A ship height measurement and overheight warning system and method for bridge area waters

Abstract

The present invention discloses a ship height measurement and super-elevation warning system and method for bridge area waters. The super-elevation monitoring module and the super-elevation warning module are both connected to the super-elevation height measurement calculation module; the super-elevation monitoring module includes a long-short focus camera array, an IP subnet and a network cloud, and the long-short focus camera array is connected to the network cloud through the IP subnet; wherein the long-short focus camera array is respectively installed on both sides of the river. Beneficial effects of the present invention: the present invention installs a long-short focus camera array on both sides of the river, and the long-short focus camera array measures the height of the passing ship, and then the relevant data of the ship is calculated by the super-elevation height measurement calculation module, and the data is fed back to the super-elevation warning module and the system management personnel. If the measurement result shows that the passing ship is super-elevated, the super-elevation warning module will work to prompt the ship's cab, and at the same time, the relevant alarm signal will be fed back to the system management personnel and the relevant supervision department, which can avoid the risk of super-elevation when the ship passes through the bridge.

Description

Technical Field

[0001] The present invention relates to the field of ship and shipping technology, and in particular to a ship height measurement and overheight warning system and method for bridge area waters. Background Art

[0002] my country's inland waterway shipping is well developed. As an important hub connecting the two sides of the river, cross-water bridges play an important role in the development of the transportation industry. In recent years, the number of cross-water bridges has also increased year by year. At the same time, bridges have also brought many constraints to ship navigation. In recent years, accidents of ship damage and bridge damage caused by insufficient estimation of the clearance height of bridges by ship drivers have occurred from time to time. At the same time, how to reduce the probability of ship collision accidents caused by the clearance height limit of bridge navigation has always been the focus of attention of bridge operators, and it has also brought huge pressure on the safety supervision of management agencies. In order to ensure the safe navigation of ships in the bridge area and the safety of bridges, it is particularly important for ships to accurately grasp the clearance height of bridges before passing through the bridge. At present, bridges in navigable waters usually use bridge navigation aids and warning signs to remind passing ships. There are also physical anti-collision devices such as rubber and buoys in the waters of bridge areas for passive anti-collision. On the one hand, due to the large number of bridges in a certain water area and the changeable water level and water flow, it is often difficult for ship drivers to accurately grasp the real-time height of a bridge above the water surface. On the other hand, since passive anti-collision devices are usually set at a certain distance outside the bridge or physically protect a bridge pier, when a ship touches or is about to collide with the top of the bridge or a bridge pier, the passive anti-collision device is usually difficult to provide effective protection.

[0003] According to the Inland Waterway Navigation Standard (GB50139-2014) and the Seagoing Vessel Channel Navigation Standard (JTS180-3-2018), the bridge navigation clearance height is the sum of the height above the waterline of the representative ship type and the surplus height, and the starting surface is the designed highest navigable water level. The surplus height is the surplus amount set to ensure the safety of ships crossing the bridge. To ensure the safe passage of ships through the bridge, the height from the waterline of the ship to the highest fixed point must be less than the bridge navigation clearance height. Under normal circumstances, the water level will change to a certain extent due to factors such as water level control, tides, and strong winds. For ship drivers, the real-time navigation height of the bridge is difficult to accurately judge, and there is a certain risk in ship navigation. Therefore, it is very important to master the accurate real-time navigation height and other major navigation environment information in advance to ensure the safety of ship navigation and bridge safety.

[0004] 1. For example, a Chinese patent discloses a real-time display system for the navigation height of a bridge (application number: CN201420284353.4), which includes a navigation height display screen, a main control board, a water level gauge, a wireless transmission terminal, an LED monitoring board and a solar power supply module. The main control board is connected to the water level gauge and the wireless transmission terminal respectively, and the output end of the main control board is connected to the input end of the navigation height display screen through the LED monitoring board, and the input end of the main control board is connected to the output end of the solar power supply module.

[0005] 2. For example, a Chinese patent discloses a ship navigation warning system for bridge area waters (application number: CN202020372017.0), which integrates and normalizes the measured water level and water flow information through a remote integrated terminal to obtain navigation information of the bridge area waters including the real-time clearance height, water depth, flow velocity and flow direction of the bridge, and transmits the navigation information of the bridge area waters to a short-range display terminal for display, so that ship drivers can intuitively obtain navigation information of the waters to be passed through the bridge area before crossing the bridge; the remote integrated terminal transmits the navigation information of the bridge area waters to the ship navigation warning platform for the bridge area waters, and the ship drivers can proactively query the navigation information of the waters to be passed through the bridge area in advance through the autonomous query terminal; the utility model realizes the early warning of navigation information through the "short-range direct display + remote active query" method, thereby effectively reminding the ship drivers to make reasonable judgments in advance to reduce the probability of accidents such as ship damage and bridge damage.

[0006] The above-mentioned prior art uses the measurement of the water level of the bridge for early warning, but does not measure the height of the navigable ships. If some ships are high and exceed the height of the bridge, without early warning, accidents of ships colliding with the bridge may also occur.

[0007] In addition, the existing technology also uses an over-height and over-limit laser warning system to determine whether a ship is over-height, but the passage of birds and internal instability of the system may cause the system to falsely alarm.

[0008] Therefore, it is necessary to propose a ship height measurement and overheight warning system and method for the bridge area waters to solve the above problems. Summary of the invention

[0009] In view of the above-mentioned deficiencies in the prior art, the object of the present invention is to provide a ship height measurement and overheight warning system and method for bridge area waters to solve the above-mentioned problems.

[0010] A ship height measurement and super-altitude warning system for bridge area waters comprises a super-altitude monitoring module, a super-altitude measurement calculation module and a super-altitude warning module, wherein the super-altitude monitoring module and the super-altitude warning module are both connected to the super-altitude measurement calculation module.

[0011] Preferably, the ultra-high monitoring module includes a long-focus and short-focus camera array, an IP subnet and a network cloud, and the long-focus and short-focus camera array is connected to the network cloud via the IP subnet; wherein the long-focus and short-focus camera arrays are respectively installed on both sides of the river.

[0012] Preferably, the long-focus and short-focus camera array includes a long-focus camera, a short-focus camera, a POE switch, a pole and a control box. The control box is installed on the pole. The long-focus camera, the short-focus camera and the POE switch are all installed in the control box. A lens attitude fine-tuning gimbal is installed at the bottom of the long-focus camera.

[0013] Preferably, the superelevation height measurement calculation module includes an inland river ship face feature library, a ship face recognition module, a ship high point recognition module, a ship height measurement calculation module, a bridge clear height limit processing module and an integrated server; the inland river ship face feature library is connected to the integrated server through the ship face recognition module; the ship high point recognition module, the ship height measurement calculation module and the bridge clear height limit processing module are all electrically connected to the integrated server.

[0014] Preferably, the ultra-high warning module includes an early warning AIS message transmitting device, a VHF voice broadcasting device, an early warning text message transmitting device, a tweeter and a high-intensity searchlight.

[0015] Its working steps are:

[0016] S1. The short-focus camera obtains the water surface image, and then uses the deep learning neural network to detect and identify whether there is a ship in the image. The obtained image is used as a training set for the ship mast recognition neural network to enhance the network's detection and recognition capabilities;

[0017] S2, training the neural network to obtain a ship detection identifier with good recognition performance;

[0018] S3, when the short-focus camera detects and identifies a ship, the long-focus camera is used to segment the moving ship;

[0019] S4, accurately measuring the identified ship mast using imagery;

[0020] S5. Feedback whether the system measurement result is over-height and the measured height to the ship and traffic management system.

[0021] A method for ship height measurement and over-height warning in bridge area waters, including a height measurement method with a long- and short-focus combined camera, the method steps are as follows:

[0022] S61. The telephoto cameras on both sides of the river simultaneously obtain images of the ship's ultra-high area, and take the images of the two images respectively.

[0023] From the optical center to the pixel point at the highest point of the ship, we get h r ,hl ;

[0024] S62, after the cameras are installed, the positions of the two telephoto cameras are accurately measured by RTK to obtain a baseline distance T between the two cameras;

[0025] S63, the focal lengths f of the two telephoto cameras are obtained by the large scene camera internal parameter calibration method r 、f l ;

[0026] S64, by pixel h r ,h l , the baseline distance T between the two cameras, and the focal lengths of the two cameras are used to calculate the super-high height ΔH:

[0027]

[0028] It also includes a large-scene camera internal parameter calibration method, the method steps of which are: using an RTK measuring instrument to obtain the elevation information of a reference object and the image information of the reference object, and using the formula of similar triangles to calculate the focal length f of the camera:

[0029]

[0030] Where T is the straight-line distance from the reference object to the camera, which is accurately measured by the RTK measuring instrument; H is the elevation of the reference object, and h is the number of pixels occupied by the reference object in the image.

[0031] It also includes a graded warning method, which is as follows: when the short-focus camera obtains the image of the ship, a prompt will be given in the background to remind the platform to be prepared in advance; after the multi-camera array device measures the height of the ship, if it is too high, strong light, voice broadcast and other methods will be used to remind the ship, and the background will remind the system management personnel that the ship is too high, and send text messages to relevant regulatory departments, thereby reminding incoming ships.

[0032] It also includes a hierarchical recognition and detection method using a combination of long- and short-focus cameras; the method is that a short-focus camera acquires a water surface image to detect and identify a ship, and when a ship is detected passing by, a long-focus camera is used to segment the ship's super-elevated area to obtain the ship's highest point, and then visual measurement is performed.

[0033] Compared with the prior art, the present invention has the following beneficial effects: the present invention installs a long- and short-focus camera array on both sides of the river, and the long- and short-focus camera array measures the height of the passing ships. Then, the relevant data of the ships is calculated by the super-elevation height measurement calculation module, and the data is fed back to the super-elevation warning module and the system management personnel. If the measurement result shows that the passing ship is super-elevated, the super-elevation warning module will work to prompt the ship's cab, and at the same time, the relevant alarm signal will be fed back to the system management personnel and relevant regulatory departments. The technical solution of the present invention can avoid the risk of super-elevation when the ship passes through the bridge. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 The diagram is a principle block diagram of the ship height measurement and overheight warning system for bridge area waters of the present invention.

[0035] Figure 2 This is a structural diagram of a long- and short-focus camera array of the present invention;

[0036] Figure 3 This is a schematic diagram of the working principle of the long-focus and short-focus camera arrays of the present invention;

[0037] Figure 4 A scene diagram of a ship entering a detection area of ​​the present invention;

[0038] Figure 5 The ship height measurement workflow diagram of the present invention;

[0039] Figure 6 This is a schematic diagram of the ultra-high detection principle of the telephoto camera of the present invention;

[0040] Figure 7 and Figure 8 It is a schematic diagram of the ultra-high measurement principle based on variant binoculars of the present invention;

[0041] Fig. 9 This is a schematic diagram of the camera calibration principle of the present invention. DETAILED DESCRIPTION

[0042] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0043] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first", "second", and the like are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, features defined as "first", "second", and the like may explicitly or implicitly include one or more of the features. In the description of the present invention, unless otherwise specified, "multiple" means two or more.

[0044] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood by specific circumstances.

[0045] The embodiments of the present invention are described in detail below with reference to the accompanying drawings, but the present invention can be implemented in many different ways as defined and covered by the claims.

[0046] like Figure 1 Combined with Figures 2 to 9 As shown, a ship height measurement and super-altitude warning system for bridge area waters includes an super-altitude monitoring module 20, an super-altitude measurement calculation module 30 and an super-altitude warning module 40, and the super-altitude monitoring module 20 and the super-altitude warning module 40 are both connected to the super-altitude measurement calculation module 30.

[0047] Furthermore, the ultra-high monitoring module 20 includes a long-focus and short-focus camera array 10, an IP subnet 21 and a network cloud 22, wherein the long-focus and short-focus camera array 10 is connected to the network cloud via the IP subnet; wherein the long-focus and short-focus camera arrays are installed on both sides of the river respectively. The network cloud 22 is a server.

[0048] Furthermore, the long-focus and short-focus camera array includes a long-focus camera 1, a short-focus camera 2, a POE switch 4, a pole 5 and a control box 6. The control box 6 is installed on the pole 5. The long-focus camera 1, the short-focus camera 2 and the POE switch 4 are all installed in the control box 6. A lens attitude fine-tuning gimbal 3 is installed at the bottom of the long-focus camera 1.

[0049] Furthermore, the superelevation height measurement calculation module 30 includes an inland river ship face feature library 31, a ship face recognition module 32, a ship high point recognition module 33, a ship height measurement calculation module 34, a bridge clear height limit processing module 35 and an integrated server 36; the inland river ship face feature library 31 is connected to the integrated server 36 through the ship face recognition module 32; the ship high point recognition module 33, the ship height measurement calculation module 34 and the bridge clear height limit processing module 35 are all electrically connected to the integrated server 36.

[0050] Furthermore, the ultra-high warning module 4 includes a warning AIS message transmitting device 41, a VHF voice broadcasting device 42, a warning text message transmitting device 43, a tweeter 44 and a high-intensity searchlight 45.

[0051] Its working steps are:

[0052] S1. The short-focus camera obtains the water surface image, and then uses the deep learning neural network to detect and identify whether there is a ship in the image. The obtained image is used as a training set for the ship mast recognition neural network to enhance the network's detection and recognition capabilities;

[0053] S2, training the neural network to obtain a ship detection identifier with good recognition performance;

[0054] S3, when the short-focus camera detects and identifies a ship, the long-focus camera is used to segment the moving ship;

[0055] S4, accurately measuring the identified ship mast using imagery;

[0056] S5. Feedback whether the system measurement result is over-height and the measured height to the ship and traffic management system.

[0057] In inland waters, a short-focus lens is used to obtain a water surface image, and then a deep learning neural network is used to detect and identify whether there is a ship in the image. If there is a ship in the image, the telephoto camera works, and the neural network is used to detect and identify the possible super-high area of ​​the ship (usually the ship's mast), and the image of the possible super-high area is saved as a training set for the neural network for subsequent super-high ship part detection and identification. Finally, the ship is image-measured based on the detected super-high area of ​​the ship to determine whether the ship is super-high.

[0058] A method for ship height measurement and over-height warning in bridge area waters, including a height measurement method with a long- and short-focus combined camera, the method steps are as follows:

[0059] S61. The telephoto cameras on both sides of the river simultaneously obtain images of the ship's ultra-high area, and take the pixel points from the optical center to the highest point of the ship in the two images to obtain h r ,h l ;

[0060] S62, after the cameras are installed, the positions of the two telephoto cameras are accurately measured by RTK to obtain a baseline distance T between the two cameras;

[0061] S63, the focal lengths f of the two telephoto cameras are obtained by the large scene camera internal parameter calibration method r 、f l ;

[0062] S64, by pixel h r ,h l , the baseline distance T between the two cameras, and the focal lengths of the two cameras are used to calculate the super-high height ΔH:

[0063]

[0064] It also includes a large-scene camera internal parameter calibration method, the method steps of which are: using an RTK measuring instrument to obtain the elevation information of a reference object and the image information of the reference object, and using the formula of similar triangles to calculate the focal length f of the camera:

[0065]

[0066] Where T is the straight-line distance from the reference object to the camera, which is accurately measured by the RTK measuring instrument; H is the elevation of the reference object, and h is the number of pixels occupied by the reference object in the image.

[0067] It also includes a graded warning method, which is as follows: when the short-focus camera obtains the image of the ship, a prompt will be given in the background to remind the platform to be prepared in advance; after the multi-camera array device measures the height of the ship, if it is too high, strong light, voice broadcast and other methods will be used to remind the ship, and the background will remind the system management personnel that the ship is too high, send text messages to relevant regulatory departments, and then remind incoming ships.

[0068] It also includes a hierarchical recognition and detection method using a combination of long- and short-focus cameras; the method is that a short-focus camera acquires a water surface image to detect and identify a ship, and when a ship is detected passing by, a long-focus camera is used to segment the ship's super-elevated area to obtain the ship's highest point, and then visual measurement is performed.

[0069] Compared with the prior art, the present invention has the following beneficial effects: the present invention installs a long- and short-focus camera array on both sides of the river, and the long- and short-focus camera array measures the height of the passing ships. Then, the relevant data of the ships is calculated by the super-altitude measurement calculation module, and the data is fed back to the super-altitude warning module and the system management personnel. If the measurement result shows that the passing ship is super-altitude, the super-altitude warning module will work to prompt the ship's cab, and at the same time, the relevant alarm signal will be fed back to the system management personnel and relevant regulatory departments. The technical solution of the present invention can avoid the risk of super-altitude when the ship passes through the bridge; the neural network is used to analyze whether the picture is a bird or a ship, thereby avoiding the possibility of false alarm of the warning.

[0070] The present invention adopts the non-traditional binocular ranging principle to build models and measure the height of ships, and achieves good results; adopts a brand-new camera internal parameter calibration method to solve the camera internal parameter problem in large scenes; adopts a brand-new multi-camera array device, that is, two pole-type long-short focal length combined camera devices installed on both sides of the river and arranged oppositely; uses a multi-camera array to detect, identify, segment and measure ships at different scales; uses vision to accurately measure in large scenes (≥1000m) in inland waters; has high measurement accuracy, and the minimum height error is 0.02m on a 1.5km baseline; and in view of the complex and changeable environment of inland waters, multiple The camera array device is not limited by the width of the water area in the bridge area and can accurately measure in waters of various widths; the ship recognition algorithm inside the system has high recognition accuracy, and there is a specific data set for inland ships to train the neural network, and the data set can be continuously supplemented according to the monitored ship images to continuously improve the accuracy of ship detection and recognition; based on online video monitoring, it can monitor in real time whether the ship is too high, and can issue an over-height warning in time for over-height ships, so that the over-height ships have enough reaction time to avoid ship-bridge collision accidents; the device is highly practical and has strong measurement robustness, and can still maintain relatively accurate measurements when there is a certain degree of error in the focal length.

[0071] like Figure 4 As shown in the scene diagram when the ship enters the detection area, long and short focus camera arrays 10 are installed on both sides of the river 11. The long and short focus camera arrays 10 measure the height and width of the passing ship 15, and then the relevant data of the ship 15 is calculated by the super-elevation height measurement calculation module 30 (matched with the height of the bridge 13 and the distance between the two piers 14), and the data is fed back to the super-elevation warning module 40 and the system management personnel.

[0072] Embodiment 1:

[0073] When the ship 15 enters the water area 12 of the bridge area, the short-focus camera 2 obtains the water surface image and determines whether there is a ship 15 through the neural network; if not, it means that something such as a bird has passed through and caused the alarm; if yes, the long-focus camera 1 obtains the image of the super-high area of ​​the ship 15, and uses the ship image to train the neural network (use the neural network to judge the ship), segment the moving ship and measure the height; determine whether the height is super-high, if not, continue monitoring, if yes, an super-high alarm is issued, and its early warning AIS message transmitter 41, VHF voice broadcast device 42, early warning SMS transmitter 43, tweeter 44 and strong light searchlight 45 start working, the tweeter 44 and the strong light searchlight 45 give sound and light prompts to the super-high ship 15, and the early warning AIS message transmitter 41, VHF voice broadcast device 42, early warning SMS transmitter 43 push alarm information to the ship's wheelhouse, system management personnel (maintenance eyes of the early warning system) and relevant regulatory departments (Maritime Administration and bridge management department) respectively.

[0074] Embodiment 2:

[0075] The specific operation of the opposing pole-type long and short focal length camera array is as follows:

[0076] During installation, adjust the telephoto camera to be in a horizontal state, and use the camera attitude fine-tuning gimbal to adjust the optical centers of the two cameras to face each other (adjustment standard: the two sides appear at the center of the image (the center point position can be obtained by finding full screen pixels / 2 in the screenshot).); the elevation of the optical axis is the same as the height limit of the bridge.

[0077] 1. Short-focus camera identification and detection of ships

[0078] A neural network is trained with a large number of data sets of possible super-high parts such as inland ship masts and sand-carrying ship conveyor belts; a detection and recognition device for super-high parts of ships with excellent performance is obtained; Faster-RCNN is used as the network architecture for inland ship detection and recognition. This network architecture has a more comprehensive feature extraction of images, thereby having a higher recognition accuracy rate for target images.

[0079] By inputting the data set of inland river ships into the neural network, the weight of the neural network is adjusted, so that when the neural network is input with the picture of inland river ships again, it can accurately classify the objects in the picture as ships.

[0080] 2. Segmentation of moving ships based on KNN

[0081] Segmentation steps:

[0082] (1) Define a KNN background segmentor object; (2) Define a video object; (3) Read the network video stream; (4) Calculate the foreground mask; (5) Binarization operation; (6) Dilation operation; (7) Find the contour; (8) Contour screening; (9) Draw the contour (in the original image); (10) Traverse the image to get the highest point of segmentation; (11) Method for determining whether the telephoto camera is super-high.

[0083] 3. How to determine whether the telephoto camera is super high

[0084] like Figure 6 As shown in the figure, ultra-high detection is based on camera images. According to the basic principle of optics: light propagates in a straight line in the same uniform medium, usually referred to as linear propagation; it is an important foundation of geometric optics. The same object reflects natural light to the imaging surfaces of two cameras. For two cameras at the same elevation, if the object is imaged at the imaging center of the two cameras, it can be inferred that the elevation of the object imaging center is the same as the elevation of the optical center of the two cameras.

[0085] 4. Ultra-high measurement based on variant binoculars

[0086] The two cameras have the same elevation and their optical axes coincide. Take the pixel points from the optical center to the top of the pole in the left and right images respectively to get h r ,h l , according to the geometric relationship:

[0087] T l +T r =T (1)

[0088]

[0089]

[0090] From the above formulas (2) and (3), the target bar superelevation height ΔH can be obtained as:

[0091]

[0092] The schematic diagram of its principle is as follows Figure 7 and Figure 8 As shown; f r 、f l is the focal length of the left and right telephoto cameras; h r ,h l is the pixel point; T is the straight-line distance from the reference object to the camera.

[0093] 5. Camera calibration

[0094] Move the target so that the target is clearest in the field of view of the camera on one side, and use the telephoto camera to obtain the image of the target at this time, record the point coordinates of the position, and the upper and lower RTK elevation information of the target; the same steps are followed on the other side;

[0095] Using similar triangles to find f:

[0096]

[0097]

[0098] Where T is the straight-line distance from the reference object to the camera, which is accurately measured by the RTK measuring instrument; H is the elevation of the reference object, and h is the number of pixels occupied by the reference object in the image; Fig. 9 shown.

[0099] 6. Error measurement and optimization of binocular ultra-high altitude measurement system based on measurement error

[0100] (1) Extremely high judgment error

[0101] Determine superelevation error = |camera elevation - target pole elevation|

[0102] Since it is impossible to keep the two telephoto cameras at the same elevation in reality, a differential locator is used to measure the elevation of the target pole at the center of the two cameras and the camera elevation respectively. Through multiple measurements, the elevation difference is used to integrate and reduce the error of over-elevation judgment.

[0103] (2) Ultra-high measurement error

[0104] Superelevation measurement error = |actual elevation difference - measured elevation difference|

[0105] First of all, in reality, the camera has problems such as tilt and different elevations. Secondly, there is a certain numerical difference between the obtained camera focal length and the actual focal length. Therefore, these errors are specifically manifested in that the superelevation elevation calculated by the actual visual theory method is different from the actually measured superelevation elevation. In order to show these errors, a differential locator is used to measure the actual superelevation elevation of the superelevation object respectively. Through multiple measurements, the superelevation measurement error is used to integrate and reduce the error of the superelevation measurement system.

[0106] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A ship height measurement and overheight warning system for bridge area waters, characterized by: The invention comprises an ultra-high monitoring module (20), an ultra-high altitude measurement calculation module (30) and an ultra-high early warning module (40), wherein the ultra-high monitoring module (20) and the ultra-high early warning module (40) are both connected to the ultra-high altitude measurement calculation module (30); the ultra-high monitoring module (20) comprises a long- and short-focus camera array (10), an IP subnet (21) and a network cloud (22), wherein the long- and short-focus camera array (10) is connected to the network cloud (22) via the IP subnet (21); wherein the long- and short-focus camera array (10) is respectively installed on both sides of a river (11), and when installed, the optical centers of the two cameras are opposite to each other; The height measurement method of the superaltitude altimetry calculation module (30) comprises the following steps: S61. The telephoto cameras on both sides of the river simultaneously obtain images of the ship's ultra-high area, and take the pixel points from the optical center to the highest point of the ship in the two images to obtain h r ,h l ; S62, after the cameras are installed, the positions of the two telephoto cameras are accurately measured by RTK to obtain a baseline distance T between the two cameras; S63, the focal lengths f of the two telephoto cameras are obtained by the large scene camera internal parameter calibration method r 、f l ; S64, by pixel h r ,h l , the baseline distance T between the two cameras, and the focal lengths of the two cameras are used to calculate the super-high height ΔH:

2. A ship height measurement and overheight warning system for bridge area waters as claimed in claim 1, characterized in that: The long-focus and short-focus camera array (10) comprises a long-focus camera (1), a short-focus camera (2), a POE switch (4), a vertical pole (5) and a control box (6); the control box (6) is mounted on the vertical pole (5); the long-focus camera (1), the short-focus camera (2) and the POE switch (4) are all mounted in the control box (6); and a lens attitude fine-tuning gimbal (3) is mounted at the bottom of the long-focus camera (1).

3. The ship height measurement and overheight warning system for bridge area waters as claimed in claim 1, characterized in that: The superelevation height measurement calculation module (30) comprises an inland river ship face feature library (31), a ship face recognition module (32), a ship high point recognition module (33), a ship height measurement calculation module (34), a bridge clear height limit processing module (35) and an integrated server (36); the inland river ship face feature library (31) is connected to the integrated server (36) via the ship face recognition module (32); the ship high point recognition module (33), the ship height measurement calculation module (34) and the bridge clear height limit processing module (35) are all electrically connected to the integrated server (36).

4. The ship height measurement and overheight warning system for bridge area waters as claimed in claim 1, characterized in that: The super-altitude warning module (40) comprises a warning AIS message transmitting device (41), a VHF voice broadcasting device (42), a warning text message transmitting device (43), a tweeter (44) and a strong light searchlight (45).

5. The ship height measurement and overheight warning system for bridge area waters as claimed in claim 1, characterized in that: Its working steps are: S1. The short-focus camera obtains the water surface image, and then uses the deep learning neural network to detect and identify whether there is a ship in the image. The obtained image is used as a training set for the ship mast recognition neural network to enhance the network's detection and recognition capabilities; S2, training the neural network to obtain a ship detection identifier with good recognition performance; S3, when the short-focus camera detects and identifies a ship, the long-focus camera is used to segment the moving ship; S4, accurately measuring the identified ship mast using imagery; S5. Feedback whether the system measurement result is over-height and the measured height to the ship and traffic management system.

6. The ship height measurement and overheight warning system facing the bridge area water area as claimed in claim 1, characterized in that: It also includes a large-scene camera internal parameter calibration method, the method steps of which are: using an RTK measuring instrument to obtain the elevation information of a reference object and the image information of the reference object, and using the formula of similar triangles to calculate the focal length f of the camera: Where T is the straight-line distance from the reference object to the camera, which is accurately measured by the RTK measuring instrument; H is the elevation of the reference object, and h is the number of pixels occupied by the reference object in the image.

7. The ship height measurement and overheight warning system facing the bridge area water area as claimed in claim 1 is characterized by: It also includes a graded warning method, which is as follows: when the short-focus camera obtains the image of the ship, a prompt will be given in the background to remind the platform to be prepared in advance; after the long- and short-focus camera array measures the height of the ship, if it is too high, a strong searchlight and voice broadcast will be used to remind the ship, and the background will remind the system manager that the ship is too high, and send a text message to the relevant regulatory department, thereby reminding the incoming ships.

8. The ship height measurement and overheight warning system facing the bridge area water area as claimed in claim 1, characterized in that: It also includes a hierarchical recognition and detection method using a combination of long- and short-focus cameras; the method is that a short-focus camera acquires a water surface image to detect and identify a ship, and when a ship is detected passing by, a long-focus camera is used to segment the ship's super-elevated area to obtain the ship's highest point, and then visual measurement is performed.

Images