Zheping Shao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zheping Shao has a total of 16 co-authors. Here are the top co-authors by number of shared publications:
- Jiacai Pan
Navigation Institute, Jimei University, Xiamen 361021, China
4 shared publications Xianbiao Ji
Jimei University, No.1 Jiageng Road, XIAMEN, China, 361021
3 shared publicationsCunbao Tang
Jimei University, No.1 Jiageng Road, Jimei, XIAMEN, China, 361021
2 shared publicationsXianxin Huang
Yangtze River Pilot Center Pilot Station, Taicang, Jiangsu Province, Taicang City, Yangzhou Road 53-8, 215400
2 shared publicationsJia-Cai Pan
Jimei University, No.1 Jiageng Road, Jimei, XIAMEN, China, 361021;
1 shared publicationJiacai Pan
Jimei University, No.1 Jiageng Road, Jimei, XIAMEN, China, 361021
1 shared publication
Latest Publications
- Open Access
- 16 March 2023
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 11 (3) , 625
A Novel Intelligent Ship Detection Method Based on Attention Mechanism Feature Enhancement
The intelligent perception ability of the close-range navigation environment is the basis of autonomous decision-making and control of unmanned ships. In order to realize real-time perception of the close-range environment of unmanned ships, an enhanced attention mechanism YOLOv4 (EA-YOLOv4) algorithm is proposed. First of all, on the basis of YOLOv4, the convolutional block attention module (CBAM) is used to search for features in channel and space dimensions, respectively, to improve the model’s feature perception of ship targets. Then, the improved-efficient intersection over union (EIoU) loss function is used to replace the complete intersection over union (CIoU) loss function of the YOLOv4 algorithm to improve the algorithm’s perception of ships of different sizes. Finally, in the post-processing of algorithm prediction, soft non-maximum suppression (Soft-NMS) is used to replace the non-maximum suppression (NMS) of YOLOv4 to reduce the missed detection of overlapping ships without affecting the efficiency. The proposed method is verified on the large data set SeaShips, and the average accuracy rate of mAP0.5–0.95 reaches 72.5%, which is 10.7% higher than the original network YOLOv4, and the FPS is 38 frames/s, which effectively improves the ship detection accuracy while ensuring real-time performance.
- 16 March 2023
- Published by MDPI AG in Journal of Marine Science and Engineering
- Vol. 11 (3) , 625
Open AccessA Novel Intelligent Ship Detection Method Based on Attention Mechanism Feature Enhancement
The intelligent perception ability of the close-range navigation environment is the basis of autonomous decision-making and control of unmanned ships. In order to realize real-time perception of the close-range environment of unmanned ships, an enhanced attention mechanism YOLOv4 (EA-YOLOv4) algorithm is proposed. First of all, on the basis of YOLOv4, the convolutional block attention module (CBAM) is used to search for features in channel and space dimensions, respectively, to improve the model’s feature perception of ship targets. Then, the improved-efficient intersection over union (EIoU) loss function is used to replace the complete intersection over union (CIoU) loss function of the YOLOv4 algorithm to improve the algorithm’s perception of ships of different sizes. Finally, in the post-processing of algorithm prediction, soft non-maximum suppression (Soft-NMS) is used to replace the non-maximum suppression (NMS) of YOLOv4 to reduce the missed detection of overlapping ships without affecting the efficiency. The proposed method is verified on the large data set SeaShips, and the average accuracy rate of mAP0.5–0.95 reaches 72.5%, which is 10.7% higher than the original network YOLOv4, and the FPS is 38 frames/s, which effectively improves the ship detection accuracy while ensuring real-time performance.
- 21 September 2015
- Published by American Society of Civil Engineers (ASCE)
- p. 2057-2064
Lateral Distribution Regularity of Ships’ Sailing Positions within a Channel Based on AIS Data
In order to offer the general view and regularity of choosing right transverse position and reduce the uncertainty of choosing lateral position to the ship officers when navigating within the two-way fairway, with the theory of marine traffic engineering and data mining technology, the regularity of transverse distribution of ship’ sailing position within the two-way fairway is mined and counted using the shipborne AIS data. The regularity reveals positive effect quantitatively on the safe navigation of different ships caused by the Regulations for Preventing Collisions at Sea, 1974 and extent of different length of ship complying with rules. The result can be used for choosing the safe and suitable transverse sailing position when the ship goes through two-way fairway and also for marine traffic organization, layout and planning of fairway conducted by the authorities.
- 21 September 2015
- Published by American Society of Civil Engineers (ASCE)
- p. 2057-2064
Lateral Distribution Regularity of Ships’ Sailing Positions within a Channel Based on AIS Data
In order to offer the general view and regularity of choosing right transverse position and reduce the uncertainty of choosing lateral position to the ship officers when navigating within the two-way fairway, with the theory of marine traffic engineering and data mining technology, the regularity of transverse distribution of ship’ sailing position within the two-way fairway is mined and counted using the shipborne AIS data. The regularity reveals positive effect quantitatively on the safe navigation of different ships caused by the Regulations for Preventing Collisions at Sea, 1974 and extent of different length of ship complying with rules. The result can be used for choosing the safe and suitable transverse sailing position when the ship goes through two-way fairway and also for marine traffic organization, layout and planning of fairway conducted by the authorities.
- 9 October 2013
- Published by American Society of Civil Engineers (ASCE)
- p. 44-51
Very Large Vessels into River Navigable Environment Analysis and Security Control Methods Research
With the Yangtze river estuary within 12.5 meters deep water channel to inward extend, make into the river's quantity, tonnage of vessel and vessel's draft also continues to increase, especially the super large vessels into the river, has a certain influence on navigation environment, brings new challenge to the pilot personnel, and creates higher requirements on piloting skills. Based on the AIS data statistics of Yangtze estuary is 12.5m deep-water channel of vessel traffic characteristic. Combination of Yangtze estuary is 12.5m deep water channel and characteristics of various performance, analysis of ultra large vessel through the Yangtze estuary is 12.5 m deep water channel into river navigable environment. For very large vessels sailing on the Yangtze estuary the 12.5 m deep water channel influences the navigational safety of vessel problem, from the viewpoint of fluid dynamics, the analysis influence the navigational safety of vessel increase absorption principle, put forward how to avoid the passing vessels phenomena of adverse influence on navigation safety. Combined with the split type pilot's actual experience, in the actual voyage by radar, AIS observation speed ratio and effective outlook discover the phenomenon of passing vessels and take appropriate measures to achieve safe navigation. In addition, aiming at this kind of vessel maneuvering characteristics puts forward operating precautions to ensure safe navigation.
- 9 October 2013
- Published by American Society of Civil Engineers (ASCE)
- p. 44-51
Very Large Vessels into River Navigable Environment Analysis and Security Control Methods Research
With the Yangtze river estuary within 12.5 meters deep water channel to inward extend, make into the river's quantity, tonnage of vessel and vessel's draft also continues to increase, especially the super large vessels into the river, has a certain influence on navigation environment, brings new challenge to the pilot personnel, and creates higher requirements on piloting skills. Based on the AIS data statistics of Yangtze estuary is 12.5m deep-water channel of vessel traffic characteristic. Combination of Yangtze estuary is 12.5m deep water channel and characteristics of various performance, analysis of ultra large vessel through the Yangtze estuary is 12.5 m deep water channel into river navigable environment. For very large vessels sailing on the Yangtze estuary the 12.5 m deep water channel influences the navigational safety of vessel problem, from the viewpoint of fluid dynamics, the analysis influence the navigational safety of vessel increase absorption principle, put forward how to avoid the passing vessels phenomena of adverse influence on navigation safety. Combined with the split type pilot's actual experience, in the actual voyage by radar, AIS observation speed ratio and effective outlook discover the phenomenon of passing vessels and take appropriate measures to achieve safe navigation. In addition, aiming at this kind of vessel maneuvering characteristics puts forward operating precautions to ensure safe navigation.
- 9 October 2013
- Published by American Society of Civil Engineers (ASCE)
- p. 2752-2760
Mining Algorithm of Ship Arrival Pattern Based on AIS Data
The analysis of the pattern of vessels' passing through a main channel using the traffic volume and time differences between successive arrival vessels can provide theoretical basis and support for maritime authorities in protecting the marine navigation environment, reducing ship accidents, and improving the efficiency of traffic control. Currently, with the popularity of the AIS (Automatic Identification System) equipment installed, AIS information could well reflect the characteristics of vessel traffic in ports. This paper examines vessel arrival pattern with details and accuracy using AIS information. AIS information in port waters collected by AIS receiving equipment and INTERNET in real time is processed and then stored in the database for further study. By the ETL (Extract Transform Load) technique, AIS data is transferred from the original database to the data warehouse, which can support the complex multidimensional query and analysis functions. According to the characteristics of the AIS data and by using OLAP (On-Line Analytical Processing) technology, the observation records of each ship near the threshold in a certain time range are searched and listed in order, so as to make statistics on the AIS data for traffic through the threshold of the main channel. Time differences of each vessel through the threshold can be quickly calculated using the algorithm of time-interval between ships. This paper, taking the vessel arrival regularity in the main channel of Xiamen port as a study case, analyzes the time-interval between ships entering/leaving the port, draws the histogram for the time-interval between the ships, finds out that the density distribution of time-interval between the ships entering/leaving the port is subject to a negative exponential distribution, and validates the algorithm and the model. The result can be used as a basis for the maritime authority in ship traffic control.
- 9 October 2013
- Published by American Society of Civil Engineers (ASCE)
- p. 2752-2760
Mining Algorithm of Ship Arrival Pattern Based on AIS Data
The analysis of the pattern of vessels' passing through a main channel using the traffic volume and time differences between successive arrival vessels can provide theoretical basis and support for maritime authorities in protecting the marine navigation environment, reducing ship accidents, and improving the efficiency of traffic control. Currently, with the popularity of the AIS (Automatic Identification System) equipment installed, AIS information could well reflect the characteristics of vessel traffic in ports. This paper examines vessel arrival pattern with details and accuracy using AIS information. AIS information in port waters collected by AIS receiving equipment and INTERNET in real time is processed and then stored in the database for further study. By the ETL (Extract Transform Load) technique, AIS data is transferred from the original database to the data warehouse, which can support the complex multidimensional query and analysis functions. According to the characteristics of the AIS data and by using OLAP (On-Line Analytical Processing) technology, the observation records of each ship near the threshold in a certain time range are searched and listed in order, so as to make statistics on the AIS data for traffic through the threshold of the main channel. Time differences of each vessel through the threshold can be quickly calculated using the algorithm of time-interval between ships. This paper, taking the vessel arrival regularity in the main channel of Xiamen port as a study case, analyzes the time-interval between ships entering/leaving the port, draws the histogram for the time-interval between the ships, finds out that the density distribution of time-interval between the ships entering/leaving the port is subject to a negative exponential distribution, and validates the algorithm and the model. The result can be used as a basis for the maritime authority in ship traffic control.
- 13 July 2011
- Published by American Society of Civil Engineers (ASCE)
- p. 2677-2682
The Application of Database Techniques in the Integrated Vessel Information Service System
Integrated Vessel Information Service System (IVISS) was a database system which collected and processed AIS (Automatic Identification System) information, and provided information service to clients. AIS information database was the core of the whole system. Large volume of data, the requirement of hard real-time capability, parallel access and wide physical distribution were features of the system, which gave some difficulties to efficient operations of database. In the paper, several database techniques were studied, including mainly data partitions, indexes, transactions and locks, procedures and functions, replications, and implied to IVISS based on features of AIS information database using SQL Server 2005 as a tool. Comprehensive applications of these techniques improved the whole performance of database, obviously increased the connection number and access rate, and also can provided good guides for other development of large database.
- 13 July 2011
- Published by American Society of Civil Engineers (ASCE)
- p. 2677-2682
The Application of Database Techniques in the Integrated Vessel Information Service System
Integrated Vessel Information Service System (IVISS) was a database system which collected and processed AIS (Automatic Identification System) information, and provided information service to clients. AIS information database was the core of the whole system. Large volume of data, the requirement of hard real-time capability, parallel access and wide physical distribution were features of the system, which gave some difficulties to efficient operations of database. In the paper, several database techniques were studied, including mainly data partitions, indexes, transactions and locks, procedures and functions, replications, and implied to IVISS based on features of AIS information database using SQL Server 2005 as a tool. Comprehensive applications of these techniques improved the whole performance of database, obviously increased the connection number and access rate, and also can provided good guides for other development of large database.
- 29 July 2009
- Published by American Society of Civil Engineers (ASCE)
- p. 3718-3723
A New AIS-Based Way to Conduct OLAP of Maritime Traffic Flow
Traditional ways of maritime traffic flow investigation include visual observation, radar scan, and aerial photography, etc. However, they have weaknesses such as time-consuming, laborious, data processing problems, biggish error, being difficult to continuously conduct real-time observation, easily being affected by weather conditions and geographical location, etc. AIS (automatic identification system) is a new kind of maritime communication and navigation system, it includes navigation information (ship identity, location, course, speed and etc.) which can be automatically exchanged between vessels, vessels and coastal stations. It is an important technology means to help avoid ship collision. But in recent years, AIS has been applied to many new areas, including VTS, tracking line drawing, ship grounding alert, maritime accident investigation, etc. This paper proposes a new AIS-based way to conduct investigation and online analytical processing (OLAP) of maritime traffic flow. There are three steps here. Firstly, receiving AIS information in different locations by AIS receivers and INTERNET in real time and putting it into ORACLE database. Secondly, obtaining the basic data of the data warehouse for traffic flow analysis by preprocessing which includes data filtering, data extraction and data reduction. Thirdly, conducting OLAP based on time, ship type, navigation direction, speed level and other dimensions. Content of this article includes research background, technical route, technical realization, OLAP instance, and conclusion.
- 29 July 2009
- Published by American Society of Civil Engineers (ASCE)
- p. 3718-3723
A New AIS-Based Way to Conduct OLAP of Maritime Traffic Flow
Traditional ways of maritime traffic flow investigation include visual observation, radar scan, and aerial photography, etc. However, they have weaknesses such as time-consuming, laborious, data processing problems, biggish error, being difficult to continuously conduct real-time observation, easily being affected by weather conditions and geographical location, etc. AIS (automatic identification system) is a new kind of maritime communication and navigation system, it includes navigation information (ship identity, location, course, speed and etc.) which can be automatically exchanged between vessels, vessels and coastal stations. It is an important technology means to help avoid ship collision. But in recent years, AIS has been applied to many new areas, including VTS, tracking line drawing, ship grounding alert, maritime accident investigation, etc. This paper proposes a new AIS-based way to conduct investigation and online analytical processing (OLAP) of maritime traffic flow. There are three steps here. Firstly, receiving AIS information in different locations by AIS receivers and INTERNET in real time and putting it into ORACLE database. Secondly, obtaining the basic data of the data warehouse for traffic flow analysis by preprocessing which includes data filtering, data extraction and data reduction. Thirdly, conducting OLAP based on time, ship type, navigation direction, speed level and other dimensions. Content of this article includes research background, technical route, technical realization, OLAP instance, and conclusion.
- 9 July 2007
- Published by American Society of Civil Engineers (ASCE)
- p. 1678-1683
Vessel Information Service System Based on ECDIS and AIS
AIS (Automatic Identification System) and ECDIS (Electronic Chart Display and Information System) are two new and very helpful technologies used in marine transportation recently. But information from AIS is simple and text-based, which imposes constraints to a great extent on the data mining of such information. Besides, the valid working range of AIS is limited. The AIS information is not available when the AIS receiver falls out of AIS transmitter's working range. Therefore, we propose in this paper to construct an information service system called VISSEA (Vessel Information Service System based on ECDIS and AIS), displaying AIS information on ECDIS platform, and relaying remote AIS information collected by AIS receivers scattered at different ports to a data server to provide remote AIS information service by the use of internet technology. In this paper, the authors discuss the purpose of constructing such a system; propose the framework of the system and principles of remote data collecting. Some key technical issues of this system are also discussed, such as remote AIS data transfer, decoding of AIS signal, displaying of AIS information on ECDIS and AIS information access based on internet. The system has been put into practice and proved to be valid in application.
- 9 July 2007
- Published by American Society of Civil Engineers (ASCE)
- p. 1678-1683
Vessel Information Service System Based on ECDIS and AIS
AIS (Automatic Identification System) and ECDIS (Electronic Chart Display and Information System) are two new and very helpful technologies used in marine transportation recently. But information from AIS is simple and text-based, which imposes constraints to a great extent on the data mining of such information. Besides, the valid working range of AIS is limited. The AIS information is not available when the AIS receiver falls out of AIS transmitter's working range. Therefore, we propose in this paper to construct an information service system called VISSEA (Vessel Information Service System based on ECDIS and AIS), displaying AIS information on ECDIS platform, and relaying remote AIS information collected by AIS receivers scattered at different ports to a data server to provide remote AIS information service by the use of internet technology. In this paper, the authors discuss the purpose of constructing such a system; propose the framework of the system and principles of remote data collecting. Some key technical issues of this system are also discussed, such as remote AIS data transfer, decoding of AIS signal, displaying of AIS information on ECDIS and AIS information access based on internet. The system has been put into practice and proved to be valid in application.