Navigation mark monitoring method, system and equipment based on image processing and storage medium

Through the image-based navigation beacon monitoring method, the automatic identification and analysis of navigation beacon status is solved, and the problems of strong subjectivity and difficulty of night patrols in traditional inspection methods are achieved, efficient and accurate navigation beacon status monitoring is achieved, and maritime traffic safety is ensured. .

CN119027691APending Publication Date: 2024-11-26XIAMEN NAVIGATION MARK OFFICE EAST CHINA SEA NAVIGATION SUPPORT CENT MINISTRY OF TRANSPORT
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411110503.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-14
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

The traditional navigation beacon inspection method relies on manual visual inspection, which has problems such as strong subjectivity, difficulty in night inspection, and poor photo quality, resulting in inaccurate monitoring of navigation beacon status and low efficiency.

Method used

Using image processing-based navigation beacon monitoring method, the navigation beacon light color recognition model and navigation beacon recognition model are constructed by collecting navigation beacon standard pictures, and the night navigation beacon images are optimized by machine learning algorithms to automatically identify and analyze the navigation beacon light colors and surface coatings. , structure and luminous cycle, and monitor the beacon status in real time.

Benefits of technology

Improve the accuracy and efficiency of navigation beacon monitoring, reduce manual intervention, timely detect abnormal navigation beacon status, ensure maritime traffic safety, optimize night shooting quality, comprehensive monitoring of navigation beacon status, improve emergency response speed and traceability of monitoring data .

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027691A_ABST
    Figure CN119027691A_ABST
Patent Text Reader

Abstract

The invention provides a navigation mark monitoring method, system and device based on image processing and a storage medium, and the method comprises the steps: collecting a navigation mark standard picture, extracting the standard chromaticity information of navigation mark light from the navigation mark standard picture, and constructing a navigation mark light color recognition model based on the standard chromaticity information; using a machine learning algorithm to carry out fusion optimization processing on the night navigation mark pictures obtained by continuous shooting to obtain a first target navigation mark picture; extracting the chromaticity information of the beacon light color in the first target navigation mark picture through a beacon light color identification model; comparing the chromaticity information with a corresponding beacon light color standard chromaticity range to obtain a chromaticity difference value, and marking the chromaticity difference value in the first target beacon picture; in response to determining that the chromaticity difference value is within the first preset range, judging that the navigation mark is normal; otherwise, judging that the navigation mark is abnormal. According to the invention, daily work efficiency of a navigation mark administrator can be improved, and fine management of navigation mark operation is further promoted.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of navigation mark management and monitoring technology, and specifically to a navigation mark monitoring method, system, device and storage medium based on image processing. Background Art

[0002] In the maritime safety assurance system, navigation marks are important navigation facilities. The accurate monitoring and maintenance of their status are directly related to the safety and efficiency of maritime traffic. With the rapid development of the shipping industry and the increase in maritime traffic density, higher requirements are placed on the accuracy and efficiency of navigation mark maintenance management. The traditional navigation mark inspection method mainly relies on manual visual inspection and handheld equipment assistance. However, in the complex and changeable marine environment, this method exposes many technical bottlenecks and shortcomings.

[0003] First, the traditional method of checking the color of the surface of visual navigation marks relies on the subjective judgment of the administrator and lacks objective and accurate quantitative standards. This subjective judgment is easily affected by multiple factors such as light and viewing angle, resulting in inconsistency and uncertainty in the judgment results, making it difficult to ensure accurate evaluation of the vividness of the color of the navigation mark.

[0004] Secondly, during nighttime inspections, the verification of the luminous chromaticity, flashing cycle and flashing rhythm of navigation lights also faces challenges. Existing technical means are often unable to capture and analyze these key parameters in real time and accurately, making it difficult for administrators to quickly determine whether the lights are in normal working condition. This not only increases the difficulty and risk of nighttime inspections, but may also lead to safety hazards due to failure to detect faults in time.

[0005] Furthermore, the bad shooting environment at sea, the shaking of the ship, the interference of the sea wind and the unprofessional operation of the photographer often lead to poor quality of the photos taken, and the navigation mark is often off the center in the picture, which increases the difficulty and time cost of post-processing. This not only affects the efficiency of the inspection, but also may miss important navigation mark status information due to poor photo quality.

[0006] In view of this, the present application proposes a navigation mark monitoring method, system, device and storage medium based on image processing, which can meet the daily operation needs of navigation mark managers and improve the efficiency and accuracy of inspection work. Summary of the invention

[0007] In order to solve the problems of strong subjectivity in existing navigation mark color inspection and difficulty in nighttime inspection, the present application provides a navigation mark monitoring method, system, device and storage medium based on image processing to solve the above-mentioned technical defects.

[0008] According to the first aspect of the present application, a navigation mark monitoring method based on image processing is proposed, the method comprising the following steps:

[0009] S1. Collect standard images of navigation marks, extract standard chromaticity information of navigation mark lights from the standard images of navigation marks, and build a navigation mark light color recognition model based on the standard chromaticity information;

[0010] S2. Using a machine learning algorithm to perform fusion optimization processing on the nighttime navigation mark images obtained by continuous shooting to obtain the first target navigation mark image;

[0011] S3, extracting the chromaticity information of the color of the navigation beacon light in the first target navigation beacon image through the navigation beacon light color recognition model constructed in step S1;

[0012] S4, comparing the chromaticity information with the corresponding standard chromaticity range of the navigation mark light color to obtain a chromaticity difference, and marking the chromaticity difference in the first target navigation mark image;

[0013] S5. In response to determining that the chromaticity difference is within the first preset range, determining that the navigation mark is normal; in response to determining that the chromaticity difference exceeds the first preset range, determining that the navigation mark is abnormal.

[0014] Preferably, the navigation mark monitoring method based on image processing provided by the present application further includes the following steps:

[0015] S6. extracting standard surface color information features and standard structure information features of the navigation mark from the navigation mark standard picture, and constructing a navigation mark recognition model based on the standard surface color information features and the standard structure information features;

[0016] S7, collecting daytime navigation mark pictures in real time, and obtaining the second target navigation mark picture after preprocessing the daytime navigation mark pictures;

[0017] S8, identifying the surface color information features and structural information features of the second target navigation mark image by using the navigation mark recognition model constructed in step S6, determining the category of the current navigation mark, and then comparing the surface color information features with the standard color range values ​​of the corresponding positions based on the category, obtaining the color color difference, and marking the color color difference in the second target navigation mark image;

[0018] S9. In response to determining that the color chromaticity difference is within the second preset range, it is determined that the navigation mark is normal; in response to determining that the chromaticity difference exceeds the second preset range, it is determined that the navigation mark is abnormal.

[0019] Preferably, in step S2, a machine learning algorithm is used to perform fusion optimization processing on the nighttime navigation mark images obtained by continuous shooting, which specifically includes the following sub-steps:

[0020] S21, continuously shooting to obtain a nighttime navigation mark image, wherein the nighttime navigation mark image includes multiple moments when the navigation mark light emits light;

[0021] S22, preprocessing the nighttime navigation mark image, the preprocessing including: resizing, color correction and noise removal, to obtain a preprocessed navigation mark image;

[0022] S23, extracting brightness feature information of the navigation light area in the pre-processed navigation image based on an image segmentation algorithm;

[0023] S24, training a brightness machine learning model based on brightness feature information, and using the trained brightness machine learning model to extract and fuse brightness feature information from multiple pre-processed navigation mark images to obtain a brightness feature map;

[0024] S25, performing noise reduction and image enhancement processing on the brightness feature map, and finally obtaining the first target navigation mark image.

[0025] Further preferably, in step S25, the brightness feature map is subjected to noise reduction and image enhancement processing, which specifically includes the following sub-steps:

[0026] S251, performing noise reduction processing on the brightness feature map by using median filtering or Gaussian filtering to obtain a noise reduction feature map;

[0027] S252, performing contrast enhancement and sharpening processing on the noise reduction feature map to obtain a first target navigation map;

[0028] S253, evaluating the signal-to-noise ratio and clarity index of the first target navigation map, and using the first target navigation map that meets the signal-to-noise ratio and clarity requirements as input data for subsequent image analysis, processing or archiving.

[0029] Preferably, the navigation mark monitoring method based on image processing provided by the present application further includes:

[0030] The lighting status of the beacon lights is captured by image acquisition equipment, and the beacon images are obtained by continuous shooting;

[0031] Automatically identify the first light emission of the navigation light from the night navigation mark pictures through image processing technology and record the time of the first light emission;

[0032] Continue to identify and record the time point of each subsequent lighting of the beacon light until the shooting is completed, and calculate the number of times the light is lit, the time interval between each light, and the time when it is not lit;

[0033] According to the recorded number of light emission and time interval, the actual light emission cycle and flashing rhythm of the navigation light within the preset time period are calculated;

[0034] Compare the actual lighting cycle with the preset standard lighting cycle and calculate the deviation percentage;

[0035] In response to determining that the deviation percentage is within the third preset range, it is determined that the lighting cycle of the beacon light is normal; in response to determining that the deviation percentage exceeds the third preset range, it is determined that the lighting cycle of the beacon light is abnormal.

[0036] Further preferably, the navigation mark monitoring method based on image processing provided by the present application also includes:

[0037] Extracting standard beacon name information from the standard beacon image, the standard beacon name information includes: standard attachment area, standard name and standard size, and building a beacon name recognition model based on the standard beacon name information;

[0038] Extract the navigation mark name target area of ​​the second target navigation mark image by using the constructed navigation mark name recognition model;

[0039] Perform OCR recognition on the names within the target area of ​​the navigation mark to obtain the name content, and compare the name content with the preset standard navigation mark name. At the same time, check whether the font, inclination and integrity of the name content meet the specifications.

[0040] Further preferably, the navigation mark monitoring method based on image processing provided by the present application further includes:

[0041] According to the attribute size of the second target navigation mark image, the second target navigation mark image is split into multiple rectangular grids, and a middle reference line is set;

[0042] Intelligently identify the navigation mark appearing in the second target navigation mark picture through the navigation mark recognition model;

[0043] Move the identified current navigation mark area to the grid area of ​​the middle reference line, so that the navigation mark image is centered in the vertical direction, and remove the two grid areas at the upper and lower edges.

[0044] In a second aspect, the present application provides a navigation mark monitoring system based on image processing, the system comprising:

[0045] The acquisition module is configured to acquire a standard image of a navigation mark, extract standard chromaticity information of the navigation mark light from the standard image of the navigation mark, and construct a navigation mark light color recognition model based on the standard chromaticity information;

[0046] An optimization module is configured to use a machine learning algorithm to perform fusion optimization processing on the nighttime navigation mark images obtained by continuous shooting to obtain a first target navigation mark image;

[0047] An extraction module is configured to extract the chromaticity information of the color of the navigation beacon light in the first target navigation beacon image through the navigation beacon light color recognition model constructed in the acquisition module;

[0048] A comparison module is configured to compare the chromaticity information with the corresponding standard chromaticity range of the navigation light color, obtain a chromaticity difference, and mark the chromaticity difference in the first target navigation mark image;

[0049] The determination module determines that the navigation mark is normal in response to determining that the chromaticity difference is within a first preset range; and determines that the navigation mark is abnormal in response to determining that the chromaticity difference exceeds the first preset range.

[0050] In a third aspect, the present application provides a terminal device, including a processor, a memory, and a computer program stored in the memory, wherein the computer program is executed by the processor to implement any of the above-mentioned image processing-based beacon monitoring methods.

[0051] In a fourth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, a navigation mark monitoring method based on image processing as described above is implemented.

[0052] Compared with the prior art, the beneficial results of this application are:

[0053] (1) Improve monitoring accuracy and efficiency: This application realizes automatic monitoring of the color, surface color, structure and luminous cycle of navigation mark lights by constructing navigation mark light color recognition models and navigation mark identification models. This method not only reduces manual intervention, but also greatly improves the accuracy and efficiency of monitoring, and can timely detect abnormal conditions of navigation marks and ensure maritime traffic safety.

[0054] (2) Optimizing the quality of night photography: In view of the complex night photography environment and weak light, this application uses machine learning algorithms to fuse and optimize the continuously shot night navigation mark images, effectively improving the clarity and stability of the images. Through preprocessing, brightness feature extraction, fusion optimization, and noise reduction enhancement, the quality of night navigation mark images is ensured, providing a reliable data basis for subsequent color recognition and luminous cycle analysis.

[0055] (3) Comprehensive monitoring of the status of the navigation mark: This application not only focuses on the color and lighting cycle of the navigation mark light, but also covers the monitoring of the surface color, structure and name information of the navigation mark. By comprehensively analyzing this information, the overall status of the navigation mark can be comprehensively evaluated, and potential safety hazards can be discovered and dealt with in a timely manner.

[0056] (4) Improve emergency response speed: When an abnormality is found in a navigation mark, this application can quickly provide abnormal information, including light chromaticity difference, surface color difference, luminous cycle deviation, etc., providing a clear basis for decision-making for the management department. This helps to speed up emergency response and reduce maritime traffic accidents caused by navigation mark failures.

[0057] (5) High degree of intelligence and automation: The entire monitoring process is highly intelligent and automated, reducing the complexity and subjectivity of manual operations. Through the application of image processing and machine learning technology, real-time and continuous monitoring of the navigation mark status is achieved, improving the scientificity and standardization of the monitoring work.

[0058] (6) Enhanced traceability of monitoring data: This application records monitoring results in the form of images and data, and marks abnormal information, which enhances the traceability of monitoring data. This provides strong support for subsequent troubleshooting, responsibility tracing, and statistical analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

[0060] Figure 1 is a flow chart of a navigation mark monitoring method based on image processing according to the present application;

[0061] Figure 2 is a structural diagram of a navigation mark monitoring system based on image processing according to the present application;

[0062] Figure 3 It is a structural diagram of a computer system suitable for implementing an electronic device of an embodiment of the present application. DETAILED DESCRIPTION

[0063] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It should also be noted that, for ease of description, only the parts related to the relevant invention are shown in the accompanying drawings.

[0064] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0065] Figure 1 The flowchart of the navigation mark monitoring method based on image processing of the present application is shown as follows: Figure 1 As shown, the method comprises the following steps:

[0066] S1. Collect standard images of navigation marks, extract standard chromaticity information of navigation mark lights from the standard images of navigation marks, and build a navigation mark light color recognition model based on the standard chromaticity information.

[0067] S2. Use a machine learning algorithm to perform fusion optimization processing on the nighttime navigation mark images obtained by continuous shooting to obtain the first target navigation mark image.

[0068] In this embodiment, special processing is performed on the pictures taken at night, and the steps are as follows:

[0069] 1. Long press the shutter of the image acquisition device (e.g. camera), during which the camera will take pictures continuously for 1 to 3 seconds. When the light is on, it will take pictures quickly to form a rapid continuous shooting.

[0070] 2. Use machine learning algorithms to fuse and optimize the continuously taken target beacon photos, integrate the brightness information of each photo, perform noise reduction and image enhancement algorithm processing on the photos, and retain the image information of the light emission of the lamp. Finally, a clear and bright photo with the light emission of the lamp is obtained for electronic archiving.

[0071] Specifically, the nighttime navigation mark images obtained by continuous shooting are fused and optimized using a machine learning algorithm, including the following sub-steps:

[0072] S21, continuously shooting to obtain night navigation pictures, wherein the night navigation pictures include multiple moments when the navigation lights emit light, so that the subsequent algorithm can make full use of the brightness information in different photos. When performing the shooting operation, the user needs to keep pressing the shutter button of the camera, and this long press action will activate the continuous shooting function of the camera. In the following 1 to 3 seconds, the camera maintains an uninterrupted shooting state and continuously captures the picture in front of the lens at a high frequency. In this process, the camera's image sensor and processor work together to ensure that each frame can be accurately recorded and temporarily stored. In particular, the camera has a built-in intelligent recognition system that can monitor the light changes in the surrounding environment in real time. Once it detects that a light is emitted, the intelligent recognition system will immediately trigger the fast capture mechanism. This means that the camera will adjust the exposure parameters in a very short time, optimize the focus, and take multiple photos continuously to capture the dynamic effect of the moment when the light emits light. This fast continuous shooting function can not only effectively reduce the blur caused by light changes, but also help users select the clearest and most expressive one or more photos from multiple continuous photos to improve the shooting effect.

[0073] S22. Preprocess the nighttime navigation mark images. The preprocessing includes: resizing, color correction, and noise removal to obtain preprocessed navigation mark images to ensure the consistency and processability of all input data. Preprocessing the nighttime navigation mark images is one of the important steps in building a navigation mark light color recognition model, which helps to improve the accuracy and robustness of the model. The following are the detailed steps for preprocessing nighttime navigation mark images:

[0074] (1) Resizing is to ensure that all input images have the same size, which helps the model maintain consistency during training and improves processing efficiency.

[0075] First, use an image processing library such as OpenCV or PIL to read the nighttime navigation mark image.

[0076] Second, set an appropriate target size (such as width and height) according to the requirements of subsequent processing steps.

[0077] Finally, use the resizing function in the image processing library to scale the image to the target size. Be careful to maintain the aspect ratio of the image, or crop or pad as needed.

[0078] (2) Since nighttime navigation images may be affected by different lighting conditions, resulting in color deviations, color correction aims to eliminate these deviations and make the colors in the images more accurate and consistent.

[0079] First, analyze the color distribution in the night navigation mark image to determine whether correction is needed.

[0080] Secondly, select a suitable color correction algorithm based on the analysis results. The color correction algorithm can be white balance adjustment, color mapping or histogram equalization.

[0081] Finally, adjust the parameters in the color correction algorithm to obtain the best color correction effect.

[0082] (3) Since pictures taken at night may contain various noises, such as Gaussian noise, etc. Noise removal aims to reduce the impact of these noises on picture quality.

[0083] Select a suitable denoising algorithm according to the type and degree of noise. The denoising algorithm can be median filtering, Gaussian filtering, bilateral filtering or wavelet denoising, etc. When using the denoising function in the image processing library to denoise an image, care should be taken to prevent the details and edge information of the image from being over-smoothed.

[0084] S23, extracting brightness feature information of the beacon light area in the pre-processed beacon image based on the image segmentation algorithm. For brightness optimization, we can focus on features such as brightness histogram, local contrast, and edge information.

[0085] S24. Train a brightness machine learning model based on brightness feature information, the brightness machine learning model including: convolutional neural network (CNN) in deep learning and support vector machine (SVM) and decision tree in traditional machine learning algorithms. Use the trained brightness machine learning model to extract and fuse brightness feature information from multiple pre-processed navigation mark images to obtain a brightness feature map.

[0086] It should be understood that when training the brightness machine learning model, it is necessary to provide a labeled training data set, including comparison photos of the light and the light without light, as well as the desired fusion optimization results. Through the iterative training process, the model parameters are continuously optimized so that it can accurately identify and fuse the brightness information in the photo while retaining the image details of the light.

[0087] After the brightness machine learning model is trained, it is applied to the actual sequence of target beacon photos. The algorithm will analyze the brightness characteristics of each photo and intelligently fuse the brightness information in these photos based on the model parameters obtained through training. During the fusion process, the algorithm will comprehensively consider the dominant areas of each photo, such as the brightest part of the lamp in a photo, the part with clearer background details in another photo, etc., to achieve uniform distribution of global brightness and highlight the lighting effect of the lamp.

[0088] S25, performing noise reduction and image enhancement processing on the brightness feature map, and finally obtaining the first target navigation mark image, specifically including the following sub-steps:

[0089] S251, using median filtering or Gaussian filtering to perform noise reduction processing on the brightness feature map to reduce noise interference in the photo and obtain a noise reduction feature map.

[0090] S252. Perform contrast enhancement and sharpening processing on the noise reduction feature map to improve the overall visual effect of the photo, make the light-emitting part of the lamp more vivid and the background clearer, and obtain the first target navigation map.

[0091] S253. Evaluate the signal-to-noise ratio and clarity index of the first target navigation map to ensure that the lighting information of the lamp is effectively retained and the overall image is clear and bright. The first target navigation map that meets the signal-to-noise ratio and clarity requirements shall be used as input data for subsequent image analysis, processing or archiving.

[0092] Continue to refer Figure 1 The navigation mark monitoring method based on image processing provided by the present application also includes the following steps:

[0093] S3. Extracting the chromaticity information of the color of the navigation beacon light in the first target navigation beacon image through the navigation beacon light color recognition model constructed in step S1.

[0094] S4. Compare the chromaticity information with the corresponding standard chromaticity range of the navigation mark light color to obtain a chromaticity difference, and mark the chromaticity difference in the first target navigation mark image.

[0095] S5. In response to determining that the chromaticity difference is within the first preset range, determining that the navigation mark is normal; in response to determining that the chromaticity difference exceeds the first preset range, determining that the navigation mark is abnormal.

[0096] For example, the standard chromaticity range values ​​of the four colors of red, green, yellow and white beacon lights are set one by one. By collecting the standard chromaticity information of the red, green, yellow and white colors in the beacon lights, a beacon light color recognition model is created. When photographing the light emission at night, the light color is first identified to determine the current beacon light emission color. The chromaticity information of the current light color is extracted from the photographed light and compared with the standard chromaticity range of the corresponding emission color to obtain the corresponding chromaticity difference and display it in the lower right corner of the photo. The chromaticity difference is displayed normally within the first preset range, otherwise it is displayed abnormally. The change in the value of each shot can remind the beacon operation to pay attention to the operation of the light emission and color of the light in advance.

[0097] Before comparing the chromaticity difference, it is necessary to predefine a first preset range. This range is set based on the standard chromaticity range of the beacon light color. For example, if the standard chromaticity value of red is X, then the first preset range is [X-Δ, X+Δ], where Δ is the allowable deviation.

[0098] In a specific embodiment, the navigation mark monitoring method based on image processing provided by the present application further includes the following steps:

[0099] S6. Extract the standard surface color information features and standard structure information features of the navigation mark from the standard image of the navigation mark, and construct a navigation mark recognition model based on the standard surface color information features and the standard structure information features.

[0100] In this embodiment, by collecting standard images of various types of beacons under different weather conditions, corresponding beacon type labels can be set for these different beacons, and then computer vision, image processing and other technologies and algorithms can be used to extract the features of standard surface color information (standard surface color information of top mark, quadpod or tripod and buoy) and standard structural information (relative positions of top mark, quadpod or tripod and buoy) of these different types of beacons, establish a beacon recognition model and continuously train and optimize it to achieve automatic recognition and classification of beacons in photos.

[0101] S7, collecting daytime navigation mark pictures in real time, preprocessing the daytime navigation mark pictures, obtaining second target navigation mark pictures, splitting the second target navigation mark pictures into multiple rectangular grids according to the attribute size of the second target navigation mark pictures, and setting a middle reference line.

[0102] S8, using the navigation mark recognition model constructed in step S6 to recognize the surface color information features and structural information features of the second target navigation mark image, move the recognized current navigation mark area to the grid area of ​​the middle reference line, center the navigation mark image in the vertical direction, and remove the two grid areas at the upper and lower edges. Determine the category of the current navigation mark, and then compare the surface color information features with the standard color range values ​​of the corresponding positions based on the category to obtain the color color difference, and mark the color color difference in the second target navigation mark image.

[0103] S9. In response to determining that the color chromaticity difference is within the second preset range, it is determined that the navigation mark is normal; in response to determining that the chromaticity difference exceeds the second preset range, it is determined that the navigation mark is abnormal.

[0104] For example, the corresponding values ​​of the chromaticity ranges of the colors such as red, black, white, yellow and green specified by the visual beacon surface color are set one by one. By identifying the beacon color information and structural information in the photo, the category of the current beacon is finally determined. The color chromaticity values ​​of the top mark, quadruped or tripod and buoy of the current beacon are respectively compared with the standard surface color chromaticity range of the corresponding positions (top mark, quadruped or tripod and buoy), and the corresponding color chromaticity difference is obtained and displayed in the lower right corner of the photo. The color chromaticity difference is normal within the second preset range, otherwise it is abnormal. The change in the value of each shot can remind the beacon operator to pay attention to the fading of each position of the beacon in advance.

[0105] Before performing the color chromaticity difference comparison, it is necessary to predefine the second preset range. The second preset range can be a numerical interval in one or more color spaces (such as RGB, HSV, Lab, etc.). For example, in the HSV color space, if the hue (H), saturation (S) and brightness (V) of the standard color are H0, S0, and V0 respectively, then the second preset range can be [H0-ΔH, H0+ΔH]×[S0-ΔS, S0+ΔS]×[V0-ΔV, V0+ΔV], where ΔH, ΔS, and ΔV are the allowable deviations respectively.

[0106] In a specific embodiment, the navigation mark monitoring method based on image processing provided by the present application also includes:

[0107] The lighting status of the beacon light is captured by the image acquisition device, and the beacon pictures are obtained by continuous shooting.

[0108] Image processing technology is used to automatically identify the first light emission of the beacon light from nighttime beacon images, and the time point of the first light emission is recorded.

[0109] Continue to identify and record the time point of each subsequent lighting of the beacon light until the shooting is completed, and calculate the number of times the light is lit, the time interval between each light, and the time when it is not lit.

[0110] According to the recorded number of light emission and time interval, the actual light emission cycle and flashing rhythm of the navigation light within the preset time period are calculated. Among them, the light emission cycle is the periodic change of the light emission and extinguishing of the light within a period of time according to a certain rule or pattern. Some lights have a light emission cycle of 1 second, some have a light emission cycle of 12 seconds, and some have a light emission cycle of 6 seconds. The flashing rhythm refers to the rhythmic lighting and extinguishing of the light within a cycle, and the specific rhythm is how many seconds it is on and how many seconds it is off. In actual application, the light emission cycle of the Q1 light buoy in Meizhou Bay is 12 seconds, and its flashing rhythm is 1.5+0.5+1.5+0.5+0.5+0.5+1.5+5.5, that is, it is on for 1.5 seconds at the beginning, off for 0.5 seconds, then on for 1.5 seconds, off for 0.5 seconds, on for 0.5 seconds, off for 0.5 seconds, on for 0.5 seconds, off for 0.5 seconds, on for 1.5 seconds, and off for 5.5 seconds; the light emission cycle of the No. 3 light buoy in the tailwater pipe is also 12 seconds, but its flashing rhythm is 1.5+0.5+1.5+0.5+0.5+1.5+5.5, that is, it is on for 1.5 seconds at the beginning, off for 0.5 seconds, then on for 1.5 seconds, off for 0.5 seconds, on for 0.5 seconds, off for 0.5 seconds, on for 1.5 seconds, and off for 5.5 seconds; although the light emission cycle of the No. 3 light buoy in the tailwater pipe is also 12 seconds, its flashing rhythm is 1.5+0.5+1.5+0.5+0.5+1.5+5.5, that is, it is on for 1.5 seconds at the beginning The light rhythm is 1.5+0.5+0.5+0.5+1.5+0.5+0.5+6.5, i.e. 1.5 seconds on, 0.5 seconds off, 0.5 seconds on, 0.5 seconds off, 1.5 seconds on, 0.5 seconds off, 0.5 seconds on, and finally 6.5 seconds off; one cycle of the Inner Harbor No. 4 Light floating light is 6 seconds, and its flashing rhythm is 0.5+1+0.5+4, i.e. 0.5 seconds on, 1 second off, 0.5 seconds on, and finally 4 seconds off.

[0111] The actual lighting period is compared with the preset standard lighting period and the deviation percentage is calculated.

[0112] In response to determining that the deviation percentage is within the third preset range, it is determined that the lighting cycle of the beacon light is normal; in response to determining that the deviation percentage exceeds the third preset range, it is determined that the lighting cycle of the beacon light is abnormal.

[0113] For example, collect information on various lighting cycles of lamps in the jurisdiction, create a 30S standard model for each flash cycle of navigation lights (the flash cycles of navigation lights in the jurisdiction are: 1S, 2S, 4S, 5S, 6S, 10S, 12S, 15S). Press the shutter for 30S each time you shoot at night, first identify the light of the lamp, and after confirming that the lamp is lit, record the current time, then identify the second light, record the second time, and so on, count the number of times the lamp lights up, the interval between lights, and the time when it does not light up within 30S. Finally, compare with the standard flash rhythm of each flash cycle within 30S, and get the current light cycle and flash rhythm of the navigation light and display them in the lower left corner of the photo. If the error with the standard flash cycle exceeds 5%, the abnormal lighting will be displayed in the photo.

[0114] In a specific embodiment, the navigation mark monitoring method based on image processing provided by the present application also includes:

[0115] Extracting standard beacon name information from the standard beacon image, the standard beacon name information includes: standard attachment area, standard name and standard size, and building a beacon name recognition model based on the standard beacon name information;

[0116] Extract the navigation mark name target area of ​​the second target navigation mark image by using the constructed navigation mark name recognition model;

[0117] Perform OCR recognition on the names within the target area of ​​the navigation mark to obtain the name content, and compare the name content with the preset standard navigation mark name. At the same time, check whether the font, inclination and integrity of the name content meet the specifications.

[0118] For example, the standard areas for attaching the corresponding names of three sizes of buoys with a buoy diameter of 3 meters, 2.4 meters and 1.8 meters are set, and the standard buoy name information of all buoys in the jurisdiction is collected. When photographing a buoy, the structure of the buoy is first identified, and a rectangular red circle is drawn around the standard area of ​​the buoy name according to the buoy name recognition model. The buoy administrator can judge whether the current buoy name is in the standard area and whether the font size is appropriate based on the rectangular red circle. The name recognition is carried out on the structural area where the buoy name is located, and the identified name is compared with the standard buoy name to confirm whether it is correct. If it is not correct, the name abnormality is prompted. The recognition of the name of the buoy includes not only whether the name of the buoy is correct, but also whether the name is written on the buoy in the correct font to prevent unprofessional phenomena such as irregular handwriting, tilted fonts, and partial detachment of the sticker. Preferably, the buoy name identified in the photo is used as the name of the storage folder, and all photos of the buoy name are stored in the folder.

[0119] Further references Figure 2 As an implementation of the above method, in the second aspect, the present application provides an embodiment of a structural diagram of a navigation mark monitoring system 200 based on image processing, which can be specifically applied to various electronic devices. The system 200 includes the following modules:

[0120] The acquisition module 210 is configured to acquire a standard image of a navigation mark, extract standard chromaticity information of the navigation mark light from the standard image of the navigation mark, and construct a navigation mark light color recognition model based on the standard chromaticity information;

[0121] The optimization module 220 is configured to use a machine learning algorithm to perform fusion optimization processing on the nighttime navigation mark images obtained by continuous shooting to obtain a first target navigation mark image;

[0122] The extraction module 230 is configured to extract the chromaticity information of the navigation light color in the first target navigation light picture through the navigation light color recognition model constructed in the acquisition module;

[0123] The comparison module 240 is configured to compare the chromaticity information with the corresponding standard chromaticity range of the navigation mark light color, obtain a chromaticity difference, and mark the chromaticity difference in the first target navigation mark image;

[0124] The determination module 250 determines that the navigation mark is normal in response to determining that the chromaticity difference value is within the first preset range; and determines that the navigation mark is abnormal in response to determining that the chromaticity difference value exceeds the first preset range.

[0125] This application scientifically and conveniently intelligently processes the following operating procedures: judging the requirements for bright color of the surface coating of the navigation mark during daytime inspections, judging the requirements for normal light color and light quality during nighttime inspections, and photographing and archiving navigation marks. This improves the daily work efficiency of navigation mark administrators and further promotes the refined management of navigation mark operations.

[0126] In a third aspect, the present application provides a terminal device, including a processor, a memory, and a computer program stored in the memory, wherein the computer program is executed by the processor to implement any of the above-mentioned image processing-based beacon monitoring methods.

[0127] In a fourth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, a navigation mark monitoring method based on image processing as described above is implemented.

[0128] Reference below Figure 3 , which shows a schematic diagram of the structure of a computer system 300 of a terminal device or server suitable for implementing an embodiment of the present application. Figure 3 The terminal device or server shown is merely an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0129] like Figure 3 As shown, the computer system 300 includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage part 308 into a random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the computer system 300 are also stored. The CPU 301, the ROM 302, and the RAM 303 are connected to each other via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0130] The following components are connected to the I / O interface 305: an input section 306 including a keyboard, a mouse, etc.; an output section 307 including a liquid crystal display (LCD), etc. and a speaker, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed, so that a computer program read therefrom is installed into the storage section 308 as needed.

[0131] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 309, and / or installed from the removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, the above functions defined in the method of the present application are executed. It should be noted that the computer-readable medium described in the present application can be a computer-readable signal medium or a computer-readable medium or any combination of the above two. The computer-readable medium can be, for example, - but not limited to - an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable media may include, but are not limited to, an electrical connection with one or more conductors, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, a computer-readable medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, an apparatus or a device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, an apparatus or a device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0132] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages, such as Java, Smalltalk, C++, and conventional procedural programming languages, such as C or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0133] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0134] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by a specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the above invention concept. For example, the above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other to form a technical solution.

Claims

1. A navigation mark monitoring method based on image processing, characterized in that: The method comprises the following steps: S1. Collecting a standard image of a navigation mark, extracting standard chromaticity information of the navigation mark light from the standard image of the navigation mark, and constructing a navigation mark light color recognition model based on the standard chromaticity information; S2. Using a machine learning algorithm to perform fusion optimization processing on the nighttime navigation mark images obtained by continuous shooting to obtain the first target navigation mark image; S3, extracting the chromaticity information of the color of the navigation beacon light in the first target navigation beacon image through the navigation beacon light color recognition model constructed in step S1; S4, comparing the chromaticity information with the corresponding standard chromaticity range of the navigation light color to obtain a chromaticity difference, and marking the chromaticity difference in the first target navigation mark picture; S5. In response to determining that the chromaticity difference is within a first preset range, determining that the navigation mark is normal; in response to determining that the chromaticity difference exceeds the first preset range, determining that the navigation mark is abnormal.

2. The navigation mark monitoring method based on image processing according to claim 1 is characterized in that: The following steps are also included: S6, extracting standard surface color information features and standard structural information features of the navigation mark from the navigation mark standard picture, and constructing a navigation mark recognition model based on the standard surface color information features and standard structural information features; S7, collecting daytime navigation mark pictures in real time, and obtaining second target navigation mark pictures after preprocessing the daytime navigation mark pictures; S8, identifying the surface color information features and structural information features of the second target navigation mark image by the navigation mark identification model constructed in step S6, determining the category of the current navigation mark, and then comparing the surface color information features with the standard chromaticity range values ​​of the corresponding positions based on the category to obtain color chromaticity difference values, and marking the color chromaticity difference values ​​in the second target navigation mark image; S9, in response to determining that the color chromaticity difference is within a second preset range, determining that the navigation mark is normal; In response to determining that the chromaticity difference value exceeds the second preset range, it is determined that the navigation mark is abnormal.

3. The navigation mark monitoring method based on image processing according to claim 1 is characterized in that: In step S2, a machine learning algorithm is used to perform fusion optimization processing on the nighttime navigation mark images obtained by continuous shooting, which specifically includes the following sub-steps: S21, continuously shooting to obtain a nighttime navigation mark picture, wherein the nighttime navigation mark picture includes multiple moments when the navigation mark light emits light; S22, preprocessing the nighttime navigation mark picture, wherein the preprocessing includes: size adjustment, color correction and noise removal, to obtain a preprocessed navigation mark picture; S23, extracting brightness feature information of the navigation light area in the pre-processed navigation mark image based on an image segmentation algorithm; S24, training a brightness machine learning model based on the brightness feature information, and using the trained brightness machine learning model to extract and fuse brightness feature information from a plurality of pre-processed navigation mark images to obtain a brightness feature map; S25, performing noise reduction and image enhancement processing on the brightness feature map, and finally obtaining the first target navigation mark image.

4. The navigation mark monitoring method based on image processing according to claim 3 is characterized in that: In step S25, noise reduction and image enhancement processing are performed on the brightness feature map, which specifically includes the following sub-steps: S251, performing noise reduction processing on the brightness feature map by using median filtering or Gaussian filtering to obtain a noise reduction feature map; S252, performing contrast enhancement and sharpening processing on the noise reduction feature map to obtain the first target navigation map; S253, evaluating the signal-to-noise ratio and clarity index of the first target navigation map, and using the first target navigation map that meets the signal-to-noise ratio and clarity requirements as input data for subsequent image analysis, processing or archiving.

5. The navigation mark monitoring method based on image processing according to claim 1 is characterized in that: Also includes: The light-emitting state of the navigation mark lamp is captured by an image acquisition device, and the navigation mark pictures are obtained by continuous shooting; Automatically identifying the first light emission of the navigation light from the night navigation mark picture by image processing technology, and recording the time point of the first light emission; Continue to identify and record the time point of each subsequent lighting of the navigation light until the shooting is completed, and calculate the number of lighting times, the time interval between each lighting, and the non-lighting time; According to the recorded number of light emission and time interval, the actual light emission cycle and flashing rhythm of the navigation light within the preset time period are calculated; Comparing the actual lighting cycle with a preset standard lighting cycle, and calculating the deviation percentage; In response to determining that the deviation percentage is within a third preset range, determining that the lighting cycle of the navigation light is normal; In response to determining that the deviation percentage exceeds a third preset range, it is determined that there is an abnormality in the lighting cycle of the navigation light.

6. The navigation mark monitoring method based on image processing according to claim 2 is characterized in that: Also includes: Extracting standard beacon name information of the beacon from the standard beacon image, the standard beacon name information including: standard attachment area, standard name and standard size, and constructing a beacon name recognition model based on the standard beacon name information; Extracting the navigation mark name target area of ​​the second target navigation mark image by using the constructed navigation mark name recognition model; Perform OCR recognition on the names in the target area of ​​the navigation mark name to obtain the name content, and compare the name content with the preset standard navigation mark name, and at the same time verify whether the font, inclination and integrity of the name content meet the specifications.

7. The navigation mark monitoring method based on image processing according to claim 2 is characterized in that: Also includes: According to the attribute size of the second target navigation mark image, split the second target navigation mark image into a plurality of rectangular grids, and set a middle reference line; Intelligently identifying the navigation mark appearing in the second target navigation mark picture through the navigation mark identification model; The identified area where the current navigation mark is located is moved to the grid area of ​​the middle reference line, so that the navigation mark image is centered in the vertical direction, and the areas of two grids at the upper and lower edges are eliminated.

8. A navigation mark monitoring system based on image processing, characterized in that: The system comprises: The acquisition module is configured to acquire a standard image of a navigation mark, extract standard chromaticity information of the navigation mark light from the standard image of the navigation mark, and construct a navigation mark light color recognition model based on the standard chromaticity information; An optimization module is configured to use a machine learning algorithm to perform fusion optimization processing on the nighttime navigation mark images obtained by continuous shooting to obtain a first target navigation mark image; An extraction module configured to extract the chromaticity information of the color of the navigation beacon light in the first target navigation beacon image through the navigation beacon light color recognition model constructed in the acquisition module; A comparison module, configured to compare the chromaticity information with a corresponding standard chromaticity range of the navigation light color, obtain a chromaticity difference, and mark the chromaticity difference in the first target navigation mark picture; The determination module determines that the navigation mark is normal in response to determining that the chromaticity difference is within a first preset range; and determines that the navigation mark is abnormal in response to determining that the chromaticity difference exceeds the first preset range.

9. A terminal device, characterized in that: The method comprises a processor, a memory and a computer program stored in the memory, wherein the computer program is executed by the processor to implement the navigation mark monitoring method based on image processing as claimed in any one of claims 1 to 8.

10. A computer-readable storage medium, wherein a computer program is stored in the medium, and when the computer program is executed by a processor, the navigation mark monitoring method based on image processing as claimed in any one of claims 1 to 8 is implemented.