Automatic Classification Using Machine Learning for Non-Conventional Vessels on Inland Waters

The prevalent methods for monitoring ships are based on automatic identification and radar systems. This applies mainly to large vessels. Additional sensors that are used include video cameras with different resolutions. Such systems feature cameras that capture images and software that analyze the selected video frames. The analysis involves the detection of a ship and the extraction of features to identify it. This article proposes a technique to detect and categorize ships through image processing methods that use convolutional neural networks. Tests to verify the proposed method were carried out on a database containing 200 images of four classes of ships. The advantages and disadvantages of implementing the proposed method are also discussed in light of the results. The system is designed to use multiple existing video streams to identify passing ships on inland waters, especially non-conventional vessels.

[1]  Palanisamy Shanmugam,et al.  Ship Recognition by Integration of SAR and AIS , 2012 .

[2]  Andrzej Stateczny,et al.  Radar and Automatic Identification System Track Fusion in an Electronic Chart Display and Information System , 2015 .

[3]  Jason Cong,et al.  Caffeine: Towards uniformed representation and acceleration for deep convolutional neural networks , 2016, 2016 IEEE/ACM International Conference on Computer-Aided Design (ICCAD).

[4]  Mario Miličević,et al.  Data Augmentation and Transfer Learning for Limited Dataset Ship Classification , 2018 .

[5]  Peter Kovesi,et al.  Fast Almost-Gaussian Filtering , 2010, 2010 International Conference on Digital Image Computing: Techniques and Applications.

[6]  Kuldip K. Paliwal,et al.  Accurate prediction of protein contact maps by coupling residual two-dimensional bidirectional long short-term memory with convolutional neural networks , 2018, Bioinform..

[7]  Christopher Hunt,et al.  Notes on the OpenSURF Library , 2009 .

[8]  Weihong Deng,et al.  Very deep convolutional neural network based image classification using small training sample size , 2015, 2015 3rd IAPR Asian Conference on Pattern Recognition (ACPR).

[9]  Anne-Claire Boury-Brisset,et al.  Ship Classification Using Deep Learning Techniques for Maritime Target Tracking , 2018, 2018 21st International Conference on Information Fusion (FUSION).

[10]  Jason Cong,et al.  Caffeine: Toward Uniformed Representation and Acceleration for Deep Convolutional Neural Networks , 2019, IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems.

[11]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[12]  Luca Iocchi,et al.  Camera based target recognition for maritime awareness , 2012, 2012 15th International Conference on Information Fusion.

[13]  Ting Liu,et al.  Recent advances in convolutional neural networks , 2015, Pattern Recognit..

[14]  Erhan Gundogdu,et al.  Generic and attribute-specific deep representations for maritime vessels , 2017, IPSJ Transactions on Computer Vision and Applications.

[15]  Natalia Wawrzyniak,et al.  Automatic Watercraft Recognition and Identification on Water Areas Covered by Video Monitoring as Extension for Sea and River Traffic Supervision Systems , 2018 .

[16]  U. Rajendra Acharya,et al.  Deep convolutional neural network for the automated detection and diagnosis of seizure using EEG signals , 2017, Comput. Biol. Medicine.

[17]  Andrzej Stateczny.  Sensors in River Information Services of the Odra River in Poland: Current State and Planned Extension , 2017, 2017 Baltic Geodetic Congress (BGC Geomatics).