A Hierarchical Maritime Target Detection Method for Optical Remote Sensing Imagery

Abstract

Maritime target detection from optical remote sensing images plays an important role in related military and civil applications and its weakness lies in its compromised performance under complex uncertain conditions. In this paper, a novel hierarchical ship detection method is proposed to overcome this issue. In the ship detection stage, based on Entropy information, we construct a combined saliency model with self-adaptive weights to prescreen ship candidates from across the entire maritime domain. To characterize ship targets and further reduce the false alarms, we introduce a novel and practical descriptor based on gradient features, and this descriptor is robust against clutter introduced by heavy clouds, islands, ship wakes as well as variation in target size. Furthermore, the proposed method is effective for not only color images but also gray images. The experimental results obtained using real optical remote sensing images have demonstrated that the locations and the number of ships can be determined accurately and that the false alarm rate is greatly decreased. A comprehensive comparison is performed between the proposed method and the state-of-the-art methods, which shows that the proposed method achieves higher accuracy and outperforms all the competing methods. Furthermore, the proposed method is robust under various backgrounds of maritime images and has great potential for providing more accurate target detection in engineering applications.

Vessel detection and classification from spaceborne optical images: A literature survey

“…A relatively large fraction of the authors (35%) used Google Earth images as the direct input data source in their research on vessel detection – either as exported data or simply as a print screen –, or as a source for collecting the greater amount of test data for machine learning methods (An et al, 2013, p. 201; Deng et al, 2013, Dong et al, 2013, Gan et al, 2015, Guo et al, 2015, Han et al, 2014, Hong et al, 2007, Huang et al, 2016, Johansson, 2011, Ju, 2015, Ma et al, 2010, p. 201; Shi et al, 2014, Xu et al, 2017, Xu et al, 2011, p. 201; Xu and Liu, 2016, Xu et al, 2014, Yang et al, 2017, Yang et al, 2014, You and Li, 2011, Zhang et al, 2016, Zou and Shi, 2016). One author has used data from Microsoft Virtual Earth (Yin et al, 2016).…”

“…As seen in Table 2, around half of the vessel detection methods indicated that their input is only the PAN band, neglecting the MS data, and thus relying more on spatially oriented and less on spectral techniques (35 authors or 52%: Fernandez Arguedas, 2015, Bi et al, 2012, Buck et al, 2007, Corbane et al, 2010, Corbane et al, 2008, Corbane et al, 2008, Guang et al, 2011, Guo and Zhu, 2012, Harvey et al, 2010, Huang et al, 2011, Hung, 2016, Jin and Zhang, 2015, Jubelin and Khenchaf, 2014, Li et al, 2016, Li et al, 2016, Liu et al, 2014, Pan et al, 2015, Proia and Page, 2010, Qi et al, 2015, Shi et al, 2014, Tang et al, 2015, Wang et al, 2005, Willhauck et al, 2005, Xu et al, 2017, Xu and Liu, 2016, Yang et al, 2014, Yang et al, 2015, Yang et al, 2017, Yao et al, 2016, Yokoya and Iwasaki, 2015, Zou and Shi, 2016). On the other hand, 30% (20 authors) use only MS data, and 18% (12 authors) use both PAN and MS data (Fig.…”

“…Zhang et al2016Google Earthn/a0.12, 0.25−n/aDeep learning method: ship proposal extraction convolution neural networks−yesVessel detectionR. Zhang et al2016GaoFen-1, VRRS-1, Google EarthPAN2 (GF-1), 16 (VRSS-1)−20 pixelsDeep learning method: convolutional neural network, Singular value decomposition algorithmDiscrimination: SVM classifieryesVessel detectionZou and Shi2017Google Earthn/a−−n/aComputer vision method: rotation and scale-invariant method based on the pose consistency voting−yesInshore vessel detectionHe et al2017Google EarthPAN2−n/aSalient-based method: maximum symmetric surround method, cellular automata dynamic evolution model, Otsu algorithmDiscrimination: histogram of oriented gradient, AdaBoost classifieryesVessel detectionWang et al2017Google EarthB, G, R−−10 pixelsSalient-based method: combined saliency map model through a self-adaptive threshold based on Entropy informationDiscrimination: based on gradient featuresyesVessel detectionXu et al2017Google EarthPAN2−n/aSalient-based method: Histogram-based contrast method, phase spectrum of a Fourier transform, surface regular indexDiscrimination: Simple shape analysis, structure-local binary pattern, AdaBoost algorithmyesVessel detectionYang et al…”

Ship Detection in Optical Remote Sensing Images Based on Wavelet Transform and Multi-Level False Alarm Identification

“…The framework is similar to the two stages in [36]. However, the extraction and discrimination features are different from there.…”

Region-of-Interest Extraction Based on Local–Global Contrast Analysis and Intra-Spectrum Information Distribution Estimation for Remote Sensing Images

“…They are very time-consuming because global searching is an indispensable part in the processing. Moreover, with the rapid development of remote sensing technology, the resolution of remote sensing images increases, and the intensity, structure, shape and texture information are more abundant [5][6][7][8]. Due to the irregular shape, unfixed size and other characteristics of ROIs, the extraction accuracy cannot be guaranteed when Aggregating neighboring pixels into superpixels can not only reduce the complexity of subsequent processing but also maintain the boundary information.…”