Airship Artificial Intelligence Detection System and Detection Method

Abstract

The disclosed document of the present invention provides an airship artificial intelligence detection system and a detection method. By recording the navigation images of a standard airship and a test airship, and then through the analysis method of artificial intelligence, the change curve of the angle between the airship body and the water surface and the change curve of the pixel points of the highest point of the airship bow and the highest point of the water surface are iteratively judged in the model, so as to realize the intelligent detection of the attitude of the aerodynamic airship. The airship artificial intelligence detection system of the present invention uses artificial intelligence technology to add an intelligent analysis function to the production quality detection of airships, avoiding the influence of human factors, improving the detection accuracy. At the same time, it uses high-definition image data for intelligent analysis, does not require fixed equipment to be installed on the test airship, and only needs to use a high-definition camera on the shore or on the shooting airship to shoot the images of the test airship and the standard airship, thus reducing the detection cost. The system also uses 5G communication technology to upload high-definition image data and download the intelligent analysis results, greatly improving the detection efficiency.

Description

Technical Field

[0001] The present invention relates to a quality detection device and method for an airboat, and particularly to a comprehensive production quality detection device and method for an airboat. Background Art

[0002] An airboat is a device that can sail smoothly in complex waters such as shallow waters, shoals, floating ice, ice and snow, and dense waterweeds. The research and development of airboats in China is still in its infancy. Due to the different structures and driving methods of airboats from ordinary boats, the hull of an ordinary boat is heavier, and the engine is mainly placed underwater to drive the boat by propelling the water flow. While an airboat places a high-power engine on the upper part of the hull and propels the movement through the aerodynamic force generated by the propeller. Due to the particularity of the mechanical structure and power form of the airboat, the quality of the airboat plays a very important decisive role in the operation safety of the airboat. At present, there are no suitable means and methods for testing the structural performance and maneuverability of airboats at home and abroad. There are generally two detection methods. One is the method of visual observation by humans, which analyzes and compares the differences in the structural forms and maneuverability between the test boat type and the prototype. This method is affected by many factors such as the environment, psychological activities, and experience. And to obtain relatively accurate conclusions, it is necessary to record the airboat driving data multiple times and have multiple people conduct visual analysis, which is time-consuming, laborious, has poor objectivity, and extremely low data reliability. The improvement of product performance has been greatly hindered, and the production capacity has increased slowly. The other is the sensing acquisition detection method, which mainly installs multiple sensors for displacement, speed, and acceleration on the test boat to collect the boat's driving attitude data, and uploads it to the upper computer PC terminal through wireless communication technology, and professional personnel organize, analyze, and judge the production quality of the test boat. This method is not conducive to the production quality detection of large-scale airboats due to the complex installation of sensors and long time consumption. Also, because high-precision sensors must be selected to improve the detection accuracy and they are expensive, this method is only suitable for laboratory use.

[0003] In summary, the currently commonly used visual observation method and sensing acquisition method cannot meet the requirements of airboat production quality detection. There is an urgent need for a device and method that can quickly and accurately detect the boat body state. Summary of the Invention

[0004] To solve the problems existing in the prior art, the present application's disclosure provides an airboat artificial intelligence detection system and detection method. By recording the navigation images of the standard boat and the test boat, and through the analysis method of artificial intelligence, the change curve of the angle between the boat body and the water surface and the change curve of the pixel points of the highest point of the boat head and the highest point of the water surface are iteratively judged in the model, so as to realize the intelligent detection of the airboat's boat attitude.

[0005] An airboat artificial intelligence detection system for detecting the navigation state of an airboat during navigation; it includes:

[0006] An acquisition module, configured to acquire the navigation videos and images of the test boat and the standard boat;

[0007] A processing module, configured to store and analyze the navigation videos and images of the test boat and the standard boat acquired by the acquisition module, form the parameters of the navigation processes of the test boat and the standard boat, and compare the parameters of the navigation processes of the test boat and the standard boat to obtain a test result;

[0008] A data management module, configured to display and retrieve the parameters and the test result; and

[0009] A transmission module, configured to transfer data among the acquisition module, the processing module and the data management module.

[0010] Furthermore, the acquisition module includes a shooting boat and a camera disposed on the shooting boat.

[0011] Furthermore, the processing module includes an image processing module and a storage module, wherein,

[0012] The image processing module is configured to process the images of the test boat and the standard boat acquired by the acquisition module, perform algorithm analysis, and determine whether the quality of the test boat is qualified according to the analysis result;

[0013] The storage module is configured to store the images of the test boat and the standard boat acquired by the acquisition module, the data and the test result processed by the image processing module.

[0014] Furthermore, the algorithm analysis is to perform iterative judgment on the change curves of the angles between the hulls of the test boat and the standard boat and the water surface and the change curves of the pixel points of the highest points of the boat heads and the high points of the water surface in a model, so as to obtain the parameter comparison between the test boat and the standard boat during the driving process, and use it to determine whether the quality of the test boat is qualified.

[0015] Furthermore, the data management module is an application program installed in a computer or a mobile phone

[0016] An artificial intelligence detection method for an air boat, comprising the following steps:

[0017] Step 1, acquire the dynamic images of the standard boat during navigation;

[0018] Step 2, store the navigation images of the standard boat;

[0019] Step 3, acquire the dynamic images of the test boat during navigation;

[0020] Step 4, store the navigation images of the test boat;

[0021] Step 5, repeat Steps 3 to 4 until the tests of all test boats are completed;

[0022] Step 6: Process the navigation images of the standard boat and each test boat to generate the navigation data of the standard boat and the test boats.

[0023] Step 7: Compare the navigation reference data of the standard boat with the navigation data of the test boats to draw a conclusion on whether the test boats are qualified.

[0024] Step 8: Store the conclusion and display it through a display terminal.

[0025] Furthermore, Step 1 is to use a shooting boat to travel simultaneously with the standard boat in the same water area, and use a camera installed on the shooting boat to capture the side sailing posture of the standard boat.

[0026] Furthermore, Step 3 is to use a shooting boat to travel simultaneously with the test boats in the same water area, and use a camera installed on the shooting boat to capture the side sailing posture of the test boats.

[0027] Furthermore, Step 6 specifically uses an artificial intelligence video analysis algorithm to calculate the change curve of the angle between the hull of the standard boat and the water surface during the driving process, the change curve of the pixel points of the highest point of the bow and the highest point of the water surface, the change curve of the angle between the hull of the test boat and the water surface, and the change curve of the pixel points of the highest point of the bow and the highest point of the water surface.

[0028] Furthermore, Step 7 is to perform iterative judgment on the data in Step 6 in the model, so as to obtain the parameter comparison between the test boats and the standard boat during the driving process, and use the data to judge the quality of the test boats.

[0029] The airboat artificial intelligence detection system provided by the present invention basically solves the problems and disadvantages existing in the prior art in the human eye observation method and the sensing acquisition method: the human eye observation method is greatly affected by many factors such as the environment, psychological activities, and experience, with very low detection accuracy and low efficiency. Although the sensing acquisition method has high detection accuracy, due to its high cost, low efficiency, poor applicability and other disadvantages, it cannot be applied in the large-scale airboat production quality detection. The airboat artificial intelligence detection system of the present invention uses artificial intelligence technology to add an intelligent analysis function to the airboat production quality detection, avoiding the influence of human factors and improving the detection accuracy. At the same time, it uses high-definition image data for intelligent analysis, does not require fixed equipment to be installed on the test boats, and only needs to use a high-definition camera on the shore or on the shooting boat to capture the images of the test boats and the standard boat, thereby reducing the detection cost. The system also uses 5G communication technology to upload high-definition image data and download intelligent analysis results, greatly improving the detection efficiency. Description of the Drawings

[0030] Figure 1 It is a schematic structural diagram of an airboat artificial intelligence detection system disclosed by the present invention.

[0031] Figure 2 Schematic flow diagram of an airship artificial intelligence detection method disclosed by the present invention.

[0032] Figure 3 Schematic diagram of standard airship detection.

[0033] Figure 4 Schematic diagram of test airship detection.

[0034] Figure 5 Schematic diagram of the positions of airship body detection points. Detailed implementation manners

[0035] The present invention will be described in detail below with reference to the accompanying drawings. The airships mentioned hereinafter include standard airships and test airships. The standard airship is an airship with the same model as the test airship and qualified quality. The system uses the test data of the standard airship as the judgment criterion to judge the quality of the test airship. The test airship is a number of airships to be tested to determine whether the quality is qualified.

[0036] As Figure 1 shown, the airship artificial intelligence detection system of the present invention is used to detect the navigation state of an airship during navigation; it includes a collection module, a processing module, a data management module, and a transmission module.

[0037] As Figures 2 - 4 shown, the collection module is used to collect the navigation videos and images of the test airship 105 and the standard airship 102, including: a high-definition camera 103, a shooting boat 101, and a positioning buoy 104. Among them, the high-definition camera 103 can also be replaced by other devices that can collect videos and images of fast-moving objects; preferably, the high-definition camera is a camera with a 4k resolution. The high-definition camera 103 is usually arranged on the shooting boat 101 for shooting the side driving postures of the standard airship 102 and the test airship. Of course, the camera 103 can also be arranged on a car or other mobile tools that can travel along the shore; or directly fixed at a set position on the shore. The positioning buoy 104 is arranged at a set position in the shooting water area for determining the reference point for shooting and data processing, that is, recording the starting points of the navigation of the test airship 105 and the standard airship 102.

[0038] As Figure 1 、 2 shown, the processing module is used to store and analyze the navigation videos and images of the standard airship 102 and the test airship 105 collected by the collection module 1, and perform intelligent analysis and calculation based on the image data to judge whether the test airship is qualified. The processing module includes an image processing module and a storage module. The storage module is used to store the images taken by the high-definition camera 103 and the data and test results after being processed by the image processing module. As Figure 2 、 3As shown, the video processing module can analyze and process images according to requirements. The main steps include data cleaning of video or image data in the storage module, refining data features to form a change curve of the required data, and iteratively judging whether the quality of the test boat is qualified in the model group.

[0039] For example, preferably, the change curve of the angle between the hull and the water surface and the change curve of the pixel points of the highest point of the boat head and the highest point of the water surface are calculated through the algorithm of artificial intelligence video analysis. The data is iteratively judged in the model to obtain the parameter comparison between the test boat and the standard boat during the driving process, and the quality of the test boat is judged by the data, so as to realize the detection of whether the boat posture of the test boat is qualified.

[0040] Such as Figure 1 、 2 As shown, the data management module is used to display and retrieve the images collected by the collection module and the various parameters and test results obtained by the processing module through image processing. The data management module is usually an application installed on mobile phones and computers, and can also be viewed by logging in to a web page. The data displayed by the data management module mainly includes: a rowing boat record list: displaying various rowing boat record information stored in the processing module such as the operation name, boat number, rowing date, average speed, average height, average angle, height curve similarity, and angle curve similarity. It also has functions such as creating a new standard boat or test boat, retrieving the various data of each boat; exporting boat information and detection records, etc.

[0041] The transmission module is used to transfer data between the collection module, the processing module, and the data management module. Usually, devices that can wirelessly transmit data such as 4G or 5G mobile networks are used. Of course, wired transmission devices can also be used.

[0042] It should be noted that the system of the present invention will be carried out on the standard boat and each test boat under the same or similar weather conditions. The weather environment mentioned here refers to the weather factors that affect the boat speed, such as: wind speed, temperature, water surface width, etc. To avoid increasing test errors due to weather conditions. Multiple groups of tests can also be carried out under various weather conditions according to the actual boat type requirements.

[0043] Such as Figure 3 、 4 As shown, during the test, the shooting boat (vehicle) will start at the same time as the standard boat 102 and each test boat. The high-definition camera on the shooting boat (vehicle) is used to shoot the standard boat and each test boat. After passing the starting point determined by the positioning buoy 104, it is necessary to ensure that the standard boat or each test boat is always within the shooting range of the high-definition camera. Preferably, the shooting boat (vehicle) travels on the water surface at the same speed as the test boat or the standard boat.

[0044] Preferably, the shooting is carried out in a closed test water area.

[0045] The attitude of the hull collected by the acquisition module, preferably the side attitude, is input into the processing module through the transmission module. After the processing module cleans and arranges the image, the key points of the bow are extracted through the hull contour extraction, so that the distance from the bow to the bottom of the hull and the angle with the water surface can be calculated, and a curve changing with time is formed. The similarity of the height and angle between the bow and the water surface of the standard boat and the test boat is calculated through the dynamic time warping algorithm, so as to determine the difference between the driving attitude of the test boat and the standard boat.

[0046] It should be noted that in a two-dimensional picture, due to the lack of three-dimensional information, the real distance from the bow to the water surface cannot be measured. Alternatively, the algorithm can output the ratio of the distance from the bow to the bottom of the hull to the height of the hull. This value has the same physical meaning as the actual distance and can approximately approximate the effect of the actual distance.

[0047] Detection method

[0048] As shown in the figure, preparation work needs to be carried out before the test, including: 1. Prepare rescue equipment for the test personnel; 2. Water surface positioning markers: floating buoys; 3. Prepare test tools: anemometer, rangefinder, meter stick; 4. Prepare recording equipment: high-definition cameras, video cameras, cameras, etc.

[0049] Standard boat data acquisition

[0050] The steps to collect the speed of the standard airboat are as follows:

[0051] (1) Before the test, use an anemometer to detect the wind speed of the test site, and select the speed of the hull at the detection point during the test. For specific details, please refer to Table 1. The reason for testing the wind speed: The airboat is propelled by aerodynamic force, so the sailing speed of the hull at the detection point is affected by the wind speed and water flow speed. The influence of the wind speed on the airboat is greater than that of the water flow, about 1.2 - 1.3 times that of the water flow. When the airboat sails against the wind, its speed is the still-air sailing speed minus the wind speed value, that is, the speed indicated on the speedometer. When the airboat sails with the wind, its sailing speed is the still-air speed plus the wind speed value, that is, the speed indicated on the speedometer.

[0052] Serial number Wind speed level Measuring point hull speed 1 Levels 0 - 2 40 Km / h 2 Level 3 30 Km / h 3 Level 4 20 Km / h

[0053] Table 1

[0054] (2) As shown in Figure 3, set the positioning float 104 on the test water surface to determine the reference point for shooting and data processing.

[0055] (3) As shown in Figure 3, place the high-definition camera 103 or video camera and its bracket in the acquisition module on the shooting boat 101 at a fixed position to shoot the side driving attitude of the standard boat;

[0056] It should be noted that in the test, the maximum sailing speed of the standard boat and the test boat can also be used as a detection index in the airship artificial intelligence detection technology. The height of this test point is related to the overall height and width of the boat. Generally, as the height and width of the hull increase, the height of the test point relative to the bottom of the boat must also increase. The angle should be determined by the specific structure of the airship, related to many aspects: power, water resistance, center of gravity, wind resistance, etc. For ordinary airships, this angle range is between 25° and 35°, preferably between 27° and 33°.

[0057] For example: When the size of the airship is 6.03m × 2.21m × 2.7m (length x width x height) and the fastest sailing speed value is in the range of 65Km / h - 70Km / h, then when the sailing attitude of the test point on the hull satisfies that the vertical water surface height of the test point is 760 ± 20mm and the angle between the test point and the water surface is 30° ± 1°, it is regarded as the maximum speed index of the test point on the hull being qualified.

[0058] (4) Take the driving image (attitude) of the standard boat: As Figure 3 、 5 shown, the standard boat and the shooting boat start from the starting point (i.e., the position of the positioning buoy 104) at the same sailing speed and drive straight. There should be a set distance between the shooting boat 101 and the standard boat 102. Preferably, the distance between the two boats is 8 - 10m. When shooting, both the shooting boat 101 and the standard boat 102 need to maintain a constant sailing speed so that the detection point 106 of the standard boat 102 is always within the shooting range of the high-definition camera 103 on the shooting boat. Starting from the starting point, the shooting boat 101 and the standard boat 102 must sail a set distance. When the sailing speed is between 65 - 75km, the sailing distance is preferably 200 - 1000 meters, preferably 300 - 500 meters, and most preferably 375 meters.

[0059] (5) Transmit the captured driving image of the standard boat to the image processing module;

[0060] (6) Take the driving attitude of the test boat: As Figure 4 、 5 shown, the test boat 105 and the shooting boat 101 start from the starting point (i.e., the position of the positioning buoy 104) at the same sailing speed and drive straight. There should be a set distance between the shooting boat 101 and the test boat 105. Preferably, the distance between the two boats is 8 - 10m. When shooting, both the shooting boat 101 and the test boat 105 need to maintain a constant sailing speed so that the detection point 106 of the test boat 105 is always within the shooting range of the high-definition camera 103 on the shooting boat. Starting from the starting point, the shooting boat 101 and the test boat 105 must sail a set distance. When the sailing speed is between 65 - 75km, the sailing distance is preferably 200 - 1000 meters, preferably 300 - 500 meters, and most preferably 375 meters.

[0061] (7) Transmit the captured navigation images of the standard boat to the image processing module;

[0062] Repeat steps (5) and (6) until all test boats complete the test;

[0063] (8) After the shooting is completed, perform pre - image processing on the high - definition images of the standard boat and each test boat, such as data cleaning, etc. After processing, transmit the data to the image processing module through the transmission module for data processing, such as using artificial intelligence AI for reasoning and drawing the pixel point map of the boat body running track. Use the algorithm of artificial intelligence video analysis to iteratively judge the change curve of the angle between the boat body and the water surface and the pixel point change curve of the highest point of the boat head and the water surface high point in the model. Preferably, use the dynamic time warping algorithm (DTW) to obtain the parameter comparison between the test boat and the standard boat during the driving process, and use the data to judge the quality of the production boat. The height similarity between the standard boat and the test boat ≤ 10, and the angle similarity between the standard boat and the test boat ≤ 100. If both of the above two indicators meet the requirements, it is regarded as the production quality of the test boat being qualified.

[0064] (9) Store and display data such as test results in the data management module, and users can edit, retrieve, and export relevant data through the data management module.

[0065] From the principle and process of the above - mentioned artificial intelligence detection system, it can be seen that the system detection operation is very simple. Since the artificial intelligence algorithm analyzes high - definition image data, it avoids the influence of human factors on the detection results, thus improving the accuracy of system detection. Also, during the detection process, only a high - definition camera is needed to shoot the high - definition video images of the driving postures of the test boat and the standard boat, and upload them to the image processing module through 5G communication technology for data processing. Due to the combined effect of efficient 5G communication transmission and powerful computing power, the detection efficiency of the power performance of the air boat has been greatly improved. At the same time, the development of the detection system's mobile terminal enables the detection of the power performance of the air boat not to be limited to the detection center. High - definition video images can be uploaded at any location and at any time, and intelligent algorithms can be used for analysis. Then, the detection results can be viewed at any time and anywhere, and multiple people can also supervise the detection process and results through corresponding permissions, avoiding errors during the detection process that affect the detection results.

[0066] Beneficial effects

[0067] 1. Labor saving: The time cost has been reduced from more than 5 people discussing repeatedly for nearly a week in the past to 2 or 3 people discussing for as little as 1 day. Calculated based on the original production scale of 50 sets per year and the detection cost, the production volume can be increased to 150 sets per year, and the detection cost per set can be saved by 24,000 yuan. It is estimated that it can save about 3.6 million yuan in cost and increase efficiency for the enterprise every year.

[0068] 2. The application of artificial intelligence technology to assist in the detection of the overall performance of powerboats represents a breakthrough from scratch in the industry. It has improved the manufacturing level of airboats, breaking the industry status quo where the observation of air-powered boats by human eyes lacks objective judgment and data support. It has extremely enhanced the accuracy of detection and judgment. As a result, the company's authority in the airboat industry can be greatly enhanced, consolidating the leading position of the company's airboats in the industry and increasing their potential market value.

[0069] 3. Social benefits of the project

[0070] Currently, the application fields of airboats mainly focus on national and local emergency rescue agencies at all levels, border defense forces, reserve forces, army training bases, the Ministry of Environmental Protection, civilian rescue teams and other departments. Through user feedback, it shows that this equipment has strong power, is flexible and maneuverable, easy to use, and has good passability in complex media when performing special tasks such as rescue, reconnaissance and patrol, delivery, and handling emergencies in conditions such as shallow water marshes, ice floe waters, ice and snow interfaces, and complex environmental waters. It is an innovative product with a wide range of applications, advanced performance, reliable quality, and strong practicality. The airboat structure analysis and driving attitude AI detection program established by using artificial intelligence can achieve on-site comparison, and directly analyze the results on the mobile phone. Almost accurate analysis and comparison results can be obtained simultaneously when uploading data, and it only takes 1 hour from the test to obtaining the final results. As a result, both the product quality and production capacity of airboats have been greatly improved. As a device applied in complex waters such as shallow water, shoals, ice floes, ice and snow, and dense waterweeds, airboats have incomparable advantages in performing tasks such as emergency rescue, flood fighting and disaster relief, ice and snow rescue, command and detection, wetland patrol, and reconnaissance and patrol in the above special environments.

[0071] Due to technical limitations and the product being in the initial stage of promotion, the production capacity of airboats is relatively low. In ice and snow rescue, flood fighting and disaster relief, shallow water rescue, command and monitoring and other operations, rescue soldiers mostly use inflatable boats as water rescue equipment, and its limitation is that it cannot meet the rescue requirements in complex waters. With the large-scale production of airboats, the above problems will be solved. While people's lives and property are protected, the status of emergency rescue teams in the hearts of the people will also be further enhanced.

[0072] The development of the AI detection program has further improved the performance and product output of airboats. The use of this product will produce more extensive social benefits in ensuring people's lives and property safety and reducing the losses of the country and people's lives and property.

[0073] 4. The first attempt to apply artificial intelligence technology to airship products will be a good start. Currently, the R & D team of our company is developing a driverless airship. Relying on artificial intelligence technology will greatly shorten the R & D process of driverless airships. In the future, the informatization level of airships will become higher and higher. Electric airships and amphibious airships that can travel on land, water, and in the air will also be realized one by one with the progress of technology.

[0074] Since the present invention is susceptible of various modifications, as described above, and numerous apparently widely different embodiments thereof are possible within the scope of the claims without departing from that scope, it is intended that all matter contained in the specification shall be interpreted as illustrative only and not in a limiting sense.

Claims

1. An airship artificial intelligence detection system for detecting the navigation status of an airship during navigation ； It includes: A collection module for collecting navigation videos and images of a test airship and a standard airship; A processing module; For storing and analyzing the navigation videos and images of the test airship and the standard airship collected by the collection module, forming parameters of the navigation processes of the test airship and the standard airship, and comparing the parameters of the navigation processes of the test airship and the standard airship to obtain a test result; The processing module includes: an image processing module and a storage module, where, The image processing module is used to process the images of the test airship and the standard airship collected by the collection module, and perform algorithm analysis, and judge whether the quality of the test airship is qualified according to the analysis result; The storage module is used to store the images of the test airship and the standard airship collected by the collection module, the data and test results processed by the image processing module; The algorithm analysis is to iteratively judge the change curves of the angles between the hulls of the test airship and the standard airship and the water surface and the change curves of the pixel points of the highest points of the bow and the highest points of the water surface in the model, so as to obtain the parameter comparison between the test airship and the standard airship during the driving process, and use it to judge whether the quality of the test airship is qualified; A data management module for displaying and retrieving the parameters and test results; and A transmission module for transmitting data between the collection module, the processing module and the data management module.

2. The detection system according to claim 1, Characterized in that, The collection module includes a shooting airship and a camera arranged on the shooting airship.

3. The detection system according to claim 1, Characterized in that, The data management module is an application program installed in a computer or a mobile phone.

4. An airship artificial intelligence detection method using the detection system according to claim 2, Including the following steps: Step 1, collect dynamic images of the standard airship during navigation; Step 2, store the dynamic images of the standard airship during navigation; Step 3, collect dynamic images of the test airship during navigation; Step 4, store the dynamic images of the test airship during navigation; Step 5, repeat steps 3 to 4 until all test airships are tested; Step 6, process the navigation images of the standard airship and each test airship to generate navigation data of the standard airship and the test airship. Specifically, use the algorithm of artificial intelligence video analysis to calculate the change curve of the angle between the hull of the standard airship and the water surface and the change curve of the pixel points of the highest point of the bow and the highest point of the water surface during the driving process, and the change curve of the angle between the hull of the test airship and the water surface and the change curve of the pixel points of the highest point of the bow and the highest point of the water surface; Step 7, compare the navigation reference data of the standard airship and the navigation data of the test airship to obtain a conclusion on whether the test airship is qualified; Step 8, store the conclusion and display it through a display terminal.

5. The detection method according to claim 4, Characterized in that, In step 1, use the shooting airship to travel in the same water area as the standard airship at the same time, and use the camera arranged on the shooting airship to shoot the side driving posture of the standard airship.

6. The detection method according to claim 4, Characterized in that, Step 3 is to have the shooting boat and the test boat travel in the same water area simultaneously, and use the camera installed on the shooting boat to capture the side sailing posture of the test boat.

7. According to the detection method described in claim 4, characterized in that step 7 is to perform iterative judgment on the data in step 6 in the model, so as to obtain the parameter comparison between the test boat and the standard boat during the driving process, and use the data to judge the quality of the test boat.

Images