A ship body identification method, device and electronic equipment

Abstract

The application discloses a ship body identification method and device and electronic equipment. The method obtains point cloud data of a target region and a target compression ratio. The point cloud data of the target region is compressed along a direction perpendicular to a water surface and a length direction of a river channel according to the target compression ratio, and compressed target region point cloud is obtained. The ship body contour in the point cloud data is identified based on the compressed target region point cloud. The technical scheme provided by the application can improve the accuracy of ship body contour identification to a certain extent.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of ship identification, and particularly relates to a ship body identification method and device and electronic equipment. BACKGROUND

[0002] In ship detection, a voxel grid is divided, and a convolutional neural network is used to extract features. The calculation complexity of this method is very high, and it is difficult to be practically applied. The method using projection will cause information loss.

[0003] Using a clustering method, a threshold is set to determine whether the clustering is a ship. When the clustering radius is small, a ship is easily clustered into several categories. When the clustering radius is large, two ships that are close to each other are easily clustered into one ship. The main reason is that when the laser radar is far away from the ship, part of the collected ship point cloud data will be blocked, so that a large number of point clouds in the middle of the ship body will be missing, and thus the error in the direct point cloud recognition process will be large. SUMMARY

[0004] Embodiments of the present application provide a ship body identification method, device and electronic equipment, thereby improving the accuracy of ship contour identification to at least some extent.

[0005] Other characteristics and advantages of the present application will become apparent from the following detailed description, or will be learned by practice of the present application.

[0006] According to a first aspect of an embodiment of the present application, a ship body identification method is provided, the method comprising:

[0007] Obtaining point cloud data of a target region, and obtaining a target compression ratio;

[0008] According to the target compression ratio, the point cloud data of the target region is compressed along a direction perpendicular to the water surface and a length direction of the river channel, to obtain compressed target region point cloud;

[0009] Based on the compressed target region point cloud, a ship body contour in the point cloud data is identified.

[0010] In some embodiments of the present application, based on the foregoing scheme, the target compression ratio is obtained, comprising:

[0011] Obtaining a discrete value of a reference ship body point cloud of a reference ship body in a ship body width direction, a ship body length direction, and a ship body height direction, the discrete value being used to represent the density of the ship body point cloud in the corresponding direction;

[0012] According to the discrete value of the reference ship body point cloud in the ship body width direction, the ship body length direction, and the ship body height direction, a candidate compression ratio is determined.

[0013] compress the reference hull point cloud in the hull width direction and the hull length direction according to the candidate compression ratio, to obtain a compressed reference hull point cloud;

[0014] If the discrete values of the compressed reference hull point cloud in the hull width direction, the hull length direction and the hull height direction are approximately equal, the candidate compression ratio is taken as the target compression ratio.

[0015] In some embodiments of the present application, based on the foregoing scheme, the discrete values of the reference hull point cloud of the reference hull in the hull width direction, the hull length direction and the hull height direction include:

[0016] Obtaining the hull point cloud of the at least one reference hull marked in advance, to obtain at least one reference hull point cloud;

[0017] Obtaining the standard deviation of the point cloud coordinate values in the hull width direction, the standard deviation of the point cloud coordinate values in the hull length direction and the standard deviation of the point cloud coordinate values in the hull height direction of each reference hull point cloud;

[0018] Setting the standard deviation of the point cloud coordinate values in the hull width direction of each reference hull point cloud as unit 1, as the reference discrete value of each reference hull point cloud in the hull width direction, and based on the standard deviation of the point cloud coordinate values in the hull length direction and the standard deviation of the point cloud coordinate values in the hull height direction of each reference hull point cloud, converting the reference discrete value in the hull length direction and the reference discrete value in the hull height direction of each reference hull point cloud;

[0019] Calculating the average value of the reference discrete value in the hull width direction of each reference hull point cloud, the average value of the reference discrete value in the hull length direction of each reference hull point cloud and the average value of the reference discrete value in the hull height direction of each reference hull point cloud, as the discrete values of the reference hull point cloud of the reference hull in the hull width direction, the hull length direction and the hull height direction, respectively.

[0020] In some embodiments of the present application, based on the foregoing scheme, the candidate compression ratio is determined according to the discrete values of the reference hull point cloud in the hull width direction, the hull length direction and the hull height direction, including:

[0021] Taking the ratio of the discrete value of the reference hull point cloud in the hull width direction to the discrete value in the hull length direction as the candidate compression ratio of the reference hull point cloud in the hull length direction;

[0022] A ratio of a discrete value of the reference hull point cloud in a hull width direction to a discrete value in a hull height direction is taken as a candidate compression ratio of the reference hull point cloud in the hull height direction.

[0023] In some embodiments of the present application, based on the foregoing scheme, the identifying the hull contour in the point cloud data based on the compressed target region point cloud comprises:

[0024] Obtaining a preset clustering radius;

[0025] Performing hull clustering on the compressed target region point cloud according to the clustering radius to identify a hull contour in the compressed target region point cloud;

[0026] Based on the hull contour in the compressed target region point cloud, identifying the hull contour in the point cloud data through the mapping relationship between the compressed target region point cloud and the point cloud data.

[0027] In some embodiments of the present application, based on the foregoing scheme, after identifying the hull contour in the point cloud data, the method further comprises:

[0028] Performing plane fitting on the point cloud belonging to the hull contour to obtain a plurality of groups of plane point clouds;

[0029] Selecting a plane point cloud satisfying a preset condition from the plurality of groups of plane point clouds as a candidate plane point cloud, wherein the preset condition is that a face included angle between a plane on which the plane point cloud is located and a horizontal plane is less than a preset threshold;

[0030] Calculating an area of a plane constituted by each candidate plane point cloud, and selecting a candidate plane point cloud with the largest area as a hull deck point cloud;

[0031] Calculating a deck height according to the hull deck point cloud, and taking the deck height as a sheer height.

[0032] In some embodiments of the present application, based on the foregoing scheme, the method further comprises:

[0033] Determining a reference point in the point cloud belonging to the hull contour, and recording a moving distance and a moving direction of the reference point within a set time;

[0034] Taking a ratio between the moving distance and the set time as a hull speed, and taking the moving direction as a hull heading;

[0035] In a direction perpendicular to the hull heading, taking a longest point cloud distance of the point cloud belonging to the hull contour and located on a horizontal plane as a hull width.

[0036] In some embodiments of the present application, based on the foregoing scheme, the method further comprises:

[0037] determining a reference line;

[0038] acquiring a moving time of a point cloud belonging to the hull profile from contacting the reference line to leaving the reference line;

[0039] multiplying the moving time by the hull speed as the hull length.

[0040] According to a second aspect of embodiments of the present application, a hull recognition device is provided, the device comprising:

[0041] an acquisition unit configured to acquire point cloud data of a target region and a target compression ratio;

[0042] a compression unit configured to compress the point cloud data of the target region along a direction perpendicular to a water surface and a length direction of a river channel according to the target compression ratio to obtain compressed target region point cloud;

[0043] a recognition unit configured to recognize a hull profile in the point cloud data based on the compressed target region point cloud.

[0044] According to a third aspect of embodiments of the present application, a hull recognition electronic device is provided, the hull recognition electronic device comprising one or more processors and one or more memories, the one or more memories storing at least one piece of program code, the at least one piece of program code being loaded and executed by the one or more processors to implement the method according to any one of the preceding claims.

[0045] In the present application, in the process of recognizing a hull profile by hull point cloud data, there is some missing data in the actually collected hull point cloud data, for example, in the hull length or height direction, the point cloud displayed by some local hull is blank, and this part of missing data will affect the accuracy of the overall recognition of the hull profile. According to some embodiments of the present application, by acquiring a target compression ratio, the point cloud data along a direction perpendicular to a water surface and a length direction of a river channel is compressed, the target hull point cloud obtained by compression can reduce the range of point cloud blank (i.e. missing data), thereby reducing the negative impact of missing data on the overall recognition of the hull profile, and therefore the present application can improve the accuracy of hull profile recognition to a certain extent.

[0046] The beneficial effects of the above-mentioned second aspect to third aspect embodiments can refer to the beneficial effects of the above-mentioned first aspect and the beneficial effects of each embodiment of the first aspect, which will not be repeated here.

[0047] It should be understood that the foregoing general description and the following detailed description are only examples and explanatory, and are not restrictive of the application. BRIEF DESCRIPTION OF DRAWINGS

[0048] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and serve to explain the principles of the application. It is apparent that the accompanying drawings are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor on the basis of these drawings. In the drawings:

[0049] Figure 1 A schematic diagram of a hull point cloud is shown;

[0050] Figure 2 A flow chart of hull recognition in the embodiments of the present application is shown;

[0051] Figure 3 A flow chart of obtaining a target compression ratio in the embodiments of the present application is shown;

[0052] Figure 4 A detailed flow chart of obtaining discrete values of a reference hull point cloud in the hull width direction, the hull length direction, and the hull height direction of a reference hull in the embodiments of the present application is shown;

[0053] Figure 5 A detailed flow chart of determining a candidate compression ratio according to the discrete values of the reference hull point cloud in the hull width direction, the hull length direction, and the hull height direction in the embodiments of the present application is shown;

[0054] Figure 6 A detailed flow chart of recognizing a hull contour in the point cloud data based on the compressed target region point cloud in the embodiments of the present application is shown;

[0055] Figure 7 A block diagram of a hull recognition device in the embodiments of the present application is shown;

[0056] Figure 8 A structural schematic diagram of a hull recognition electronic device in the embodiments of the present application is shown. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be described clearly and completely below in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0058] Furthermore, the described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the application. One skilled in the relevant art will recognize, however, that the application can be practiced without one or more of the specific details, or with other methods, components, materials, and so forth. In other instances, well-known structures, devices, implementations, or operations are not shown or described in detail to avoid obscuring aspects of the application.

[0059] The block diagrams in the drawings show only the functionality of the features and can not imply that the functionality must be implemented in the order shown. Some of the features can be implemented in hardware, software, firmware in one or more computer systems or components thereof.

[0060] The flow diagrams shown in the various figures, which can typically be executed

[0061] In the description of the application, it should be understood that the terms "first", "second", etc. are used only to describe different features, and do not imply or suggest relative importance or a number of the features indicated. Thus, the features defined with "first", "second" can explicitly or implicitly include one or more of the features. In the description of the application, unless otherwise specified, the meaning of "a plurality" is two or more.

[0062] In order for those skilled in the art to better understand the background of the present application in the process of the present inventors proposing the present application, the following will be described in conjunction with Figure 1 Brief Description of the Drawings

[0063] With reference to Figure 1 A schematic diagram 100 of a ship body point cloud is shown. In the process of identifying a ship, the following two situations are often encountered: because the point cloud distance on the hull profile of two different ships (A and B in the middle) close to each other is too close, the point cloud data of the two ships is easily identified as one ship during identification; or because the same ship (B in the middle) is partially (such as the middle) blocked (or other reasons) and the local point cloud data is missing, the point cloud data originally belonging to one ship (B in the middle) is easily identified as two ships (A and B in the middle) during identification. Figure 1 Figure 1 Figure 1 Figure 1 ​​​In order to solve the above technical problems, a new ship body identification method needs to be provided.

[0064] The implementation details of the technical solutions of the embodiments of the present application are described below in detail:

[0065] Referring to Figure 2 , a flowchart of a ship body identification method according to an embodiment of the present application is shown, which can be executed by a device with computing processing function. The ship body identification method includes at least steps S1 to S3, which are described in detail as follows:

[0066] In S1, point cloud data of a target region is obtained, and a target compression ratio is obtained.

[0067] In some embodiments of the present application, the target compression ratio can be obtained according to the steps shown in Figure 3

[0068] Referring to Figure 3 , a flowchart of obtaining a target compression ratio according to an embodiment of the present application is shown, which includes steps S11 to S14:

[0069] S11, obtaining discrete values of reference ship body point cloud of a reference ship body in the ship body width direction, the ship body length direction, and the ship body height direction, the discrete values being used to represent the density of the ship body point cloud in the corresponding direction;

[0070] S12, determining a candidate compression ratio according to the discrete values of the reference ship body point cloud in the ship body width direction, the ship body length direction, and the ship body height direction;

[0071] S13, compressing the reference ship body point cloud in the ship body height direction and the ship body length direction according to the candidate compression ratio, to obtain compressed reference ship body point cloud;

[0072] S14, if the discrete values of the compressed reference ship body point cloud in the ship body width direction, the ship body length direction, and the ship body height direction are approximately equal, the candidate compression ratio is taken as the target compression ratio.

[0073] It should be noted that the scale of "approximately equal" mentioned in step S14 can be determined according to the actual needs of the user, for example, in the discrete values of the ship body point cloud in the ship body width direction, the ship body length direction, and the ship body height direction, the difference between any two discrete values does not exceed a preset threshold, which can meet the "approximately equal".

[0074] ​It should be noted that the interpretation of the reference ship point cloud of the reference ship is as follows: the ship body point cloud of a ship in the target area or outside the target area and the application scene is adapted to each other, and if the ship body point cloud is compressed according to the candidate compression ratio, the discrete values in the ship body width direction, the ship body length direction and the ship body height direction are approximately equal, then the candidate compression ratio is determined as the target compression ratio for compressing the point cloud data of the target area.

[0075] In some embodiments of the present application, the discrete values of the reference ship point cloud of the reference ship in the ship body width direction, the ship body length direction and the ship body height direction can be obtained according to the steps shown in the following figure. Figure 4

[0076] Referring to Figure 4 , a detailed flowchart for obtaining the discrete values of the reference ship point cloud of the reference ship in the ship body width direction, the ship body length direction and the ship body height direction in the embodiments of the present application is shown, which specifically includes steps S111 to S114:

[0077] S111, obtaining the ship body point cloud of at least one reference ship marked in advance to obtain at least one reference ship point cloud;

[0078] S112, obtaining the standard deviation of the point cloud coordinate values in the ship body width direction, the standard deviation of the point cloud coordinate values in the ship body length direction and the standard deviation of the point cloud coordinate values in the ship body height direction of each reference ship point cloud;

[0079] S113, setting the standard deviation of the point cloud coordinate values in the ship body width direction of each reference ship point cloud as unit 1 as the reference discrete value of each reference ship point cloud in the ship body width direction, and converting the reference discrete value of each reference ship point cloud in the ship body length direction and the reference discrete value of each reference ship point cloud in the ship body height direction based on the standard deviation of the point cloud coordinate values in the ship body length direction and the standard deviation of the point cloud coordinate values in the ship body height direction of each reference ship point cloud;

[0080] S114, calculating the average value of the reference discrete value of each reference ship point cloud in the ship body width direction, the average value of the reference discrete value of each reference ship point cloud in the ship body length direction and the average value of the reference discrete value of each reference ship point cloud in the ship body height direction as the discrete value of the reference ship point cloud of the reference ship in the ship body width direction, the ship body length direction and the ship body height direction, respectively.

[0081] It should be noted that the reference ship point cloud of the reference ship obtained in step S111 can be one, multiple or all, and the more reference ship point clouds obtained, the more accurate the target compression ratio value obtained, which is more suitable for the present use scenario.

[0082] ​It should be noted that in step S113, "setting the standard deviation of the point cloud coordinate value of each reference ship body point cloud in the ship body width direction as unit 1 as the reference dispersion value of each reference ship body point cloud in the ship body width direction, and converting the reference dispersion value of each reference ship body point cloud in the ship body length direction and the reference dispersion value in the ship body height direction based on the standard deviation of the point cloud coordinate value of each reference ship body point cloud in the ship body length direction and the standard deviation of the point cloud coordinate value in the ship body height direction", for example, if the standard deviation of the point cloud coordinate value of a certain reference ship body point cloud in the ship body width direction, the standard deviation of the point cloud coordinate value in the ship body length direction, and the standard deviation of the point cloud coordinate value in the ship body height direction are 2, 4, and 6 respectively, through step S113, the reference dispersion value of the point cloud coordinate value of the reference ship body point cloud in the ship body width direction, the reference dispersion value of the point cloud coordinate value in the ship body length direction, and the reference dispersion value of the point cloud coordinate value in the ship body height direction can be obtained. In this application, the standard deviation of the point cloud coordinate value of each reference ship body point cloud in the ship body width direction is set as unit 1, which has the advantage of facilitating the subsequent calculation of the average value of each reference dispersion value.

[0083] In other embodiments of the present application, the reference dispersion value of the point cloud coordinate value of the reference ship body point cloud in the ship body width direction, the reference dispersion value of the point cloud coordinate value in the ship body length direction, and the reference dispersion value of the point cloud coordinate value in the ship body height direction can also be determined based on the variance of the point cloud coordinate value of the reference ship body point cloud in the ship body width direction, the variance of the point cloud coordinate value in the ship body length direction, and the variance of the point cloud coordinate value in the ship body height direction. It can be understood that there are many ways to determine the reference dispersion value of the reference ship body point cloud in each ship body direction, and they are not limited to those exemplified above.

[0084] It should be understood that the above steps obtain the dispersion values in each direction of the reference ship body point cloud of the reference ship body most accurately conforming to the identification scene in the target area by averaging the standard deviations in different dimensional directions of all ships in the target area. Different identification scenes include fishing boats at the seaside, large cargo ships, steamships, etc. Therefore, the dispersion values in different directions of ships with different functions and scenes are different.

[0085] In some embodiments of the present application, the candidate compression ratio is determined according to the dispersion values of the reference ship body point cloud in the ship body width direction, the ship body length direction, and the ship body height direction, which can be performed according to the steps shown in Figure 5

[0086] Referring to Figure 5 ​, a detailed flowchart of determining the candidate compression ratio according to the discrete values of the reference hull point cloud in the hull width direction, the hull length direction and the hull height direction is shown, and specifically includes steps S121 to S122:

[0087] S121, taking the ratio of the discrete value of the reference hull point cloud in the hull width direction to the discrete value in the hull length direction as the candidate compression ratio of the reference hull point cloud in the hull length direction;

[0088] S122, taking the ratio of the discrete value of the reference hull point cloud in the hull width direction to the discrete value in the hull height direction as the candidate compression ratio of the reference hull point cloud in the hull height direction.

[0089] It should be understood that the candidate compression ratio is determined by the ratio, which is more accurate and faster than slowly adjusting the compression ratio.

[0090] The above proposes specific determination details of the target compression ratio. Of course, the determination of the target compression ratio is not limited to the above determination by using the ratio. It should be noted that the target compression ratio can be determined directly in addition to the above-mentioned manner. The target compression ratio can be calculated artificially or can be calculated according to the point cloud data outside the target region.

[0091] In the present application, the way to obtain the target compression ratio can be various and is not limited to those mentioned above.

[0092] Continue to refer to Figure 2 In S2, the point cloud data of the target region is compressed along the direction perpendicular to the water surface and the length direction of the river channel according to the target compression ratio, and the compressed target region point cloud is obtained.

[0093] It should be noted that the "direction perpendicular to the water surface" in step S2 corresponds to the "candidate compression ratio of the reference hull point cloud in the hull height direction" in step S121, and the "length direction of the river channel" in step S2 corresponds to the "candidate compression ratio of the reference hull point cloud in the hull length direction" in step S122.

[0094] In the present application, by compressing the point cloud data in the direction perpendicular to the water surface (that is, the height direction of most ship bodies) and the length direction of the river channel (that is, the length direction of most ship bodies arranged on the river channel) according to the target compression ratio, the influence of point cloud missing in the original ship body point cloud on the overall recognition of the ship body can be reduced, and the accuracy of recognizing the ship body contour by the obtained target region point cloud after compression in the subsequent step can be improved. At the same time, since the point cloud data in the ship body width direction (that is, the river channel width) is not compressed, the recognition accuracy of multiple ship bodies in the parallel condition will not be affected.

[0095] With reference to the foregoing Figure 2 In S3, the ship body contour in the point cloud data is recognized based on the compressed target region point cloud.

[0096] In some embodiments of the present application, recognizing the ship body contour in the point cloud data based on the compressed target region point cloud can be performed according to the steps shown in Figure 6

[0097] With reference to the foregoing Figure 6 , a detailed flowchart for recognizing the ship body contour in the point cloud data based on the compressed target region point cloud in the embodiments of the present application is shown, which specifically includes steps S31 to S33:

[0098] S31, a preset clustering radius is obtained;

[0099] S32, ship clustering is performed based on the compressed target region point cloud according to the clustering radius, so as to recognize the ship body contour in the compressed target region point cloud;

[0100] S33, the ship body contour in the point cloud data is recognized based on the ship body contour in the compressed target region point cloud through the mapping relationship between the target ship body point cloud and the point cloud data.

[0101] It should be noted that in step S31, the "clustering radius" can be set to 1 / 4 of the ship width, or 2 / 5 of the ship width. In the present application, the size of the clustering radius can be determined according to actual needs, and the present application does not make specific limitations thereon.

[0102] ​It should be noted that in step S32, the interpretation of "clustering the compressed target region point cloud according to the clustering radius" is that if the distance between any two point clouds belonging to the ship point cloud or near the contour is less than the preset clustering radius, the two point clouds are clustered into the same ship point cloud. Since the original ship point cloud is compressed in the length direction, even if the compressed ship point cloud is missing local point cloud data, the distance between any two point clouds in the compressed ship contour is less than the preset clustering radius, so they are clustered into the same ship point cloud. Therefore, the influence of missing point cloud data in the middle of the ship on the actual ship contour recognition is reduced, so it is difficult to identify a ship point cloud with missing middle point cloud data as two ships; and since the two adjacent ship point clouds are compressed in the length direction, the distance between the two ships in the width direction is not compressed, so the distance between the two ships in the width direction is greater than the preset clustering radius, so the two ship point clouds are difficult to be identified as the same ship point cloud, so the result of identifying the target ship point cloud is more accurate.

[0103] It should be noted that in step S33, the "mapping relationship between the compressed target region point cloud and the point cloud data" can be the index relationship between the ship contour in the pre-set point cloud data before compression and the target region point cloud after compression, or it can be manually matched through a matching table.

[0104] It should be understood that the identification method is not limited to clustering alone, and there are many other ways, and this step is only to protect the identification of point clouds belonging to the same ship during clustering identification.

[0105] The beneficial effects that can be achieved by the embodiments of the application are: in the process of identifying the ship contour through the ship point cloud data, since there is some missing data in the actual collected ship point cloud data, for example, in the length or height direction of the ship, some local point clouds of the ship are blank, and this part of missing data will affect the accuracy of the overall identification of the ship contour. Through the target compression ratio obtained by the embodiments of the application, the target ship point cloud obtained by compressing the point cloud data of the target region along the direction perpendicular to the water surface and the length direction of the river can reduce the range of point cloud blank (i.e. missing data), thereby reducing the negative influence of missing data on the overall identification of the ship contour. Therefore, the application can improve the accuracy of ship contour identification to a certain extent.

[0106] In some embodiments of the application, after identifying the ship contour in the point cloud data, the method further comprises:

[0107] Fitting a plane to the point cloud belonging to the ship contour to obtain a plurality of groups of plane point clouds;

[0108] selecting, from the plurality of sets of planar point clouds, a planar point cloud satisfying a preset condition as a candidate planar point cloud, wherein the preset condition is that an included angle between a plane of the planar point cloud and a horizontal plane is less than a preset threshold value;

[0109] calculating an area of a plane formed by each candidate planar point cloud, and selecting a candidate planar point cloud with a largest area as a hull deck point cloud;

[0110] calculating a deck height according to the hull deck point cloud, and taking the deck height as a ship side height.

[0111] It should be noted that the "preset threshold value" in the step "selecting, from the plurality of sets of planar point clouds, a planar point cloud satisfying a preset condition as a candidate planar point cloud, wherein the preset condition is that an included angle between a plane of the planar point cloud and a horizontal plane is less than a preset threshold value" can be 30 degrees. The purpose of this step is to screen out all the point clouds of the planes located in the horizontal plane, and discard the point clouds of the hull side.

[0112] It should be understood that, because the top of the hull cabin also has a small area of point clouds located in the horizontal plane, the part of point clouds needs to be removed, so it is necessary to screen out the point cloud with the largest area from all the point clouds located in the horizontal plane as the height of the deck and the height of the ship side.

[0113] In some embodiments of the present application, the method further comprises:

[0114] determining a reference point in the point cloud belonging to the hull contour, and recording a moving distance and a moving direction of the reference point within a set time;

[0115] taking a ratio between the moving distance and the set time as a hull speed, and taking the moving direction as a hull heading;

[0116] in a direction perpendicular to the hull heading, taking a longest point cloud distance of the point cloud belonging to the hull contour and located in the horizontal plane as a hull width.

[0117] Specifically, the length direction of the ship can be obtained by taking one of the hull point clouds as a reference point cloud and measuring the heading of the movement of the reference point cloud within the set time. Based on the length direction of the ship and the hull point cloud, the length, width and other information of the hull can be obtained.

[0118] It should be understood that, based on the heading of the hull, a direction perpendicular to the hull heading of the ship is obtained, and then the direction located in the horizontal plane in the direction perpendicular to the hull heading is screened out as the width direction of the hull, and the longest point cloud distance of the point cloud belonging to the hull contour in the width direction of the hull is taken as the ship width. This process can directly obtain the accurate ship width value from the measured point cloud data on the premise that the point cloud data is not missing, without relying on other measurement data, which is relatively convenient and fast.

[0119] In some embodiments of the present application, the method further comprises:

[0120] determining a reference line;

[0121] acquiring a moving time of the point cloud belonging to the hull contour from the time of contacting the reference line to the time of leaving the reference line;

[0122] multiplying the moving time by the hull speed to obtain the length of the hull.

[0123] Specifically, a reference line can be set to track the bow when the bow passes through the detection line. The average speed of the ship and the running direction of the ship can be calculated by the displacement of the bow in dozens of frames, so as to obtain the time of the hull from contacting the reference line to leaving the reference line and the speed, and multiply the time of the hull from contacting the reference line to leaving the reference line by the speed to obtain the actual length of the hull, so that the length of the hull can be obtained.

[0124] It can be understood that in the above method of obtaining the length of the hull, only the moving time of the hull from contacting the reference line to leaving the reference line and the hull speed are needed to obtain the length, and the hull point cloud data which may be missing is not needed, compared with the calculation of length, width and height by relying on the hull point cloud data and the heading, the length and width of the ship can be obtained by the speed in the case of incomplete ship point cloud, so that the calculation is more accurate.

[0125] The device embodiments of the present application are introduced below, which can be used to execute the hull recognition method in the above embodiments of the present application. For details not disclosed in the device embodiments of the present application, please refer to the above embodiments of the hull recognition method of the present application.

[0126] Referring to Figure 7 , a block diagram of a hull recognition device in an embodiment of the present application is shown, which comprises an acquisition unit 1101, a compression unit 1102 and a recognition unit 1103.

[0127] The acquisition unit 1101 is configured to acquire point cloud data of a target region and a target compression ratio; the compression unit 1102 is configured to compress the point cloud data of the target region along a direction perpendicular to the water surface and a length direction of the river channel according to the target compression ratio to obtain compressed target region point cloud; and the recognition unit 1103 is configured to recognize a hull contour in the point cloud data based on the compressed target region point cloud.

[0128] In some embodiments of the present application, based on the foregoing scheme, the acquisition unit 1101 is configured to: acquire discrete values of the reference hull point cloud of the reference hull in the hull width direction, the hull length direction, and the hull height direction, the discrete values being used to represent the density of the hull point cloud in the corresponding direction; determine a candidate compression ratio according to the discrete values of the reference hull point cloud in the hull width direction, the hull length direction, and the hull height direction; compress the reference hull point cloud in the hull height direction and the hull length direction according to the candidate compression ratio to obtain a compressed reference hull point cloud; and if the discrete values of the compressed reference hull point cloud in the hull width direction, the hull length direction, and the hull height direction are approximately equal, the candidate compression ratio is taken as the target compression ratio.

[0129] In some embodiments of the present application, based on the foregoing scheme, the acquisition unit 1101 is further configured to: acquire hull point clouds of at least one reference hull marked in advance to obtain at least one reference hull point cloud; acquire the standard deviation of the point cloud coordinate values of each reference hull point cloud in the hull width direction, the standard deviation of the point cloud coordinate values in the hull length direction, and the standard deviation of the point cloud coordinate values in the hull height direction; set the standard deviation of the point cloud coordinate values of each reference hull point cloud in the hull width direction as unit 1 as the reference discrete value of each reference hull point cloud in the hull width direction, and convert the reference discrete value of each reference hull point cloud in the hull length direction and the reference discrete value in the hull height direction based on the standard deviation of the point cloud coordinate values of each reference hull point cloud in the hull length direction and the standard deviation of the point cloud coordinate values in the hull height direction; calculate the average of the reference discrete values of each reference hull point cloud in the hull width direction, the average of the reference discrete values of each reference hull point cloud in the hull length direction, and the average of the reference discrete values of each reference hull point cloud in the hull height direction, respectively, as the discrete values of the reference hull point cloud of the reference hull in the hull width direction, the hull length direction, and the hull height direction.

[0130] In some embodiments of this application, based on the foregoing scheme, the acquisition unit 1101 is further configured to: acquire hull point clouds of at least one pre-marked reference hull, thereby obtaining at least one reference hull point cloud; acquire the standard deviation of the point cloud coordinate values ​​of each reference hull point cloud in the hull width direction, the standard deviation of the point cloud coordinate values ​​in the hull length direction, and the standard deviation of the point cloud coordinate values ​​in the hull height direction; set the standard deviation of the point cloud coordinate values ​​of each reference hull point cloud in the hull width direction to a unit of 1, as a reference discrete value of each reference hull point cloud in the hull width direction, and based on the standard deviation of the point cloud coordinate values ​​of each reference hull point cloud in the hull height direction... The standard deviations of the point cloud coordinates along the length direction and the height direction are used to convert the reference discrete values ​​of each reference hull point cloud along the length and height directions. The average values ​​of the reference discrete values ​​along the width direction, length direction, and height direction of each reference hull point cloud are calculated and used as the discrete values ​​of the reference hull point cloud along the width, length, and height directions, respectively.

[0131] In some embodiments of this application, based on the foregoing scheme, the acquisition unit 1101 is further configured to: use the ratio of the discrete value of the reference hull point cloud in the hull width direction to the discrete value in the hull length direction as a candidate compression ratio of the reference hull point cloud in the hull length direction; and use the ratio of the discrete value of the reference hull point cloud in the hull width direction to the discrete value in the hull height direction as a candidate compression ratio of the reference hull point cloud in the hull height direction.

[0132] In some embodiments of this application, based on the aforementioned scheme, the identification unit 1103 is configured to: obtain a preset clustering radius; perform ship clustering on the target ship point cloud according to the clustering radius to identify the ship outline in the target ship point cloud; and identify the ship outline in the point cloud data based on the ship outline in the compressed target region point cloud through the mapping relationship between the compressed target region point cloud and the point cloud data.

[0133] Based on the same inventive concept, a third aspect of the embodiments of this application also provides a ship hull identification electronic device, referencing... Figure 8 The diagram shows a schematic of the structure of a ship identification electronic device according to an embodiment of this application. The ship identification electronic device includes one or more memories 1204, one or more processors 1202, and at least one computer program (program code) stored in the memory 1204 and executable on the processor 1202. When the processor 1202 executes the computer program, it implements the ship identification method of the first aspect mentioned above.

[0134] Among them, Figure 8In this document, a bus architecture (represented by bus 1200) is used. Bus 1200 may include any number of interconnected buses and bridges, linking various circuits including one or more processors represented by processor 1202 and memory represented by memory 1204. Bus 1200 may also link various other circuits such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. Bus interface 1205 provides an interface between bus 1200 and receiver 1201 and transmitter 1203. Receiver 1201 and transmitter 1203 may be the same element, i.e., a transceiver, providing a unit for communicating with various other devices over a transmission medium. Processor 1202 is responsible for managing bus 1200 and general processing, while memory 1204 can be used to store data used by processor 1202 during operation.

[0135] The beneficial effects of the embodiments of the second to third aspects described above can be referred to the beneficial effects of the first aspect and the embodiments of the first aspect described above, and will not be repeated here.

[0136] The functions described herein may be implemented in hardware, software executed by a processor, firmware, or any combination thereof. If implemented in software executed by a processor, the functions may be stored as one or more instructions or codes on or transmitted via a computer-readable medium. Other examples and embodiments are within the scope and spirit of this application and the appended claims. For example, due to the nature of software, the functions described above may be implemented using software executed by a processor, hardware, firmware, hardwired, or any combination thereof. Furthermore, the functional units may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit.

[0137] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0138] The units described as separate components may or may not be physically separate. Similarly, the components of the control device may or may not be physical units; they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0139] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0140] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of the claims of this application.

Claims

1. A method for identifying ship hulls, characterized in that, include: The process involves acquiring point cloud data of the target region and obtaining discrete values ​​of the reference hull point cloud in the width, length, and height directions of the hull. These discrete values ​​characterize the density of the hull point cloud in the corresponding directions. A candidate compression ratio is determined based on these discrete values. The reference hull point cloud is then compressed in the height and length directions according to the candidate compression ratio to obtain a compressed reference hull point cloud. If the discrete values ​​of the compressed reference hull point cloud in the width, length, and height directions are approximately equal, the candidate compression ratio is used as the target compression ratio. The approximate equality is satisfied when the difference between any two discrete values ​​in the width, length, and height directions does not exceed a preset threshold. According to the target compression ratio, the point cloud data of the target area is compressed along the direction perpendicular to the water surface and the length of the river channel to obtain the compressed point cloud of the target area. Based on the compressed target region point cloud, the ship hull outline in the point cloud data is identified.

2. The method according to claim 1, characterized in that, The process of obtaining the discrete values ​​of the reference hull point cloud in the hull width direction, hull length direction, and hull height direction includes: Obtain the hull point cloud of at least one pre-labeled reference hull to obtain at least one reference hull point cloud; Obtain the standard deviation of the point cloud coordinates of each reference hull in the hull width direction, the standard deviation of the point cloud coordinates in the hull length direction, and the standard deviation of the point cloud coordinates in the hull height direction. The standard deviation of the point cloud coordinates of each reference hull point cloud in the hull width direction is set to 1, which is used as the reference discrete value of each reference hull point cloud in the hull width direction. Based on the standard deviation of the point cloud coordinates of each reference hull point cloud in the hull length direction and the standard deviation of the point cloud coordinates in the hull height direction, the reference discrete values ​​of each reference hull point cloud in the hull length direction and the reference discrete values ​​in the hull height direction are converted. Calculate the average value of the reference discrete values ​​of each reference hull point cloud in the hull width direction, the average value of the reference discrete values ​​of each reference hull point cloud in the hull length direction, and the average value of the reference discrete values ​​of each reference hull point cloud in the hull height direction, respectively, and use them as the discrete values ​​of the reference hull point cloud in the hull width direction, hull length direction, and hull height direction.

3. The method according to claim 1, characterized in that, The step of determining the candidate compression ratio based on the discrete values ​​of the reference hull point cloud in the hull width direction, hull length direction, and hull height direction includes: The ratio of the discrete values ​​of the reference hull point cloud in the width direction to the discrete values ​​in the length direction is used as the candidate compression ratio of the reference hull point cloud in the length direction. The ratio of the discrete values ​​of the reference hull point cloud in the width direction to the discrete values ​​in the height direction is used as the candidate compression ratio of the reference hull point cloud in the height direction.

4. The method according to claim 1, characterized in that, The step of identifying the ship's hull outline in the point cloud data based on the compressed target region point cloud includes: Obtain the preset cluster radius; Based on the compressed target region point cloud, the ship hull is clustered according to the clustering radius to identify the ship hull outline in the compressed target region point cloud. Based on the ship's hull outline in the compressed target region point cloud, the ship's hull outline in the point cloud data is identified through the mapping relationship between the compressed target region point cloud and the point cloud data.

5. The method according to claim 4, characterized in that, After identifying the hull outline in the point cloud data, the method further includes: Planar fitting is performed on the point cloud belonging to the hull outline to obtain multiple sets of planar point clouds; Among the multiple sets of planar point clouds, planar point clouds that meet preset conditions are selected as candidate planar point clouds, wherein the preset condition is that the angle between the plane containing the planar point cloud and the horizontal plane is less than a preset threshold. Calculate the area of ​​the plane formed by each candidate planar point cloud, and select the candidate planar point cloud with the largest area as the hull deck point cloud; The deck height is calculated based on the hull deck point, and this deck height is used as the ship's side height.

6. The method according to claim 4, characterized in that, The method further includes: A reference point is determined in the point cloud that is within the outline of the hull, and the moving distance and direction of the reference point are recorded within a set time. The ratio between the distance traveled and the set time is taken as the ship's speed, and the direction of travel is taken as the ship's heading. In a direction perpendicular to the ship's heading, the longest point cloud distance within the ship's outline on the horizontal plane is taken as the ship's width.

7. The method according to claim 6, characterized in that, The method further includes: Determine the reference line; Obtain the movement time of the point cloud belonging to the hull outline from the moment it touches the reference line until it leaves the reference line; The product of the travel time and the ship's speed is taken as the ship's length.

8. A ship identification device, characterized in that, The device includes: An acquisition unit is used to acquire point cloud data of a target region and to acquire discrete values ​​of the reference hull point cloud in the width, length, and height directions of the hull, wherein the discrete values ​​characterize the density of the hull point cloud in the corresponding directions; a candidate compression ratio is determined based on the discrete values ​​of the reference hull point cloud in the width, length, and height directions; the reference hull point cloud is compressed in the height and length directions according to the candidate compression ratio to obtain a compressed reference hull point cloud; if the discrete values ​​of the compressed reference hull point cloud in the width, length, and height directions are approximately equal, the candidate compression ratio is used as the target compression ratio; wherein, the approximate equality is satisfied when the difference between any two discrete values ​​in the width, length, and height directions of the hull point cloud does not exceed a preset threshold. A compression unit is used to compress the point cloud data of the target area along the direction perpendicular to the water surface and the length direction of the river channel according to the target compression ratio, so as to obtain the compressed point cloud of the target area. The identification unit is used to identify the hull outline in the point cloud data based on the compressed target area point cloud.

9. An electronic device, characterized in that, The method includes one or more processors and one or more memories, wherein at least one piece of program code is stored in the one or more memories, and the at least one piece of program code is loaded and executed by the one or more processors to implement the method as claimed in any one of claims 1-7.

Images