CN-121997142-A - Ship driving situation awareness-based crewman abnormal behavior identification method and device

CN121997142ACN 121997142 ACN121997142 ACN 121997142ACN-121997142-A

Abstract

The embodiment of the invention discloses a shipman abnormal behavior identification method and device based on ship running situation awareness. The method comprises the steps of dividing a ship deck corresponding to a target ship, positioning a crew in a deck area set to determine the target deck area, extracting behavior characteristics of the crew in the target deck area according to a real-time area video corresponding to the target deck area, extracting task characteristics of each executable task in an executable task list corresponding to the target deck area, determining real-time ship running state information corresponding to the target ship, and generating crew behavior description information aiming at the target deck area according to the crew behavior characteristics, the executable task characteristic list and the real-time ship running state information. This embodiment enables an efficient identification of abnormal behaviour of a crew operating the deck of the vessel.

Inventors

WU XIANGFENG
XIA ZHE
CHEN ZHENG
LI XIAOHONG
CAI YONGGUANG
LIU YAHUI
LI XIAOBIN
QI ZHENG
ZHANG XIAOMING
WU LIXIN
WANG ZHISHENG
TIAN JIANGWEI

Assignees

河北港口港航发展股份有限公司

Dates

Publication Date: 20260508
Application Date: 20260130

Claims (10)

1. The method for identifying the abnormal behavior of the crew based on ship driving situation awareness is characterized by comprising the following steps of: Dividing the ship deck corresponding to the target ship into areas to obtain a deck area set, wherein the area description information corresponding to the deck area comprises boundary fences and executable task lists, the boundary fences represent boundary electronic fences corresponding to the deck area, and the executable tasks represent executable job tasks of crews in the corresponding deck area; Positioning a crew member in the deck area set to determine a target deck area, wherein the target deck area is a deck area containing the crew member; According to the real-time region video corresponding to the target deck region, behavior feature extraction is carried out on crews in the target deck region, and crewman behavior features are obtained; Extracting task characteristics of each executable task in the executable task list corresponding to the target deck area to obtain an executable task characteristic list; determining real-time ship running state information corresponding to the target ship; and generating crew behavior description information aiming at the target deck area according to the crew behavior characteristics, the executable task characteristic list and the real-time ship running state information.
2. The method for identifying abnormal behavior of a crew based on ship driving situation awareness according to claim 1, wherein the method further comprises: Responding to the crewman behavior description information meeting the early warning condition, and sending risk early warning information to the crewman in the target deck area; generating an early warning record according to the risk early warning information, the crewman behavior description information, the real-time ship running state information and the video storage address corresponding to the real-time area video; And storing the early warning record into an in-ship risk record log.
3. The method for identifying abnormal behavior of a crew based on ship driving situation awareness according to claim 2, wherein the performing area division on the ship deck corresponding to the target ship to obtain the deck area set comprises: Acquiring a ship deck plan corresponding to the target ship and a ship operation task list corresponding to the navigation task, wherein a crew movable area and a crew non-movable area are calibrated in the ship deck plan in advance; Performing region segmentation on a crew movable region included in the ship deck plan to obtain a region unit set, wherein the unit sizes of the region units in the region unit set are consistent; for each ship operation task in the ship operation task list, distributing a task tag corresponding to the ship operation task to an area unit bound with the ship operation task in the area unit set; carrying out unit combination on the area units in the area unit set according to the task tag group corresponding to the area units to obtain the deck area set; For each deck area in the deck area set, generating boundary fences included in the deck area corresponding area description information according to the unit positions corresponding to at least one area unit included in the deck area; And for each deck area in the deck area set, determining a duplication elimination label union of at least one task label group corresponding to at least one area unit included in the deck area as an executable task list included in the deck area corresponding area description information.
4. A method of identifying abnormal behavior of a crew based on ship driving situation awareness according to claim 3, wherein said locating the crew of the deck areas in the set of deck areas to determine the target deck area comprises: Acquiring an ultra-wideband signal set, wherein the ultra-wideband signal set is acquired by at least four positioning base stations arranged in the movable area of the crew, and the ultra-wideband signal is sent out by a low-power-consumption positioning tag worn by the crew; Performing signal filtering on each ultra-wideband signal in the ultra-wideband signal set to generate a filtered ultra-wideband signal, thereby obtaining a filtered ultra-wideband signal set; Performing code division signal address division on the filtered ultra-wideband signals in the filtered ultra-wideband signal set to obtain a filtered ultra-wideband signal set, wherein the ultra-wideband signals in the filtered ultra-wideband signal set correspond to the same low-power-consumption positioning tag; for each filtered ultra wideband signal group in the set of filtered ultra wideband signal groups, performing the following processing steps: Determining signal health degree corresponding to each filtered ultra-wideband signal in the filtered ultra-wideband signal group, wherein the signal health degree is used for measuring non-line-of-sight propagation error intensity corresponding to the filtered ultra-wideband signal; positioning the label position of a low-power-consumption positioning label corresponding to the ultra-wideband signal group according to the ultra-wideband signal group after filtering and the signal health degree corresponding to the ultra-wideband signal in the ultra-wideband signal group after filtering, and taking the label position as a crewman position; And performing position matching according to the obtained crew position set and the deck area set to obtain the target deck area.
5. The method for identifying abnormal behavior of a crew based on ship driving situation awareness according to claim 4, wherein the extracting behavior characteristics of the crew in the target deck area according to the real-time area video corresponding to the target deck area to obtain the behavior characteristics of the crew comprises: Determining corresponding environmental state information of the real-time region video during acquisition, wherein the environmental state information represents the region environmental state of a navigation region where the target ship is located; generating an activation factor group according to the environment state information, wherein the activation factors in the activation factor group are used for determining whether the corresponding video enhancement module is activated or not; responding to the characteristics of all the activating factors in the activating factor group without activating the corresponding video enhancement module, and carrying out crew identification on the real-time region video through a target positioning network to obtain the crew behavior characteristics; Responding to the fact that the active factor characterization exists in the active factor group and the corresponding video enhancement module is required to be activated, and carrying out video enhancement on the real-time region video according to the active factor group and a video enhancement network to obtain an enhanced video, wherein the video enhancement network and the target positioning network are included in a behavior feature extraction network, and the video enhancement network contains video enhancement modules with the same number as the active factors in the active factor group; and performing crew identification on the enhanced video through a target positioning network to obtain the crew behavior characteristics.
6. The method for identifying abnormal behavior of a crew based on ship driving situation awareness according to claim 5, wherein the task feature extraction is performed on each executable task in the executable task list corresponding to the target deck area, and the method comprises the steps of: performing position coding on a task execution position corresponding to the executable task to obtain a task execution position characteristic; Performing time coding on task execution time corresponding to the executable task to obtain task execution time characteristics; Performing information coding on task related equipment information corresponding to the executable task to obtain task related equipment information characteristics; extracting features of the environment-dependent information of the executable task to obtain environment-dependent information features; performing feature stitching on the task execution position feature, the task execution time feature, the task related equipment information feature and the environment dependent information feature to obtain stitching features; And performing feature compression on the spliced features to obtain executable task features corresponding to the executable tasks in the executable task feature list.
7. The method for identifying abnormal behavior of a crew based on ship driving situation awareness according to claim 6, wherein the generating crew behavior description information for the target deck area according to the crew behavior feature, the executable task feature list and the real-time ship driving state information comprises: extracting the ship state characteristics of the real-time ship running state information to obtain ship state running state characteristics; Performing feature similarity matching according to the crew behavior feature and the executable task feature list to determine a crew behavior evaluation factor; Generating an environmental risk factor according to the ship state running state characteristics; and generating the crewman behavior description information according to the crewman behavior evaluation factor and the environment risk factor.
8. The utility model provides a shipman's abnormal behavior recognition device based on boats and ships driving situation perception which characterized in that includes: The regional division unit is configured to divide the ship deck corresponding to the target ship into regions to obtain a deck region set, wherein region description information corresponding to the deck region comprises boundary fences and executable task lists, the boundary fences represent boundary electronic fences corresponding to the deck region, and the executable tasks represent executable job tasks of crews in the corresponding deck region; a crew positioning unit configured to perform crew positioning on a deck area in the deck area set to determine a target deck area, wherein the target deck area is a deck area containing crew; The behavior feature extraction unit is configured to extract behavior features of crews in the target deck area according to the real-time area video corresponding to the target deck area, so as to obtain the behavior features of the crews; The task feature extraction unit is configured to extract task features of each executable task in the executable task list corresponding to the target deck area to obtain an executable task feature list; A determining unit configured to determine real-time ship travel state information corresponding to the target ship; And the generation unit is configured to generate crew behavior description information aiming at the target deck area according to the crew behavior characteristics, the executable task characteristic list and the real-time ship running state information.
9. An electronic device, comprising: One or more processors; A storage device having one or more programs stored thereon; when executed by the one or more processors, causes the one or more processors to implement the method of any of claims 1 to 7.
10. A computer readable medium, characterized in that a computer program is stored thereon, wherein the computer program, when executed by a processor, implements the method according to any of claims 1 to 7.

Description

Ship driving situation awareness-based crewman abnormal behavior identification method and device Technical Field The embodiment of the disclosure relates to the technical field of computers, the field of video processing and the field of machine learning, in particular to a method and a device for identifying abnormal behavior of a crew based on ship driving situation awareness. Background Ocean-going shipping has become one of the major transportation modes of global trade by virtue of its cost and traffic advantages. In this case, since the ship is required to travel on the ocean for a long time, a crew is required to perform routine inspection and maintenance on the ship state, the cargo state, and the like, so as to ensure the safety of the ship and the cargo. However, when the security requirement of the inspection maintenance is violated in the process of the inspection maintenance of the crew, or accidents such as injury of the crew are very easy to occur due to severe and complicated sea conditions and weather influence, how to timely and effectively identify abnormal behaviors of the crew becomes one of the important issues for protecting the operation security of the crew. Disclosure of Invention The disclosure is in part intended to introduce concepts in a simplified form that are further described below in the detailed description. The disclosure is not intended to identify key features or essential features of the claimed subject matter, nor is it intended to be used to limit the scope of the claimed subject matter. Some embodiments of the present disclosure propose a method and apparatus for identifying abnormal behavior of a crew based on ship driving situation awareness, to solve the technical problems mentioned in the background section above. According to the technical scheme, the method comprises the steps of dividing a ship deck corresponding to a target ship into a deck area set, obtaining a boundary fence and an executable task list, wherein the boundary fence represents a boundary electronic fence of the corresponding deck area, the executable task represents a work task executable by a crew in the corresponding deck area, positioning the deck area in the deck area set to determine a target deck area, wherein the target deck area is the deck area containing the crew, extracting behavior characteristics of the crew in the target deck area according to a real-time area video corresponding to the target deck area, obtaining the behavior characteristics of the crew, extracting the task characteristics of each executable task in an executable task list corresponding to the target deck area, obtaining the executable task characteristic list, determining real-time ship running state information corresponding to the target ship, and generating the ship deck description information for the target ship according to the behavior characteristics of the crew, the executable task characteristic list and the real-time ship running state information. In a second aspect, some embodiments of the present disclosure provide a device for identifying abnormal behavior of a crew based on ship driving situation awareness, where the device includes a region dividing unit configured to divide a ship deck corresponding to a target ship into a set of deck regions, where region description information corresponding to the deck regions includes boundary fences and executable task lists, the boundary fences characterize boundary fences of the corresponding deck regions, the executable tasks characterize executable task of a crew in the corresponding deck regions, a crew positioning unit configured to perform crew positioning on the deck regions in the set of deck regions to determine a target deck region, where the target deck region is a deck region including a crew, a behavior feature extracting unit configured to perform behavior feature extraction on the crew in the target deck region according to a real-time region video corresponding to the target deck region, to obtain crew behavior features, a task feature extracting unit configured to perform task feature extraction on each executable task in the executable task list corresponding to the target deck region, to obtain executable task features, a determining unit configured to determine the target deck region as a target deck region, and generate the behavior feature of the crew on the ship according to the real-time state information corresponding to the target deck region, and the behavior feature information describing the crew on the ship driving situation. In a third aspect, some embodiments of the present disclosure provide an electronic device comprising one or more processors, and storage means having one or more programs stored thereon, which when executed by the one or more processors, cause the one or more processors to implement the method described in any of the implementations of the first aspect. In a fourth aspect, some embodiments of the pr