Inland waterway ship stagnation state detection method and system

Through the combination of multiple dual-spectrum thermal imaging cameras and deep learning algorithms, active, real-time and accurate detection of ship stagnation in inland waterways is achieved, and the problems of poor passive equipment recognition effect and poor remote sensing image age in the existing technology are solved. , providing efficient and low-cost regulatory measures.

CN118968434AInactive Publication Date: 2024-11-15长江信达软件技术(武汉)有限责任公司
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411010094.0
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-26
Publication Date
2024-11-15
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

In the prior art, the detection of stagnant state of the waterway ship depends on passive equipment, resulting in poor identification effect, insufficient timeliness and accuracy of remote sensing images, making it difficult to meet the supervision needs of inland waterways.

Method used

Multiple dual-spectrum thermal imaging cameras are used to collect ship navigation video frames, and combined with deep learning algorithms and SURF feature matching technology to build a ship recognition and stagnation analysis model, and actively detect the images through real-time monitoring to obtain ship stagnation analysis results.

Benefits of technology

The waterway management department has realized active, real-time and accurate detection of the ship's stagnation status, avoiding timeliness and continuity defects, reducing inspection costs, and is suitable for equipment with resource-constrained.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118968434A_ABST
    Figure CN118968434A_ABST
Patent Text Reader

Abstract

The invention provides an inland waterway ship stagnation state detection method and system, and belongs to the technical field of image recognition, and the method comprises the steps: collecting a ship monitoring image of a designated waterway, recognizing a target ship in the ship monitoring image through a deep learning algorithm, and obtaining a first image frame; after a preset time threshold, matching the target ship to obtain a second image frame; and carrying out stagnation comparison analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result. Compared with an existing ship stopping detection algorithm, the method has the advantages of being active, efficient, low in cost and light in weight, can be widely applied to inland waterway ship supervision work, provides a more reliable and efficient management means for a waterway management department, and facilitates improvement of the intelligent level of waterway management services.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of image recognition technology, and in particular to a method and system for detecting the stagnant state of an inland waterway ship. Background Art

[0002] The abnormal stagnation of ships on the waterway may potentially involve some illegal situations, such as illegal sand mining, illegal transactions, etc., and will also increase the risk of accidents such as ship collisions, obstructions to navigation, etc. The detection of ship stagnation status is of great significance in waterway supervision.

[0003] At present, the main means for waterway management departments to detect the stagnant state of ships is the Automatic Identification System (AIS). This method relies on the communication between the coastal base station and the ship-borne AIS equipment, passively receives the ship's broadcast messages, and parses its position, speed, acceleration and other information. It is not proactive. When the ship does not install or illegally shuts down the AIS equipment, the ship cannot be identified. In recent years, deep learning and computer vision technology have developed rapidly. Some researchers have actively obtained remote sensing images through synthetic aperture radar (SAR) technology to identify ships. Limited by the economic and time costs of shooting, processing and interpreting remote sensing images, this method cannot obtain remote sensing images of the same area multiple times in a short period of time. This method has significant defects such as poor timeliness and poor time continuity. It can only obtain and identify the static features of the ship and is difficult to apply to the detection of stagnant state of ships. There are also target tracking algorithms to identify ship tracking trajectories, but most of them are aimed at large port ships. The application accuracy for inland waterway ships is not high, and the hardware equipment requirements are high. Summary of the invention

[0004] The present invention provides a method and system for detecting the stagnant state of a ship in an inland waterway, so as to solve the defects in the prior art that waterway ship parking detection needs to rely on passive equipment resulting in poor recognition effect, and the use of remote sensing images has the defects of poor timeliness and low accuracy.

[0005] In a first aspect, the present invention provides a method for detecting a stagnant state of an inland waterway ship, comprising: Collecting ship monitoring images of a designated waterway, identifying a target ship in the ship monitoring images using a deep learning algorithm, and acquiring a first image frame; After a preset time threshold, matching the target ship to obtain a second image frame; A stagnation comparison analysis is performed on the first image frame and the second image frame to obtain a ship stagnation analysis result.

[0006] According to a method for detecting the stagnant state of an inland waterway ship provided by the present invention, before collecting monitoring images of ships in a designated waterway, the method further comprises: Setting up a plurality of dual-spectrum thermal imaging cameras at a plurality of preset positions of the designated waterway, wherein adjacent dual-spectrum thermal imaging cameras have overlapping field of view coverage areas; The multiple dual-spectrum thermal imaging cameras are used to collect ship navigation video frames, and the ship monitoring images in the ship navigation video frames are extracted.

[0007] According to a method for detecting a stagnant state of an inland waterway ship provided by the present invention, a ship monitoring image of a designated waterway is collected, a target ship in the ship monitoring image is identified by using a deep learning algorithm, and a first image frame is obtained, including: The multiple dual-spectrum thermal imaging cameras start shooting, and if the ship monitoring image is obtained, the ship is identified, otherwise the patrol shooting continues; A data set is constructed from a number of ship monitoring images, and the data set is divided into a training set, a validation set, and a test set; Building an initial recognition model based on the YOLO v5 algorithm, training the initial recognition model using the training set based on the initial parameters of the model, and verifying and testing the trained model using the verification set and the test set to obtain a recognition model; The recognition model is used to recognize the ship monitoring image to be recognized, and the first image frame is output.

[0008] According to a method for detecting a stagnant state of an inland waterway ship provided by the present invention, after a preset time threshold, the target ship is matched to obtain a second image frame, comprising: Starting from acquiring the first image frame, after the preset time threshold, adjusting the plurality of dual-spectrum thermal imaging cameras to the same parameter state as that of capturing the first image frame, to acquire the image frame to be identified; Preprocessing the first image frame and the image frame to be recognized to obtain the preprocessed first image frame and the image frame to be recognized; Using the SURF algorithm to extract feature vectors of the preprocessed first image frame and the image frame to be identified; The feature vectors are compared to obtain a matching feature point pair, and the second image frame corresponding to the matching feature point pair is determined according to a preset similarity matching degree.

[0009] According to a method for detecting a stagnant state of an inland waterway ship provided by the present invention, the first image frame and the image frame to be identified are preprocessed to obtain the preprocessed first image frame and the image frame to be identified, including: Acquire the ship outline circumscribed frames of the first image frame and the image frame to be identified; The non-hull part of the image background in the ship outline circumscribed frame is filled with white blocks to obtain the preprocessed first image frame and the image frame to be identified.

[0010] According to a method for detecting a stagnant state of an inland waterway ship provided by the present invention, a SURF algorithm is used to extract feature vectors of the preprocessed first image frame and the image frame to be identified, comprising: Performing filtering and feature extraction on the preprocessed first image frame and the image frame to be identified in sequence at different scales to obtain a plurality of feature points; The feature vector of a preset dimension is generated for the plurality of feature points.

[0011] According to a method for detecting a stagnant state of an inland waterway ship provided by the present invention, a stagnation comparison analysis is performed on the first image frame and the second image frame to obtain a ship stagnation analysis result, including: Calculate the displacement of each pair of feature points in the feature vector of the first image frame and the second image frame within the preset time threshold; If the displacement of each pair of feature points is less than or equal to the displacement threshold, ship stagnation warning information is generated, and the ship stagnation warning information includes warning time, warning position and warning image.

[0012] According to a method for detecting a stagnant state of an inland waterway ship provided by the present invention, after performing a stagnation comparative analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result, the method further includes: Intercepting video surveillance images of a first time period before the warning time and video surveillance images of a second time period after the warning time; The video surveillance images in the first time period and the video surveillance images in the second time period constitute the supporting video material.

[0013] In a second aspect, the present invention further provides a system for detecting a stagnant state of an inland waterway vessel, comprising: An identification module is used to collect ship monitoring images of a designated waterway, identify a target ship in the ship monitoring images using a deep learning algorithm, and obtain a first image frame; A matching module, used for matching the target ship after a preset time threshold, and acquiring a second image frame; The analysis module is used to perform a stagnation comparison analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result.

[0014] In a third aspect, the present invention further provides an electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, it implements any of the above-described methods for detecting a stagnant state of an inland waterway ship.

[0015] Compared with the prior art, the present invention has the following beneficial effects: 1. Proactiveness: Compared with the AIS system, it can transform "passive" into "active". The waterway management department can actively obtain real-time video images of the front-end sensing equipment according to regulatory requirements to detect and analyze the stagnant status of ships in the waterway; 2. Efficiency: Compared with SAR technology, this method avoids the timeliness and continuity defects of SAR technology, and can flexibly set the detection threshold according to regulatory requirements, providing a more efficient detection method for ship supervision; 3. Accuracy: Constructing and labeling inland waterway ship datasets is more applicable to inland waterway supervision scenarios; 4. Low cost: Compared with existing technologies such as SAR technology and target tracking models, the method proposed in this invention has lower training cost for data samples; 5. Lightweight: The technical method proposed in the present invention has light resource requirements and can be used in resource-constrained devices. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 This is one of the flow charts of the method for detecting the stagnant state of an inland waterway ship provided by the present invention; Figure 2 This is the second flow chart of the method for detecting the stagnant state of an inland waterway ship provided by the present invention; Figure 3 This is one of the application effect diagrams provided by the present invention, wherein Figure 3 (a) is the first frame image of the visible light-sand transport ship. Figure 3 (b) is the visible light-sand-carrying ship-second frame image; Figure 4 This is the second schematic diagram of the application effect provided by the present invention, wherein Figure 4 (a) is the first frame of thermal imaging - sand mining ship, Figure 4 (b) is the thermal imaging - sand mining ship - the second frame image; Figure 5 It is a structural schematic diagram of the inland waterway ship stagnation state detection system provided by the present invention; Figure 6 It is a structural schematic diagram of the electronic device provided by the present invention. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the drawings of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] In view of the problems of lack of initiative, high cost, poor timeliness and difficulty in application in existing methods, the present invention proposes a method for detecting the stagnant state of inland waterway ships. The method obtains real-time and continuous images of the supervised waters through front-end video monitoring equipment, and constructs a ship parking detection model by combining deep learning and computer vision technologies.

[0020] Figure 1 FIG. 1 is one of the flow charts of the method for detecting the stagnant state of an inland waterway ship provided by an embodiment of the present invention, such as Figure 1 As shown, including: Step 100: collecting ship monitoring images of a designated waterway, identifying a target ship in the ship monitoring images using a deep learning algorithm, and acquiring a first image frame; Step 200: After a preset time threshold, the target ship is matched to obtain a second image frame; Step 300: performing a stagnation comparison analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result.

[0021] The inland waterway ship stagnation detection proposed in the embodiment of the present invention is applied to the detection of ship stagnation status in the Yangtze River waterway. Since the Yangtze River waterway is the main artery for the economic and social development of the basin, the main framework of the national comprehensive three-dimensional transportation network, and the main channel for economic and trade logistics along the river, there are a small number of ships staying on the Yangtze River waterway, obstructing navigation or engaging in illegal activities. The overall technical process route is as follows Figure 2 As shown, the first step is to set up front-end video surveillance equipment in the target waters; the second step is to build a ship recognition model based on deep learning, and obtain monitoring images in real time for detection; the third step is to build a ship matching model based on the feature extraction algorithm, and track and analyze the ship template; the fourth step is to analyze the parking status of the target ship.

[0022] S1, set up multiple dual-spectrum thermal imaging cameras in key monitoring waters (i.e. a section of the Yangtze River waterway), which can receive infrared and visible light at the same time, to ensure nighttime detection accuracy. After debugging, set several preset positions for each camera, and the camera automatically patrols and monitors the target waters to achieve multi-angle monitoring perception. When setting the camera position and preset angle, the camera's field of view needs to cover the entire monitoring area, and there should be a certain overlap in the field of view between adjacent cameras or preset positions to avoid blind spots.

[0023] S2, obtain the video frame input into the ship recognition module, perform real-time analysis on the monitoring screen, and detect whether there is a ship in the monitoring screen. If a ship is recognized, record the monitoring image of the frame, the current time of the camera, the rotation angle, the focal length and other parameters, and input them into the ship matching module, otherwise the camera continues to patrol.

[0024] The YOLO v5 (You Only Look Once version 5) algorithm is real-time, accurate, and lightweight. The ship identification module uses the YOLO v5 algorithm to build a ship identification model, which includes the following steps: S21, prepare data set: collect ship images. In this embodiment, data samples include ship images from open source data sets, inland waterway ship images provided by local waterway authorities (for scientific research purposes only), and ship images taken by video surveillance. Then, these images are annotated, and the types of ships are annotated, including cargo ships, sand mining ships, and sand transport ships. The collected images and annotation information are organized into a data set, which is divided into training set, validation set, and test set according to 8:1:1.

[0025] S22, model building and training: Build a model based on the YOLO v5 algorithm, use the training set to train the model, and use the official recommended values ​​of YOLO v5 for the initial model parameter settings, as shown in the following table: Table 1 Initial parameter values ​​of the model

[0026] S23, model evaluation and optimization: Use the validation set and test set to verify and test the trained model. The calculation formulas for each evaluation indicator are as follows:

[0027]

[0028]

[0029] Among them, Precision represents accuracy, Recall represents recall, TP represents the number of correct predictions, FP represents the number of predicted boxes that do not match the true boxes; FN represents the undetected true box: N represents the number of categories, mAP represents the weighted average of P with R as the weight, for example, mAP@0.5 represents the mean average precision when the IoU threshold is 0.5.

[0030] According to the evaluation results, the performance was improved by adjusting the hyperparameters and data preprocessing methods. The evaluation results before and after model tuning are shown in Table 2.

[0031] Table 2 Evaluation indicators before and after model tuning

[0032] S24, model deployment: deploy the trained model to actual applications for real-time ship detection.

[0033] S3, after the ship matching module receives the ship identification signal, it receives the first frame of surveillance image. After patrolling for 30 minutes, the camera is adjusted to the same rotation angle and focal length, and the second frame of surveillance image is captured. The feature points are extracted by the Speeded Up Robust Features (SURF) algorithm and compared with the first frame of surveillance image. If the matching rate exceeds a certain confidence level, it is determined to be the same ship, and the two frames of image information are input into the stagnation analysis module; otherwise, the camera continues to patrol.

[0034] The SURF algorithm is a feature detection algorithm in the field of computer vision and machine learning. The SURF algorithm is based on the Scale-Invariant Feature Transform (SIFT) algorithm, but it is accelerated by using the Box Filter and Integral Image technology to improve the algorithm's operation speed. Specifically, it includes the following steps: S31, preprocessing: preprocess the two frames of monitoring images to improve the accuracy and efficiency of feature extraction. According to the ship outline external frame of the ship recognition module in S2, use white blocks to fill the image background of the non-hull part.

[0035] S32, feature detection: Apply the SURF algorithm to each image to detect the feature points. The SURF algorithm filters and extracts features from images at different scales to generate a series of feature points. For each detected feature point, a 64-dimensional feature vector is generated. These feature vectors contain the local image information of the feature point for subsequent comparison.

[0036] S33, feature matching: compare the feature vectors in the two frames of images to find matching feature point pairs. Use the Hamming distance to measure the similarity of the ship features in the two frames of images. When the number of successful matching points exceeds 60% of the total number of ship feature points in the first frame of image, it can be determined that the ships in the two frames of images are the same ship.

[0037] S4, calculates the displacement of each feature point pair in the two frames of images, obtains the comprehensive displacement by arithmetic averaging, and when the comprehensive displacement is lower than the displacement threshold of 15 pixels, the ship is judged to be in a stagnant state, and a ship stagnation warning signal is issued. The warning time, position, and warning image are automatically recorded, and the video surveillance images 3 minutes before and 40 minutes after the warning time point are intercepted as supporting materials. Figure 3 This is the detection result of sand-carrying ship under visible light condition, where Figure 3 (a) is the first frame image of the visible light-sand transport ship. Figure 3 (b) is the visible light-sand-carrying ship-second frame image; Figure 4 This is the result of the sand mining ship detection application under thermal imaging. Figure 4 (a) is the first frame of thermal imaging - sand mining ship, Figure 4 (b) in the figure is thermal imaging - sand mining ship - second frame image.

[0038] The inland waterway ship stagnation state detection system provided by the present invention is described below. The inland waterway ship stagnation state detection system described below and the inland waterway ship stagnation state detection method described above can be referenced to each other.

[0039] Figure 5 FIG. 1 is a schematic diagram of the structure of the inland waterway ship stagnation state detection system provided by an embodiment of the present invention. Figure 5 As shown, it includes: an identification module 51, a matching module 52 and an analysis module 53, wherein: The recognition module 51 is used to collect ship monitoring images of a designated waterway, identify the target ship in the ship monitoring image using a deep learning algorithm, and obtain a first image frame; the matching module 52 is used to match the target ship after a preset time threshold, and obtain a second image frame; the analysis module 53 is used to perform a stagnation comparison analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result.

[0040] Figure 6 An example of a physical structure diagram of an electronic device is shown in FIG. Figure 6As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute the inland waterway ship stagnation state detection method, which includes: collecting ship monitoring images of a designated waterway, identifying the target ship in the ship monitoring image by using a deep learning algorithm, and obtaining a first image frame; after a preset time threshold, matching the target ship and obtaining a second image frame; performing stagnation comparison analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result.

[0041] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0042] On the other hand, the present invention also provides a computer program product, which includes a computer program. The computer program can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the inland waterway ship stagnation state detection method provided by the above-mentioned methods. The method includes: collecting ship monitoring images of a designated waterway, using a deep learning algorithm to identify a target ship in the ship monitoring image, and obtaining a first image frame; after a preset time threshold, matching the target ship to obtain a second image frame; performing stagnation comparative analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result.

[0043] On the other hand, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, is implemented to execute the inland waterway ship stagnation state detection method provided by the above-mentioned methods, the method comprising: collecting ship monitoring images of a designated waterway, identifying target ships in the ship monitoring images using a deep learning algorithm, and acquiring a first image frame; after a preset time threshold, matching the target ship and acquiring a second image frame; performing stagnation comparative analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result.

[0044] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0045] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0046] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for detecting the stagnant state of an inland waterway ship, It is characterized in that include: Collecting ship monitoring images of a designated waterway, identifying a target ship in the ship monitoring images using a deep learning algorithm, and acquiring a first image frame; After a preset time threshold, matching the target ship to obtain a second image frame; A stagnation comparison analysis is performed on the first image frame and the second image frame to obtain a ship stagnation analysis result.

2. The method for detecting the stagnant state of an inland waterway ship according to claim 1, It is characterized in that Before collecting ship monitoring images of a designated waterway, it also includes: Setting up a plurality of dual-spectrum thermal imaging cameras at a plurality of preset positions of the designated waterway, wherein adjacent dual-spectrum thermal imaging cameras have overlapping field of view coverage areas; The multiple dual-spectrum thermal imaging cameras are used to collect ship navigation video frames, and the ship monitoring images in the ship navigation video frames are extracted.

3. The method for detecting the stagnant state of an inland waterway vessel according to claim 2, It is characterized in that Collecting ship monitoring images of a designated waterway, identifying a target ship in the ship monitoring images using a deep learning algorithm, and acquiring a first image frame, including: The multiple dual-spectrum thermal imaging cameras start shooting, and if the ship monitoring image is obtained, the ship is identified, otherwise the patrol shooting continues; A data set is constructed from a number of ship monitoring images, and the data set is divided into a training set, a validation set, and a test set; Building an initial recognition model based on the YOLO v5 algorithm, training the initial recognition model using the training set based on the initial parameters of the model, and verifying and testing the trained model using the verification set and the test set to obtain a recognition model; The recognition model is used to recognize the ship monitoring image to be recognized, and the first image frame is output.

4. The method for detecting the stagnant state of an inland waterway vessel according to claim 1, It is characterized in that After a preset time threshold, matching the target ship to obtain a second image frame includes: Starting from acquiring the first image frame, after the preset time threshold, adjusting the plurality of dual-spectrum thermal imaging cameras to the same parameter state as that of capturing the first image frame, to acquire the image frame to be identified; Preprocessing the first image frame and the image frame to be recognized to obtain the preprocessed first image frame and the image frame to be recognized; Using the accelerated robust feature SURF algorithm to extract feature vectors of the preprocessed first image frame and the image frame to be identified; The feature vectors are compared to obtain a matching feature point pair, and the second image frame corresponding to the matching feature point pair is determined according to a preset similarity matching degree.

5. The method for detecting the stagnant state of an inland waterway vessel according to claim 4, It is characterized in that Preprocessing the first image frame and the image frame to be recognized to obtain the preprocessed first image frame and the image frame to be recognized, including: Acquire the ship outline circumscribed frames of the first image frame and the image frame to be identified; The non-hull part of the image background in the ship outline circumscribed frame is filled with white blocks to obtain the preprocessed first image frame and the image frame to be identified.

6. The method for detecting the stagnant state of an inland waterway vessel according to claim 4, It is characterized in that The SURF algorithm is used to extract feature vectors of the preprocessed first image frame and the image frame to be identified, including: Performing filtering and feature extraction on the preprocessed first image frame and the image frame to be identified in sequence at different scales to obtain a plurality of feature points; The feature vector of a preset dimension is generated for the plurality of feature points.

7. The method for detecting the stagnant state of an inland waterway vessel according to claim 1, It is characterized in that Performing a stagnation comparison analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result includes: Calculate the displacement of each pair of feature points in the feature vector of the first image frame and the second image frame within the preset time threshold; If the displacement of each pair of feature points is less than or equal to the displacement threshold, ship stagnation warning information is generated, and the ship stagnation warning information includes warning time, warning position and warning image.

8. The method for detecting the stagnant state of an inland waterway vessel according to claim 7, It is characterized in that After performing stagnation comparative analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result, the method further includes: Intercepting video surveillance images of a first time period before the warning time and video surveillance images of a second time period after the warning time; The video surveillance images of the first time period and the video surveillance images of the second time period constitute the supporting video material.

9. A system for detecting the stagnant state of ships in inland waterways. It is characterized in that include: An identification module is used to collect ship monitoring images of a designated waterway, identify a target ship in the ship monitoring images using a deep learning algorithm, and obtain a first image frame; A matching module, used for matching the target ship after a preset time threshold, and acquiring a second image frame; The analysis module is used to perform a stagnation comparison analysis on the first image frame and the second image frame to obtain a ship stagnation analysis result.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the program, the method for detecting the stagnant state of an inland waterway ship as described in any one of claims 1 to 8 is implemented.

Citation Information

Patent Citations

  • Method for preventing container truck from being hoisted based on machine vision

    CN107200274A

  • Traffic sign detection and identification method and system in vehicle driving

    CN111832388A

  • Hull attitude monitoring system and method based on visual technology

    CN115393806A

  • Ship docking system and ship docking method

    US20240182144A1