User authentication system

Abstract

A user authentication system is disclosed. The user authentication system acquires big data by accumulating response messages received from users utilizing video captured aboard a vessel for user authentication, and utilizes the acquired big data for vessel navigation by an artificial intelligence (AI) installed on the vessel.

Description

User authentication system

[0001] The present invention relates to a security technology that distinguishes whether a user attempting to connect is a legitimate user using an image.

[0002] Figure 1 shows an image of a conventional CAPTCHA.

[0003] Referring to Figure 1, CAPCHA (Completely Automated Public Turing test to tell Computers and Humans Apart) is a security technology that distinguishes whether a user is automated software (bot) or not. It sends a message to the user that can only be responded to if the user is not automated software (bot), and if a true response message is received from the user, it determines that the user is a legitimate user and blocks attacks by automated software (bot).

[0004] reCAPTCHA is a CAPTCHA system developed by Carnegie Mellon University that displays distorted word images to users and requires them to input words exactly as they appear. The data received from users was utilized to digitize books that cannot be scanned by computer. reCAPTCHA helped digitize the New York Times archives and was subsequently used for a similar purpose by Google Books.

[0005] Meanwhile, artificial intelligence installed in autonomous vessels, smart ships, and unmanned vessels utilizes video footage captured by the ship for navigation purposes, such as collision avoidance. In order for AI to use video footage for navigation, a large amount of feature point data is required; however, conventionally, human labeling of the extracted feature points was essential.

[0006] The present invention aims to provide a user authentication system that utilizes images captured on a ship for user authentication and utilizes big data of response messages received from the user for the operation of an artificial intelligence installed on the ship.

[0007] According to one aspect of the present invention for achieving the above objective, a user authentication system is provided that utilizes images captured on a ship for user authentication to acquire big data accumulated from response messages received from a user, and utilizes the acquired big data for ship operation by an artificial intelligence installed on the ship.

[0008] The above user authentication system may include: an image analysis unit that analyzes images taken on the vessel to acquire screenshots or moving images to be used for generating an authentication message; an authentication message generation unit that generates the authentication message using the screenshots or moving images acquired by the image analysis unit; and a response message analysis unit that analyzes the response message written by the user in response to the authentication message and converts it into data; and the big data of the response message converted into data by the response message analysis unit can be utilized for the operation of the vessel by the artificial intelligence installed on the vessel.

[0009] The above user authentication system may further include a response message verification unit that checks whether the response message is true; and a response message collection unit that collects response messages confirmed to be true among the response messages; and the response message analysis unit may analyze the response messages confirmed to be true and convert them into data.

[0010] The above image analysis unit can define the bow or stern, which has a fixed position in the image taken from the vessel, as a fixed body and extract the outline of the fixed body as a reference line.

[0011] The above image analysis unit can extract feature points of the outer part of the reference line excluding the fixed body.

[0012] The above image analysis unit can generate a grid from an image captured on the vessel.

[0013] The above-mentioned image analysis unit plays back an image captured on the vessel, and if a change occurs in the image captured on the vessel, it can record the position of the changed part on a grid.

[0014] The above image analysis unit can generate a grid by positioning one of the horizontal lines of the grid at the horizon position in the image taken from the vessel.

[0015] Changes occurring below the aforementioned horizon can be judged as more important information by giving them more weight than changes occurring above the aforementioned horizon.

[0016] The above-mentioned image analysis unit plays an image captured on the vessel, and if a change occurs in the image captured on the vessel, it can obtain a screenshot of the image containing the change, and the above-mentioned authentication message generation unit can generate an authentication message using the screenshot of the image containing the change obtained by the above-mentioned image analysis unit.

[0017] The above-mentioned image analysis unit plays an image captured on the vessel, and if a change occurs in the image captured on the vessel, it can acquire a moving image containing the changed part, and the above-mentioned authentication message generation unit can generate an authentication message using the moving image containing the changed part acquired by the above-mentioned image analysis unit.

[0018] The authentication message using the above-mentioned moving image may include selecting whether one or more of the following expressions are correct: “weather is cloudy”; “weather is clear”; “obstacle is approaching”; “obstacle is moving away”; “obstacle is a ship”; and “obstacle is a reef”.

[0019] When a user attempts to connect to a connection target, the connection target may request the authentication message from the user authentication system.

[0020] When the above user authentication system receives an authentication message request from the above connection target, it can generate the authentication message and transmit the authentication message to the above user.

[0021] When the user generates a response message to the authentication message and transmits the response message to the user authentication system, the user authentication system can collect the response message, and the big data of the response message collected by the user authentication system can be utilized for the operation of the artificial intelligence installed on the ship.

[0022] When the user generates a response message to the authentication message and transmits the response message to the user authentication system, the user authentication system can verify the response message received from the user, determine whether it is true or false, and then transmit the result of verifying the response message to the connection target.

[0023] The above connection target may authenticate the user and allow access if the result of verifying the response message received from the user authentication system is true, and may block access by not authenticating the user if it is false.

[0024] According to the present invention, big data accumulated from response messages received from users is obtained by utilizing images captured on a ship for user authentication, and the obtained big data can be utilized for the operation of the ship by an artificial intelligence installed on the ship.

[0025] In particular, when the user authentication system of the present invention is applied to a connection target (such as a ship's remote system) where the majority of users are experts in ships, accurate data accumulation by a group of experts is possible.

[0026] Figure 1 shows an image of a conventional CAPTCHA.

[0027] FIG. 2 schematically illustrates a user authentication system according to one embodiment of the present invention.

[0028] FIG. 3 schematically illustrates the image analysis process of a user authentication system according to one embodiment of the present invention.

[0029] FIG. 4 schematically illustrates the baseline extraction, feature point extraction, and grid generation processes of a user authentication system according to an embodiment of the present invention.

[0030] FIG. 5 schematically illustrates a user authentication method according to an embodiment of the present invention.

[0031] The structure and operation of a preferred embodiment of the present invention will be described in detail below with reference to the attached drawings. The present invention can be applied in various ways in fields such as shipbuilding and security. Furthermore, the following embodiments may be modified in various different forms, and the scope of the present invention is not limited to the following embodiments.

[0032]

[0033] The present invention utilizes video captured on a ship for user authentication to acquire big data by accumulating response messages received from users, and utilizes the acquired big data for the operation of an artificial intelligence installed on the ship. This will be examined in detail below.

[0034]

[0035] FIG. 2 schematically illustrates a user authentication system according to one embodiment of the present invention.

[0036] Referring to FIG. 2, the user authentication system of the present embodiment may include one or more of an image analysis unit (110), an authentication message generation unit (120), a response message verification unit (130), a response message collection unit (140), and a response message analysis unit (150).

[0037] The video analysis unit (110) analyzes the video taken on the ship and obtains a screenshot or moving video to be used to generate an authentication message. This will be described later with reference to FIGS. 3 and FIGS. 4.

[0038] The authentication message generation unit (120) uses a screenshot or moving image obtained by the image analysis unit (110) to generate an authentication message that can determine whether a user attempting to access a target (a remote system of a vessel, a website, an application, etc.) is a legitimate user and not an automated software (bot).

[0039] The response message verification unit (130) checks whether the response written by the user for the authentication message is true.

[0040] The response message collection unit (140) collects response messages that are confirmed as True among the response messages written by the user for the authentication message.

[0041] The response message analysis unit (150) analyzes the response messages confirmed as true collected by the response message collection unit (140) and converts them into data. The big data of the response messages converted into data by the response message analysis unit (140) can be used for ship operation (feature point extraction, marine environment judgment, collision avoidance, etc.) of artificial intelligence installed in autonomous ships, smart ships, unmanned ships, etc.

[0042]

[0043] FIG. 3 schematically illustrates the image analysis process of a user authentication system according to an embodiment of the present invention, and FIG. 4 schematically illustrates the baseline extraction, feature point extraction, and grid generation processes of a user authentication system according to an embodiment of the present invention.

[0044] Referring to FIG. 3, the image analysis process of the user authentication system of the present embodiment is performed by an image analysis unit (110) and may include one or more of the following steps: extracting a reference line (S310), extracting a feature point (S320), generating a grid (S330), playing an image (S340), determining whether a change has occurred in the image (S350), recording the position of the changed part on the grid (S360), obtaining a screenshot or moving image of the image containing the changed part (S370), and determining whether the image has ended (S380). The image of a ship used for image analysis in the present invention may be an original image in which navigation information is stored without loss.

[0045] 1) Step of extracting a baseline (S310) and step of extracting a feature point (S320)

[0046] As illustrated in FIG. 4, images captured from a ship have the characteristic that the position of the bow (or stern) is fixed. To utilize this characteristic of images captured from a ship, the fixed bow (or stern) is defined as a fixed body, and the outline of the fixed body is extracted as a reference line (BL) (Fig. 4 (b)). Additionally, feature points are extracted from the part outside the reference line (BL) (i.e., the fixed body is excluded).

[0047] 2) Step to generate the grid (S330)

[0048] Images captured from a ship are characterized by the fact that there is little change in the sea and sky in most of the images, and the position of the horizon is fixed. Therefore, as shown in FIGS. 4(b) and (c), one of the horizontal lines of the grid is positioned at the horizon (H), and then the grid is created by dividing it into equal lengths horizontally and vertically. The number of vertical divisions of the grid may be greater than the number of horizontal divisions, or the number of vertical and horizontal divisions above the horizon (sky portion) may be the same (identical to 5 in FIG. 4(c)).

[0049] 3) a step of playing the video (S340) and a step of determining whether a change has occurred in the video (S350),

[0050] Video footage captured from the vessel is played back to determine whether a change has occurred in the footage. Most of the footage captured from the vessel consists of static images where the bow (or stern) is fixed and the rest of the background is filled with the sky and sea (excluding changes in clouds and sea surface ripples). A change in the footage can be determined if a flying object (airplane, drone, etc.) appears in the sky or if an obstacle (other vessel, reef, island, etc.) appears in the sea.

[0051] If a change occurs in the video (S350’s ‘Yes’), the position of the changed part on the grid is recorded, and if no change occurs in the video (S350’s ‘No’), the video is played continuously to find the point where the change occurs.

[0052] 4) A step of recording the position of the changed part on the grid (S360) and a step of acquiring a screenshot or moving image of the video with the changed part (S370)

[0053] If a change occurs in the video (S350 'Yes'), the position of the changed part on the grid can be recorded (S360). Additionally, if a change occurs in the video (S350 'Yes'), a screenshot or moving video of the video containing the changed part can be obtained (S370). The obtained screenshot or moving video is used to generate an authentication message in the authentication message generation unit (120).

[0054] When operating a ship, changes in the sea may be more important than changes in the sky (e.g., when an obstacle appears ahead and the ship must change direction to avoid a collision). Therefore, changes occurring below the horizon (H) (sea portion) can be given more weight than changes occurring above the horizon (H) (sky portion) to be considered more important information.

[0055] 5) A step of determining whether the video has ended (S380)

[0056] While playing the video until the video ends (S340), determine whether a change has occurred in the video (S350). If a change has occurred in the video (S350 'yes'), record the position of the changed part on the grid (S360) and obtain a screenshot or moving video of the video containing the changed part (S370).

[0057]

[0058] FIG. 5 schematically illustrates a user authentication method according to an embodiment of the present invention.

[0059] Referring to FIG. 5, the user authentication method of the present embodiment comprises one of the following steps: a step in which a user (U) attempts to connect to a connection target (A) (S710); a step in which the connection target (A) requests an authentication message from the user authentication system (100) (S720); a step in which the user authentication system (100) generates an authentication message (S730); a step in which the user authentication system (100) transmits an authentication message to the user (U) (S740); a step in which the user (U) generates a response message (S750); a step in which a response message is transmitted from the user (U) to the user authentication system (100) (S760); a step in which the user authentication system (100) checks the response message received from the user (U) (S770); a step in which the user authentication system (100) collects the response message (S780); a step in which the user authentication system (100) transmits the result of checking the response message to the connection target (A) (S790); and a step in which the connection target (A) authenticates or disauthenticates the user (U) (S800). It may include the above.

[0060] 1) A step (S710) in which a user (U) attempts to connect to a connection target (A) and a step (S720) in which the connection target (A) requests an authentication message from the user authentication system (100)

[0061] When a user (U) attempts to access a connection target (a remote system of a vessel, a website, an application, etc.) (S710), the connection target (A) requests an authentication message from the user authentication system (100) to determine whether the user is a legitimate user and not an automated software (bot) (S720).

[0062] 2) A step in which the user authentication system (100) generates an authentication message (S730) and a step in which the user authentication system (100) sends an authentication message to the user (U) (S740)

[0063] When the user authentication system (100) receives a request for an authentication message from a connection target (A) (S720), the authentication message generation unit (120) of the user authentication system (100) transmits an authentication message (S730) generated using a screenshot of a video with a change part obtained through the video analysis process of the video analysis unit (110) or a moving video (S370 of FIG. 3) to the user (U) (S740).

[0064] For example, the authentication message may provide a screenshot divided into a grid as in (c) of Fig. 4 and make a request such as "Select all tiles with obstacles."

[0065] The user authentication system (100) of the present invention may provide a moving image rather than a screenshot and request a response from the user (U). For example, a request such as “Look at the image and select the correct expression from the following: the weather is cloudy, the weather is clear, an obstacle is approaching, an obstacle is moving away, the obstacle is a ship, the obstacle is a reef” may be made.

[0066] The authentication message generation unit (120) of the present invention may generate an authentication message by combining weather information on the date the image was captured and information from the ship's Automatic Identification System (AIS) with image information. By collecting the authentication message generated in this way and obtaining big data, the distortion value of the image due to weather changes can be obtained. Even on the same route, the marine environment may change, such as the color of the sea changing depending on the weather; by applying the distortion value of the image due to weather changes, the precision of the ship's operation by artificial intelligence can be increased.

[0067] 3) A step in which the user (U) generates a response message (S750) and a step in which the response message is transmitted from the user (U) to the user authentication system (100) (S760)

[0068] The user checks the authentication message and generates a response message with the answer requested in the authentication message (S750), and the generated response message is transmitted from the user (U) to the user authentication system (100) (S760).

[0069] 4) A step (S770) in which the user authentication system (100) checks the response message received from the user (U)

[0070] The response message verification unit (130) of the user authentication system (100) verifies the response message received from the user (U) and determines whether it is true or false. If the response message is true, the user can be considered a legitimate user, such as a person, and if the response message is false, the user can be considered not a legitimate user, such as automated software (bot).

[0071] 5) Step (S780) in which the user authentication system (100) collects the response message

[0072] The response message collection unit (140) of the user authentication system (100) collects response messages that are determined to be true among the response messages received from the user (U). The response messages collected by the response message collection unit (140) are transmitted to the response message analysis unit (150) and converted into data, and the response messages analyzed and converted into data by the response message analysis unit (150) are utilized for ship operation (feature point extraction, marine environment judgment, collision avoidance, etc.) of artificial intelligence installed in autonomous ships, smart ships, unmanned ships, etc.

[0073] 6) A step in which the user authentication system (100) transmits a response message confirmation result to the connection target (A) (S790) and a step in which the connection target (A) authenticates or disauthenticates the user (U) (S800)

[0074] The user authentication system (100) transmits the true / false result of the response message determined by the response message verification unit (130) to the connection target (A) (S790), and the connection target (A) authenticates the user (U) and allows the connection if the result of the response message verification received from the user authentication system (100) is true, and blocks the connection by not authenticating the user (U) if the result of the response message verification received from the user authentication system (100) is false.

[0075]

[0076] According to the present invention, big data accumulated from response messages received from users is obtained by utilizing images captured on a ship for user authentication, and the obtained big data can be utilized for the operation of the ship by an artificial intelligence installed on the ship.

[0077] In particular, when the user authentication system of the present invention is applied to a connection target (such as a ship's remote system) where the majority of users are experts in ships, accurate data accumulation by a group of experts is possible.

[0078]

[0079] It is obvious to those skilled in the art that the present invention is not limited to the above embodiments and may be implemented with various modifications or variations within the scope of the technical essence of the present invention.

Claims

1. A user authentication system that utilizes video captured on a ship for user authentication to acquire big data accumulated from response messages received from users, and utilizes the acquired big data for ship operation by artificial intelligence installed on the ship.

2. In Claim 1, A video analysis unit that analyzes images taken from the above vessel to acquire screenshots or moving images to be used for generating authentication messages; An authentication message generation unit that generates the authentication message using the screenshot or the moving image obtained by the image analysis unit; and A response message analysis unit that analyzes and converts into data the response message written by the user in response to the authentication message; A user authentication system that utilizes big data of the above-mentioned response messages, converted into data by the above-mentioned response message analysis unit, for the operation of the ship by artificial intelligence installed on the ship.

3. In Claim 2, A response message verification unit that checks whether the above response message is true; and Further including a response message collection unit that collects response messages confirmed as true among the above response messages, The above response message analysis unit is a user authentication system that analyzes the response message confirmed as true and converts it into data.

4. In Claim 2, The above image analysis unit is, A user authentication system that defines a fixed bow or stern section as a fixed body in an image taken from the above-mentioned vessel, and extracts the outline of the fixed body as a reference line.

5. In Claim 4, The above image analysis unit is, A user authentication system that extracts feature points from the outer part of the reference line excluding the fixed body.

6. In Claim 2, The above image analysis unit is, A user authentication system that generates a grid from images taken on the above-mentioned vessel.

7. In Claim 6, The above image analysis unit is, A user authentication system that plays video captured on the above-mentioned vessel and, if a change occurs in the video captured on the above-mentioned vessel, records the position of the changed part on a grid.

8. In Claim 6, The above image analysis unit is, A user authentication system that generates a grid by positioning one of the horizontal lines of a grid at the horizon position in an image taken from the above vessel.

9. In Claim 8, A user authentication system that determines the change portion occurring below the horizontal line as more important information by giving it more weight than the change portion occurring above the horizontal line.

10. In Claim 2, The above-mentioned image analysis unit plays back an image captured on the vessel, and if a change occurs in the image captured on the vessel, acquires a screenshot of the image containing the changed part, and A user authentication system in which the authentication message generation unit generates an authentication message using a screenshot of an image containing the change portion obtained by the image analysis unit.

11. In Claim 2, The above image analysis unit plays back an image captured on the vessel, and if a change occurs in the image captured on the vessel, acquires a moving image containing the changed part, and The above authentication message generation unit is a user authentication system that generates an authentication message using a moving image having the above-mentioned change portion obtained by the above-mentioned image analysis unit.

12. In Claim 11, The authentication message using the above moving image is, "Cloudy weather"; "The weather is clear"; "Obstacles approaching"; "Obstacles moving away"; "Obstacles are ships"; and "Obstacles are reefs"; A user authentication system that includes selecting whether one or more of the expressions are correct.

13. In Claim 2, A user authentication system in which, when a user attempts to connect to a connection target, the connection target requests the authentication message from the user authentication system.

14. In Claim 13, The above user authentication system is, A user authentication system that, upon receiving an authentication message request from the above-mentioned connection target, generates the authentication message and transmits the authentication message to the above-mentioned user.

15. In Claim 14, When the above user generates a response message to the above authentication message and transmits the response message to the above user authentication system, the above user authentication system collects the response message, and A user authentication system in which big data of the response messages collected by the above user authentication system is utilized for the operation of an artificial intelligence installed on a ship.

16. In Claim 14, A user authentication system that, when the user generates a response message to the authentication message and transmits the response message to the user authentication system, the user authentication system verifies the response message received from the user, determines whether it is true or false, and then transmits the result of verifying the response message to the connection target.

17. In Claim 16, The above connection target is a user authentication system that authenticates the user and allows access when the result of confirming the response message received from the above user authentication system is true, and unauthenticates the user and blocks access when it is false.

Images