Image-based ship automatic identification method and intelligent navigation beacon camera device
By combining multi-channel information comparison and image processing technology with laser ranging, accurate verification of ship identity in complex waterway environments has been achieved, solving the security and identification problems of existing systems.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- THE NAVIGATION GUARANTEE CENT OF NORTH CHINA SEA NGCN MOT
- Filing Date
- 2026-06-04
- Publication Date
- 2026-07-03
AI Technical Summary
Existing Automatic Identification Systems (AIS) lack security measures, their data can be tampered with and their identities can be cloned, and image recognition is difficult in complex waterway environments.
By comprehensively using comparison results, communication information, and cloud information, combined with laser ranging and image processing technology, information on ship objects is extracted for multi-channel comparison to verify their identity, including comparison of size, text, color, and outline features.
It improves the accuracy and security of ship identification verification and solves the problems of data tampering and identification in complex environments.
Smart Images
Figure CN122336725A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of image processing technology, and in particular to an image-based automatic ship identification method and an intelligent navigation mark checkpoint device. Background Technology
[0002] Checkpoint devices are electronic inspection stations set up at the entrances and exits of waterways, rivers, and ports. Their purpose is to record the passage information of ships and verify their identity information. Based on this information, overload detection, illegal operation capture, safety warning, traffic management, and waterway data analysis can be achieved simultaneously.
[0003] Currently, one method used for verifying the identity of a vessel is the Automatic Identification System (AIS). The system works by periodically broadcasting the vessel's information to the outside world. The checkpoint device receives and records the broadcast information. The problem with the AIS is that it lacks necessary security measures (no protocol authentication, data can be tampered with, identity can be cloned, etc.), so it can only be used as an auxiliary means.
[0004] To address the shortcomings of Automatic Identification Systems (AIS), image processing methods are currently used for proactive identity verification. However, the complex environment within waterways makes it difficult to identify ships solely by capturing images. Summary of the Invention
[0005] This application provides an image-based automatic ship identification method and an intelligent navigation mark checkpoint device, which comprehensively uses comparison results, communication information and cloud information to verify the ship's identity. This method uses multi-channel data for comprehensive comparison, which can ensure the accuracy of the verification results while verifying the ship's identity.
[0006] The above-mentioned objective of this application is achieved through the following technical solution: In a first aspect, this application provides an image-based automatic ship identification method, comprising: In response to the communication information sent by the Automatic Identification System (AIS), the communication information is parsed to obtain the ship's identity information and ship's location information; Images of the ship are collected based on its location information and recorded as the first analysis image; The position calculation parameters of a ship are determined using laser ranging, and these parameters include angles and distances. Extract the information of the first ship object included in the first analysis image. The information of the first ship object includes size information, text information, color information and contour feature information. The first ship object information is compared with the second ship object information in the database to obtain the comparison result; Verify the comparison results, communication information, and cloud information. When the comparison results, communication information, and cloud information are all consistent, the ship's identity is confirmed.
[0007] In one possible implementation of the first aspect, after obtaining the first analysis image, it further includes: The first analysis image is divided into regions of interest and regions of non-interest. Determine the edge contour and interference region of the region of interest, where the interference region is located on or inside the edge contour. Remove the interfering area; When removing interference areas, it is necessary to determine the type of interference area, which includes light interference and water surface interference.
[0008] In one possible implementation of the first aspect, dividing the first analysis image into regions of interest and non-regions of interest includes: The first analysis image is processed using an edge extraction operator to obtain a ship outline, and the number of ship outlines is at least one. When there are multiple ship outlines, the ship outlines are marked using a distance measurement method; A background model is built using the first analyzed image; Acquire a second analysis image, the second analysis image being generated later than the first analysis image; A contrast background model is generated by combining the background model with the second analysis image; Compare the background model and the contrast background model. When there are differences between the background model and the contrast background model, determine the region of interest.
[0009] In one possible implementation of the first aspect, after determining the region of interest, the method further includes correcting the lower edge of the region of interest. Correcting the lower edge of the region of interest includes: Multiple sets of analysis regions are sequentially established on both sides of the lower edge of the region of interest. Each set of analysis regions includes two analysis regions, and the two analysis regions belonging to the same set are located on both sides of the lower edge of the region of interest. Calculate the texture feature values of two analysis regions in the same group; When the texture feature values of two analysis regions in the same group are the same, move the lower edge of the region of interest until the texture feature values of the two analysis regions in the same group are different.
[0010] In one possible implementation of the first aspect, removing the interference region of type optical interference includes: Images of regions of interest are acquired multiple times over a time series; Identify and mark the locations of light interference regions on each region of interest image; Based on the location of the light interference region, determine the hidden information of the region of interest on the region of interest image excluding the light interference region; Replace the light interference region with information hidden in the region of interest.
[0011] In one possible implementation of the first aspect, the method further includes determining the shadow region, which includes: The region of interest image is converted into the HSV space to determine the hue, saturation, and brightness of the region of interest image; The region of interest image is divided according to brightness, and brightness division lines are obtained; Compare the hue and saturation of the areas on both sides of the brightness division line and determine the shadow area based on the difference in hue and saturation.
[0012] In one possible implementation of the first aspect, removing the shadow area includes: Multiple region of interest images are obtained, among which at least one region of interest image is overexposed; The multiple region of interest images are converted into HSV space to determine the hue, saturation, and brightness of the region of interest images; Based on the brightness value of the non-shaded area on one side of the brightness division line, a region is selected from the multiple region of interest images to obtain the selected region. Replace the shaded area with the selected area.
[0013] Secondly, this application provides an image-based automatic ship identification device, comprising: The communication unit is used to respond to the communication information sent by the Automatic Identification System (AIS) and parse the communication information to obtain the ship's identity information and ship's position information. The image acquisition unit is used to acquire images of the ship based on the ship's position information, which are denoted as the first analysis image; The distance calculation unit is used to determine the position calculation parameters of a ship using laser ranging. The position calculation parameters include angle and distance. The information extraction unit is used to extract information about a first ship object included in the first analysis image. The information about the first ship object includes size information, text information, color information, and contour feature information. The information comparison unit is used to compare the first ship object information with the second ship object information in the database to obtain the comparison result. The identity verification unit is used to verify the comparison results, communication information and cloud information. When the comparison results, communication information and cloud information are consistent, the identity of the ship is determined.
[0014] Thirdly, this application provides an image-based automatic identification system for ships, the system comprising: One or more memories for storing instructions; and One or more processors are configured to call and execute the instructions from the memory to perform the methods described in the first aspect and any possible implementation thereof.
[0015] Fourthly, this application provides a computer-readable storage medium, the computer-readable storage medium comprising: The program, when run by a processor, is executed as described in the first aspect and any possible implementation thereof.
[0016] Fifthly, this application provides a computer program product, including program instructions that, when run by a computing device, execute the method described in the first aspect and any possible implementation thereof.
[0017] Sixthly, this application provides a chip system including a processor for implementing the functions involved in the foregoing aspects, such as generating, receiving, transmitting, or processing the data and / or information involved in the foregoing methods.
[0018] This chip system can consist of chips or include chips and other discrete components.
[0019] In one possible design, the chip system also includes a memory for storing necessary program instructions and data. The processor and the memory can be decoupled and located on different devices, connected via wired or wireless means, or the processor and the memory can be coupled to the same device. Attached Figure Description
[0020] Figure 1 This is a flowchart illustrating the steps of an image-based automatic ship identification method provided in this application.
[0021] Figure 2 This is a schematic diagram illustrating the working principle of an intelligent navigation beacon checkpoint device provided in this application.
[0022] Figure 3 This is a schematic diagram of a region of interest and a region of non-interest provided in this application.
[0023] Figure 4 This is a schematic diagram of a ship's outline provided in this application.
[0024] Figure 5 This is a schematic diagram of marking the outline of a ship using a distance measurement method, as provided in this application.
[0025] Figure 6This is a schematic diagram of dividing a first analysis image into non-overlapping blocks, as provided in this application.
[0026] Figure 7 This is a schematic diagram of how multiple sets of analysis regions are sequentially established on both sides of the lower edge of the region of interest, as provided in this application.
[0027] Figure 8 This is a schematic diagram of replacing the light interference region with region of interest hiding information provided in this application. Detailed Implementation
[0028] The technical solutions in this application will be further described in detail below with reference to the accompanying drawings.
[0029] This application discloses an image-based automatic ship identification method. Please refer to [link to relevant documentation]. Figure 1 In some examples, the image-based automatic ship identification method disclosed in this application includes the following steps: S101, in response to the communication information sent by the Automatic Identification System (AIS), the communication information is parsed to obtain the ship's identity information and ship's position information; S102, Collect images of the ship based on the ship's position information, and record them as the first analysis image; S103, using laser ranging to determine the position calculation parameters of the ship, including angle and distance; S104, Extract the first ship object information included in the first analysis image. The first ship object information includes size information, text information, color information and contour feature information. S105, compare the first ship object information with the second ship object information in the database to obtain the comparison result; S106, verify the comparison results, communication information and cloud information. When the comparison results, communication information and cloud information are consistent, the identity of the vessel is determined.
[0030] Specifically, in step S101, the intelligent navigation beacon checkpoint device first receives communication information sent by the Automatic Identification System (AIS). At this time, the communication information sent by the AIS is first parsed to obtain the ship's identity information and ship's location information.
[0031] Vessel identification information includes MMSI code, IMO number, vessel name, call sign, and vessel identification number (registered in China).
[0032] Ship location information is its latitude and longitude or location coordinates.
[0033] In step S102, the image of the ship is acquired based on the ship's position information and is recorded as the first analysis image. Here, acquiring the image of the ship refers to the intelligent navigation beacon gate device adjusting its direction to acquire the image of the ship. When the intelligent navigation beacon gate device does not need to adjust its direction, the image of the ship is acquired directly.
[0034] In step S103, laser ranging is used to determine the position calculation parameters of the ships. The position calculation parameters include angle and distance. Here, the laser ranging method is used to determine the position calculation parameters of the ships for all ships included in the first analysis image.
[0035] For intelligent navigation mark checkpoint devices, the local location coordinates are known quantities. Therefore, the ship's position information can be calculated by using angles and distances. When the calculated ship position information matches the received ship position information, the ship's identity can be preliminarily determined.
[0036] In step S104, the first ship object information included in the first analysis image is extracted. The first ship object information includes size information, text information, color information and contour feature information. Then in step S105, the first ship object information is compared with the second ship object information in the database to obtain the comparison result. At this time, the comparison result can be either consistent or inconsistent.
[0037] Finally, in step S106, the comparison results, communication information, and cloud information are verified. When the comparison results, communication information, and cloud information are all consistent, the ship's identity is determined. Here, the comparison results refer to the ship's identity information obtained from image comparison, the communication information refers to the ship's identity information obtained from parsing the received information, and the cloud information refers to the route information actively uploaded by the ship during its voyage. The route information can display the ship's location information and ship's identity information.
[0038] Image comparison and communication information can also be used to obtain ship location information. At this time, the ship location information obtained from these three channels is compared with the ship identity information. If they are all consistent, the ship's identity is confirmed; otherwise, a warning of identity doubt is sent to the cloud.
[0039] like Figure 2 As shown, the smart checkpoint device receives broadcast information from the ship (obtaining the first set of ship location information and ship identity information), obtains the first set of ship location information and ship identity information through local image comparison and location calculation, and obtains the third set of ship location information and ship identity information by querying the cloud. The three sets of ship location information and ship identity information are consistent, thus confirming the ship's identity.
[0040] It should be noted that current route information is not necessarily transmitted to the cloud. Therefore, when route information is not available, only the comparison results and communication information are verified.
[0041] The comparison methods for size information, text information, color information, and contour feature information are as follows: Size information comparison: The ship's length, beam, and superstructure height are calculated using the ship's dimensions and actual distance included in the first analysis image. These dimensions are then compared with the ship's length and beam broadcast by AIS or the standard dimensions in the maritime registration database.
[0042] Text information comparison: OCR models such as CRNN and SVTR are used to recognize text and obtain ship names and numbers. The ship names reported by AIS are then compared with the standard names in the maritime database.
[0043] Color information comparison involves extracting the ship's main body color, cabin color, and cargo hold color from the HSV / Lab space, and then comparing them with the registered / historical features.
[0044] Contour feature information comparison involves extracting the outer contour, convex hull, aspect ratio, sharp corners, and superstructure location, and then comparing it with historical data or data in the filing database by calculating Euclidean distance or cosine similarity.
[0045] In some cases, after obtaining the first analysis image, it is processed in the following way: S201, Divide the first analysis image into regions of interest and regions of non-interest; S202, Determine the edge contour and interference region included in the region of interest, wherein the interference region is located on or inside the edge contour; S203, Remove the interference area; When removing interference areas, it is necessary to determine the type of interference area, which includes light interference and water surface interference.
[0046] Specifically, please refer to Figure 3 The region of interest here refers to the ships in the first analysis image, and the region of non-interest refers to the area in the first analysis image other than the ships. To process the first analysis image, it is necessary to first determine the ships included in the first analysis image (region of interest).
[0047] Next, the edge contour and interference area of the region of interest are determined. The interference area is located on or inside the edge contour. Here, the edge contour refers to the maximum outer contour of the ship in the region of interest. The interference area refers to the reflective area and the shadow area. The reflective area is mainly formed by the reflection of the water surface, and the shadow area is mainly caused by the three-dimensional structure of the ship.
[0048] After identifying the interference areas, they are categorized into light interference and water surface interference, and then targeted removal is carried out. This part will be introduced later.
[0049] The specific method for dividing the first analysis image into regions of interest and non-regions of interest is as follows: S301, The first analysis image is processed using an edge extraction operator to obtain a ship outline, and the number of ship outlines is at least one; S302, When there are multiple ship outlines, the ship outlines shall be marked by a distance measurement method; S303, Use the first analysis image to establish a background model; S304, acquire the second analysis image, the generation time of the second analysis image is later than the generation time of the first analysis image; S305, obtain and use the background model combined with the second analysis image to generate a contrast background model; S306, compare the background model and the contrasting background model, and determine the region of interest when there are differences between the background model and the contrasting background model.
[0050] In this section, the first analysis image is processed using an edge extraction operator. The edge extraction operator used is the Sobel operator, Prewitt operator, Laplacian operator, or Canny operator (preferred). Specifically, the process involves grayscale conversion, Gaussian blurring, edge calculation using the edge extraction operator, morphological processing, contour finding, and fitting the minimum bounding rectangle / polygon.
[0051] When the ship's outline ( Figure 4 When there are multiple ship outlines (as shown), a ranging method is used to mark them. Laser ranging is used to avoid overlap in the identification of two ship outlines. Figure 5 As shown, specifically, when two ship outlines are identified, if only image processing is used, there may be a potential overlap and confusion problem, that is, the outline of one ship may be assigned to the outline of the other ship. However, after marking the ship outlines using a distance measurement method, the ship outlines can be separated by distance.
[0052] In some possible implementations, when measuring the distance to a ship's outline, the region within the outline is first divided using color and texture, and then the distance to each of the divided regions is measured.
[0053] Then, a background model is built using the first analyzed image, as follows: The first analysis image is divided into non-overlapping blocks (e.g., 8×8 / 16×16). Then, the grayscale mean and / or texture features of each block are calculated. Based on this, a background model is constructed. The second analysis image is processed in the same way. Figure 6 As shown.
[0054] The image blocks in the first analysis image and the image blocks in the second analysis image are in one-to-one correspondence in position.
[0055] Finally, the background model and the contrasting background model are compared. When there are differences between the background model and the contrasting background model, the region of interest is determined.
[0056] When using the grayscale mean, the difference between corresponding positions of the background model and the contrasting background model should be 5~8 (normal waves, slight lighting changes) or 10~15 (large waves, strong reflections). A suitable reference value is 12. The size of the waves should be limited according to the actual location and time period.
[0057] When using texture features, grayscale variance is generally used, with a value range of 180 to 220. A suitable reference value is 200.
[0058] In some possible implementations, if the absolute difference of the mean grayscale is greater than 12 and the variance of the grayscale of the current block is greater than 200, it is determined that there is a difference between two corresponding blocks (the image block of the first analysis image and the image block of the second analysis image, respectively), and the set of image blocks with differences on the second analysis image is the region of interest.
[0059] Another method is as follows: The detection model is either the YOLO model or the Mask R-CNN model, and the processing procedure is as follows: Scale to a fixed model size (e.g., 640×640, 1280×1280); normalize pixel values (0~1 or -1~1); fill with black borders to maintain proportions and avoid stretching and deformation; The network performs convolution on the image, extracting ship features layer by layer: Shallow layer: edges, outline, horizon, straight lines of the hull; Middle layer: hull structure, deck, and cabin shape; In-depth: Overall semantic features of the ship (ship vs. dock / bridge / island / buoy); Then, the extracted features are used for judgment. This method requires more computing resources and is insufficient in terms of processing speed and timeliness. Therefore, this method is not used in this application.
[0060] In some examples, after determining the region of interest, the process also includes refining the lower edge of the region of interest. This refining includes: S401, Multiple sets of analysis regions are sequentially established on both sides of the lower edge of the region of interest. Each set of analysis regions includes two analysis regions, and the two analysis regions belonging to the same set are located on both sides of the lower edge of the region of interest. S402, calculate the texture feature values of two analysis regions in the same group; S403, when the texture feature values of two analysis regions in the same group are the same, move the lower edge of the region of interest until the texture feature values of the two analysis regions in the same group are different.
[0061] The reason for the correction is that this application uses a comparison method, which has the advantage of fast processing speed, but has certain identification errors because the water surface of the river is changing and other boats on the river are also moving.
[0062] Correcting the lower edge of the region of interest primarily addresses recognition errors caused by changes in the river's water level. Specifically, this is achieved by comparing multiple analysis regions to determine the appropriateness of the lower edge's position. Figure 7 As shown.
[0063] The texture feature value here is generally grayscale variance. The size of the analysis area is 5×5 or 7×7. When the grayscale variance of two analysis areas is less than 20, they are considered the same. In this case, the corresponding position of the lower edge of the region of interest needs to be moved. When it is greater than or equal to 20, they are considered different. In this case, the corresponding position of the lower edge of the region of interest does not need to be moved.
[0064] Both upward and downward movements are required in the direction of movement.
[0065] The specific method for removing interference areas of type optical interference is as follows: Images of regions of interest are acquired multiple times over a time series; Identify and mark the locations of light interference regions on each region of interest image; Based on the location of the light interference region, determine the hidden information of the region of interest on the region of interest image excluding the light interference region; Replace the light interference region with information hidden in the region of interest.
[0066] In the above method, the specific way to determine the light interference region on each region of interest image is to convert the region of interest image into a grayscale image. The characteristics of the light interference region are that the grayscale value is extremely high, close to saturation, and the reference value is greater than or equal to 220~245. The texture inside the region is flat and the variance is extremely low.
[0067] After filtering out regions with extremely high grayscale values, morphological operations (dilation + erosion) are used to remove noise and obtain connected components. Then, connected components with an area greater than a minimum threshold (e.g., 5×5) are filtered out and marked as light interference regions.
[0068] The location of the light interference area is based on the coordinates of feature points on the ship in the image, such as the bow position, stern position, or a certain area as (0,0). The location of the light interference area is calculated based on the (0,0) point to ensure the consistency of the coordinate system.
[0069] Since the light interference regions differ on images of different regions of interest (ROIs), region-of-interest (ROI) images excluding the light interference regions can be determined based on their locations. Then, the ROI hiding information can be used to replace the light interference regions. Figure 8 As shown.
[0070] The specific method involves cropping the image of other regions of interest (ROIs) based on the shape of the light interference region (in conjunction with its position coordinates). The cropped content is then processed using median synthesis, and finally, the light interference region is replaced with the hidden information of the ROI.
[0071] In addition, it is necessary to determine the shaded area, which is done as follows: The region of interest image is converted into the HSV space to determine the hue, saturation, and brightness of the region of interest image; The region of interest image is divided according to brightness, and brightness division lines are obtained; Compare the hue and saturation of the areas on both sides of the brightness division line and determine the shadow area based on the difference in hue and saturation.
[0072] Brightness is used here for determination. Specifically, the region of interest image is divided according to brightness to obtain brightness division lines. Then, the hue and saturation of the regions on both sides of the brightness division lines are compared, and the shadow region is determined based on the difference in hue and saturation.
[0073] When dividing the region of interest image based on brightness, the area with a brightness value between 40 and 100 is designated as the shadow area. The outline of the shadow area is the brightness division line. Then, the hue and saturation of the areas on both sides of the brightness division line are compared, and the shadow area is determined based on the difference in hue and saturation.
[0074] The purpose of the comparison is to determine that the areas on both sides of the brightness dividing line are the same, and are divided by the brightness dividing line only because of the difference in brightness. The reference value for the hue difference is 20°, and the reference value for the saturation difference is 0.2. That is, the shadow area is determined only when the hue difference between the areas on both sides of the brightness dividing line is less than 20° and the saturation difference is less than 0.2.
[0075] The following are methods to remove shadow areas: Multiple region of interest images are obtained, among which at least one region of interest image is overexposed; The multiple region of interest images are converted into HSV space to determine the hue, saturation, and brightness of the region of interest images; Based on the brightness value of the non-shaded area on one side of the brightness division line, a region is selected from the multiple region of interest images to obtain the selected region. Replace the shaded area with the selected area.
[0076] The process here still involves local selection and transfer. Specifically, the region is selected from multiple region of interest images based on the brightness value of the non-shaded area on one side of the brightness division line. The difference between the brightness value of the selected region and the brightness value of the non-shaded area on one side of the brightness division line is less than 20. In other words, when selecting, regions with similar or identical brightness values are used for replacement.
[0077] When multiple regions are selected, the median composite method is used for processing. The method for determining the location is the same as that for processing light interference regions, which will not be repeated here.
[0078] This application also provides an image-based automatic ship identification device, comprising: The communication unit is used to respond to the communication information sent by the Automatic Identification System (AIS) and parse the communication information to obtain the ship's identity information and ship's position information. The image acquisition unit is used to acquire images of the ship based on the ship's position information, which are denoted as the first analysis image; The distance calculation unit is used to determine the position calculation parameters of a ship using laser ranging. The position calculation parameters include angle and distance. The information extraction unit is used to extract information about a first ship object included in the first analysis image. The information about the first ship object includes size information, text information, color information, and contour feature information. The information comparison unit is used to compare the first ship object information with the second ship object information in the database to obtain the comparison result. The identity verification unit is used to verify the comparison results, communication information and cloud information. When the comparison results, communication information and cloud information are consistent, the identity of the ship is determined.
[0079] Furthermore, after obtaining the first analysis image, the process also includes: The first analysis image is divided into regions of interest and regions of non-interest. Determine the edge contour and interference region of the region of interest, where the interference region is located on or inside the edge contour. Remove the interfering area; When removing interference areas, it is necessary to determine the type of interference area, which includes light interference and water surface interference.
[0080] Furthermore, the first analysis image is divided into regions of interest and non-regions of interest, including: The first analysis image is processed using an edge extraction operator to obtain a ship outline, and the number of ship outlines is at least one. When there are multiple ship outlines, the ship outlines are marked using a distance measurement method; A background model is built using the first analyzed image; Acquire a second analysis image, the second analysis image being generated later than the first analysis image; A contrast background model is generated by combining the background model with the second analysis image; Compare the background model and the contrast background model. When there are differences between the background model and the contrast background model, determine the region of interest.
[0081] Furthermore, after determining the region of interest, the process also includes refining the lower edge of the region of interest. This refining includes: Multiple sets of analysis regions are sequentially established on both sides of the lower edge of the region of interest. Each set of analysis regions includes two analysis regions, and the two analysis regions belonging to the same set are located on both sides of the lower edge of the region of interest. Calculate the texture feature values of two analysis regions in the same group; When the texture feature values of two analysis regions in the same group are the same, move the lower edge of the region of interest until the texture feature values of the two analysis regions in the same group are different.
[0082] Furthermore, the areas to be removed as optical interference include: Images of regions of interest are acquired multiple times over a time series; Identify and mark the locations of light interference regions on each region of interest image; Based on the location of the light interference region, determine the hidden information of the region of interest on the region of interest image excluding the light interference region; Replace the light interference region with information hidden in the region of interest.
[0083] Furthermore, it also includes determining the shaded area, which includes: The region of interest image is converted into the HSV space to determine the hue, saturation, and brightness of the region of interest image; The region of interest image is divided according to brightness, and brightness division lines are obtained; Compare the hue and saturation of the areas on both sides of the brightness division line and determine the shadow area based on the difference in hue and saturation.
[0084] Furthermore, removing shadow areas includes: Multiple region of interest images are obtained, among which at least one region of interest image is overexposed; The multiple region of interest images are converted into HSV space to determine the hue, saturation, and brightness of the region of interest images; Based on the brightness value of the non-shaded area on one side of the brightness division line, a region is selected from the multiple region of interest images to obtain the selected region. Replace the shaded area with the selected area.
[0085] In one example, the unit in any of the above devices may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.
[0086] For example, when the units in the device can be implemented through a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor capable of calling programs. Alternatively, these units can be integrated together to form a system-on-a-chip (SOC).
[0087] In this application, various objects such as messages / information / devices / network elements / systems / apparatus / actions / operations / processes / concepts may be named. It is understood that these specific names do not constitute a limitation on the relevant objects. The names may be changed depending on the scenario, context, or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from their functions and technical effects embodied / performed in the technical solution.
[0088] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.
[0089] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.
[0090] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.
[0091] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.
[0092] It should also be understood that in the various embodiments of this application, the terms "first," "second," etc., are merely to indicate that multiple objects are different. For example, a first time window and a second time window are only to indicate different time windows. They should not have any effect on the time windows themselves, and the aforementioned terms "first," "second," etc., should not impose any limitations on the embodiments of this application.
[0093] It should also be understood that, in the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced by each other, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.
[0094] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they 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 a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a computer-readable 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 computer-readable storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.
[0095] This application also provides an image-based automatic ship identification system, the system comprising: One or more memories for storing instructions; and One or more processors are configured to retrieve and execute the instructions from the memory, performing the methods described above.
[0096] This application also provides a computer program product including instructions that, when executed, cause the terminal device and the network device to perform operations corresponding to the methods described above.
[0097] This application also provides a chip system including a processor for implementing the functions involved in the above description, such as generating, receiving, transmitting, or processing the data and / or information involved in the above methods.
[0098] This chip system can consist of chips or include chips and other discrete components.
[0099] The processor mentioned above can be a CPU, a microprocessor, an ASIC, or one or more integrated circuits that execute a program to control the method of transmitting the feedback information described above.
[0100] In one possible design, the chip system also includes a memory for storing necessary program instructions and data. The processor and the memory can be decoupled and located on different devices, connected via wired or wireless means to support the chip system in implementing the various functions described in the above embodiments. Alternatively, the processor and the memory can also be coupled to the same device.
[0101] Optionally, the computer instructions are stored in memory.
[0102] Optionally, the memory can be a storage unit within the chip, such as a register or cache. Alternatively, the memory can be a storage unit located outside the chip within the terminal, such as a ROM or other types of static storage devices that can store static information and instructions, such as RAM.
[0103] It is understood that the memory in this application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory.
[0104] Non-volatile memory can be ROM, programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory.
[0105] Volatile memory can be RAM, which is used as an external cache. There are many different types of RAM, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory.
[0106] The embodiments described in this specific implementation are preferred embodiments of this application and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.
Claims
1. An image-based automatic ship identification method, characterized in that, include: In response to the communication information sent by the Automatic Identification System (AIS), the communication information is parsed to obtain the ship's identity information and ship's location information; Images of the ship are collected based on its location information and recorded as the first analysis image; The position calculation parameters of a ship are determined using laser ranging, and these parameters include angles and distances. Extract the information of the first ship object included in the first analysis image. The information of the first ship object includes size information, text information, color information and contour feature information. The first ship object information is compared with the second ship object information in the database to obtain the comparison result; Verify the comparison results, communication information, and cloud information. When the comparison results, communication information, and cloud information are all consistent, the ship's identity is confirmed.
2. The image-based automatic ship identification method according to claim 1, characterized in that, After obtaining the first analysis image, the process also includes: The first analysis image is divided into regions of interest and regions of non-interest. Determine the edge contour and interference region of the region of interest, where the interference region is located on or inside the edge contour. Remove the interfering area; When removing interference areas, it is necessary to determine the type of interference area, which includes light interference and water surface interference.
3. The image-based automatic ship identification method according to claim 2, characterized in that, The first analysis image is divided into regions of interest and non-regions of interest, including: The first analysis image is processed using an edge extraction operator to obtain a ship outline, and the number of ship outlines is at least one. When there are multiple ship outlines, the ship outlines are marked using a distance measurement method; A background model is built using the first analyzed image; Acquire a second analysis image, the second analysis image being generated later than the first analysis image; A contrast background model is generated by combining the background model with the second analysis image; Compare the background model and the contrast background model. When there are differences between the background model and the contrast background model, determine the region of interest.
4. The image-based automatic ship identification method according to claim 3, characterized in that, After determining the region of interest, the process also includes refining the lower edge of the region of interest. This refining includes: Multiple sets of analysis regions are sequentially established on both sides of the lower edge of the region of interest. Each set of analysis regions includes two analysis regions, and the two analysis regions belonging to the same set are located on both sides of the lower edge of the region of interest. Calculate the texture feature values of two analysis regions in the same group; When the texture feature values of two analysis regions in the same group are the same, move the lower edge of the region of interest until the texture feature values of the two analysis regions in the same group are different.
5. The image-based automatic ship identification method according to claim 2, characterized in that, The interference areas to be removed as optical interference include: Images of regions of interest are acquired multiple times over a time series; Identify and mark the locations of light interference regions on each region of interest image; Based on the location of the light interference region, determine the hidden information of the region of interest on the region of interest image excluding the light interference region; Replace the light interference region with information hidden in the region of interest.
6. The image-based automatic ship identification method according to claim 5, characterized in that, This also includes determining the shaded area, which includes: The region of interest image is converted into the HSV space to determine the hue, saturation, and brightness of the region of interest image; The region of interest image is divided according to brightness, and brightness division lines are obtained; Compare the hue and saturation of the areas on both sides of the brightness division line and determine the shadow area based on the difference in hue and saturation.
7. The image-based automatic ship identification method according to claim 6, characterized in that, Removing shadow areas includes: Multiple region of interest images are obtained, among which at least one region of interest image is overexposed; The multiple region of interest images are converted into HSV space to determine the hue, saturation, and brightness of the region of interest images; Based on the brightness value of the non-shaded area on one side of the brightness division line, a region is selected from the multiple region of interest images to obtain the selected region. Replace the shaded area with the selected area.
8. An image-based automatic ship identification device, characterized in that, include: The communication unit is used to respond to the communication information sent by the Automatic Identification System (AIS) and parse the communication information to obtain the ship's identity information and ship's position information. The image acquisition unit is used to acquire images of the ship based on the ship's position information, which are denoted as the first analysis image; The distance calculation unit is used to determine the position calculation parameters of a ship using laser ranging. The position calculation parameters include angle and distance. The information extraction unit is used to extract information about a first ship object included in the first analysis image. The information about the first ship object includes size information, text information, color information, and contour feature information. The information comparison unit is used to compare the first ship object information with the second ship object information in the database to obtain the comparison result. The identity verification unit is used to verify the comparison results, communication information and cloud information. When the comparison results, communication information and cloud information are consistent, the identity of the ship is determined.
9. An image-based automatic identification system for ships, characterized in that, The system includes: One or more memories for storing instructions; and One or more processors are configured to retrieve and execute the instructions from the memory to perform the method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes: The program, when run by the processor, executes the method as described in any one of claims 1 to 7.