direction, and conversely a fusion image will be generated along the y direction. x or t y If it is greater than 0, the fusion direction is converted from the reference image to the registered image, and vice versa. Therefore, the fusion rule of the overlapping area is defined as follows:

[0053] I(x,y)=uI1(x1,y1)+vI2(x2,y2)

[0054] Where I(x,y) is the grayscale value of the fused image at (x,y), I1(x1,y1) and I2(x2,y2) are the grayscale values ​​of the registered image at (x1,y1) and (x2,y2), respectively. u and v are adaptive weighting coefficients that can be adaptively changed according to the pixel values ​​of each row and column in the overlapping area. The definitions of u and v are as follows:

[0055]

[0056] where N j is the number of pixels in the jth row or jth column of the overlapping region, and i is defined as the number of pixels traversed by the jth or jth column in the overlapping region.

[0057] The registered images are fused in the above manner to complete the stitching of the four images.

[0058] In the radar recognition device 1 described above, the flash radar 12 uses a VCSEL laser as its signal source. A VCSEL laser is a surface-emitting light source that continuously emits laser pulses toward a target during operation. The laser beam passes through a beam expander 14 to achieve a wide field of view. Laser pulses reflected from target objects within the field of view are received by a detector 15, which utilizes a highly sensitive SPAD array sensor. Each pixel in the sensor independently records the arrival time and intensity of photons. The detection signal output by the detector 15 is collected by a data acquisition card 16 and transmitted to the data processing module of the host computer.

[0059] The images collected by the wide-angle image sensor 18 and the data collected by the data acquisition card 16 are processed by the data processing module. Figure 3 As shown, the data processing module includes a stitching unit 310, a first target recognition unit 320, a second target recognition unit 330, and a target parameter determination unit 340. The stitching unit 310 stitches the four images captured by the four wide-angle image sensors 18 to form a wide-angle image; the first target recognition unit 320 recognizes the target to be measured from the wide-angle image; the second target recognition unit 330 recognizes the target to be measured based on the data collected by the data acquisition card; and the target parameter determination unit 340 determines the position and distance of the target to be measured based on the recognition results of the first target recognition unit 320 and the second target recognition unit 330.

[0060] The stitching unit 310 of the data processing module stitches the images captured by the four wide-angle image sensors 18 , and the first target recognition unit 320 obtains the position information of the target from the stitched wide-angle images.

[0061] The second target recognition unit 330 uses the ToF method to measure the distance between the target and the radar recognition device 1. The ToF ranging principle uses laser pulses as the carrier of information. The pulse width of the laser pulse is very narrow, usually at the nanosecond level. The detector 15 receives the reflected signal and derives the distance between the laser and the object to be measured by measuring the time it takes the laser pulse to propagate in space. The time it takes the laser pulse to propagate is the flight time. There is basically no angle difference between the laser and the detector 15, so it is only necessary to know the time t when the laser emits the pulse laser. start and the time t at which the detector 15 receives the signal stop , we can get the flight time t TOF :

[0062] t TOF =t stop -t start

[0063] After obtaining the flight time t TOF Then, the distance d between the radar recognition device 1 and the object to be detected can be calculated:

[0064]

[0065] In the above formula, c is the speed of light. Because the laser pulse signal transmits quickly and receives data in a very short time, the embodiments of the present application can achieve real-time measurement. Furthermore, due to the high energy and short pulse width of the laser pulse signal, sensitivity to ambient interference sources is reduced, resulting in strong anti-interference capabilities and the ability to simultaneously resolve multiple targets.

[0066] The second target recognition unit 330 uses the global exposure characteristics of the Flash lidar and the ToF time tag to bind the intensity (reflectivity) of each pixel with the depth information, construct a point cloud of the surrounding environment, and then reconstruct a three-dimensional point cloud model. A flight time histogram is constructed according to the photon flight time, and the direct reflection signal is distinguished from the multiple scattering noise. The valid target points are retained to obtain a three-dimensional point cloud model that only contains the depth and position information of the surrounding environment.

[0067] Finally, the target parameter determination unit 340 determines the position and distance of the target to be measured by integrating the recognition results of the first target recognition unit 320 and the second target recognition unit 330. Specifically, the target of interest can be locked from the recognition results of the first target recognition unit 320, and these targets can be found from the recognition results of the second target recognition unit 330 based on the position information of these targets, and then the distance of these targets can be determined based on the three-dimensional point cloud model obtained by the second target recognition unit 330.

[0068] The data processing module can reconstruct the depth, position and specific image information of the underwater environment by fusing the three-dimensional point cloud model with the 360-degree surround view image, thereby realizing the identification of underwater environment targets.

[0069] Traditional optical imaging uses a lens system to stably project light onto a specific location on the imaging sensor, forming an image corresponding to the spatial relationship. However, dynamic water flow changes the propagation path of light, generating a distorted two-dimensional image. Two adjacent points on an object may be pulled apart or squeezed in the image, causing the overall shape to be distorted. However, the distance value calculated by the time-of-flight method in the embodiment of the present application is the real-space distance information of the target point relative to the sensor along the actual light path. When light is used to construct the final point cloud or three-dimensional image, the position information of each point is the absolute spatial position directly measured and calculated. Although water flow distorts the light path, so that the position of a single point may not be the position under the ideal straight line, the laser imaging radar recognition device 1 truly records the actual distance of the point under the distorted light path and calculates its coordinates accordingly, avoiding the reliance on the stability of the entire light path in traditional imaging. The relative position relationship between points is directly assigned in three-dimensional space, rather than inferred from a distorted two-dimensional image. Therefore, the embodiment of the present application greatly reduces the image distortion caused by dynamic water flow in traditional optical imaging.

[0070] In addition, the embodiment of the present application combines the time-of-flight method with image stitching technology to reduce the interference of the complex underwater environment on the image, thereby achieving the purpose of high-resolution imaging.

[0071] The data processing module and high-pressure pump controller described herein can be implemented in combination with hardware or software, or a combination thereof. Thus, the method and apparatus of the present invention, or certain aspects or portions of the method and apparatus of the present invention, can take the form of program code (i.e., instructions) embedded in a tangible medium, such as a removable hard disk, USB flash drive, floppy disk, CD-ROM, or any other machine-readable storage medium. When the program is loaded into a machine such as a computer and executed by the machine, the machine becomes an apparatus for practicing the present invention.

Claims

1. An underwater laser imaging radar recognition device based on a pressure compensation sealing layer, comprising a laser imaging radar recognition device, characterized in that: It also includes a sealing layer, a high-pressure pump, a working fluid cylinder, a first pressure sensor, a second pressure sensor, and a high-pressure pump controller; The sealing layer includes an inner wall and an outer wall, the gap between the inner wall and the outer wall serves as a pressure compensation area, the pressure compensation area is filled with a working fluid, and an interface is provided on the inner wall, one end of the high-pressure pump is connected to the pressure compensation area through the interface, and the other end is connected to the working fluid cylinder; The first pressure sensor and the second pressure sensor are used to measure the pressure outside the sealing layer and the pressure in the pressure compensation area respectively; The high-pressure pump controller is used to control the operation of the high-pressure pump according to the output data of the first pressure sensor and the second pressure sensor; The enclosed space surrounded by the inner wall serves as a working area, and the laser imaging radar recognition device, the high-pressure pump, the working fluid cylinder and the controller are arranged in the working area.

2. The device according to claim 1, wherein The high-pressure pump controller includes: a data acquisition unit adapted to acquire pressure data P1 output by the first pressure sensor and pressure data P2 output by the second pressure sensor at a preset frequency; The comparison unit is suitable for comparing the pressure data P1 and pressure data P2 collected in the same pressure sampling period. th1 <P1-P2<P th2 , start the first control unit, if P1-P2≥P th2 , start the second control unit, if P1-P2≤P th1 , start the third control unit, where P th2 Less than the maximum pressure that the outer wall can withstand, and 0<P th1 <P th2 ; a first control unit, adapted to send a control signal to the high-pressure pump to stop the high-pressure pump; a second control unit, adapted to send a control signal to the high-pressure pump to enable the working fluid to flow from the working fluid cylinder to the pressure compensation area; The third control unit is adapted to send a control signal to the high-pressure pump to enable the working fluid to flow from the pressure compensation area to the working fluid cylinder.

3. The device according to claim 1, wherein The high-pressure pump controller further includes: a detection unit, adapted to determine whether a leak occurs in the sealing layer based on pressure data output by the first pressure sensor and the second pressure sensor; and The alarm unit is adapted to send out an alarm message when a leakage point occurs in the sealing layer.

4. The device according to claim 3, wherein The determining whether a leak occurs in the sealing layer according to the pressure data output by the first pressure sensor and the second pressure sensor includes: If the pressure data output by the first pressure sensor shows an upward trend, and the pressure data output by the second pressure sensor shows a downward trend, it is determined that a leakage point exists in the outer wall and / or the inner wall.

5. The device according to claim 1, wherein The radar recognition device includes a laser controller, a flash radar, a frequency multiplier, a beam expander, a detector, a data acquisition card, a power supply, four wide-angle image sensors, and a data processing module; The control signal output end of the laser controller is connected to the control signal input end of the flash radar. The laser output by the flash radar passes through the frequency doubling plate and the beam expander in sequence and then irradiates the target to be measured. The laser returned by the target to be measured is received by the detector. The signal output end of the detector is connected to the signal input end of the data acquisition card. The signal output end of the data acquisition card is connected to the data processing module. The power supply is used to power the laser controller and the data acquisition card. The four wide-angle image sensors are arranged around the radar recognition device, and the signal output ends of the four wide-angle image sensors are connected to the data processing module; Windows are arranged on the sealing layer at positions corresponding to the beam expander, the detector, and the four wide-angle image sensors.

6. The device according to claim 5, characterized in that Part of the cables in the working area passes through the inner wall, the pressure compensation area and the outer wall in sequence and extends to the outside of the outer wall. The part of the cables is sealed together with the window on the inner wall.

7. The device according to claim 6, characterized in that The data processing module includes: a stitching unit, adapted to stitch the four images together to form a wide-angle image; a first target recognition unit, adapted to recognize a target to be detected from the wide-angle image; A second target recognition unit, adapted to recognize the target to be measured based on the data collected by the data acquisition card; and The target parameter determination unit is adapted to determine the position and distance of the target to be measured according to the recognition results of the first target recognition unit and the second target recognition unit.

Images