Rough-Sea-Horizon-Line Detection using a Novel Color Clustering and Least Squares Regression Method
Sea horizon line (SHL) detection is the first step in maritime image processing. It is aimed at object detection, navigation of autonomous aerial and sea surface vehicles, and distance estimation. Many methods have been proposed to detect SHL, however, their focus remains on SHL detection in calm sea conditions. For this reason, this study is designed to fill this gap and investigate an efficient method to detect a sea horizon line under rough sea conditions. A novel color clustering and least-squares regression-based method is proposed to solve the issue. Minimizing computational cost, our method identifies the candidate region of interest (CROI) from grayscale image ROIs by analyzing the modality of its pixel intensity histogram. It then applies k-mean clustering to highlight potential sea-sky regions. The regional boundary pixel coordinates are used to construct a horizon line by applying the least-squares regression method. The results of the proposed method were compared to the Canny Edge and Hough Transform (CEHT) method. The gale state sea images were used to test the efficiency of both methods, correctly detecting the starting and ending coordinates of a horizon line and its slope. The results highlight the superiority of the proposed method over the CEHT method. On average, the proposed method identified the horizon line within one degree of error as opposed to CEHT method with the average of five degrees. In half of the images, the error in detecting the horizon starting and ending coordinates for the proposed method was within five or fewer pixels. The overall results show the superiority of the proposed method over the CEHT method in rough sea conditions. The novelty of this study is two-fold. Firstly, it is a pioneering study that proposes a novel method to detect SHL under rough sea conditions. Secondly, the proposed method yielded superior SHL detection results compared to its peers.
Keywords: Sea Horizon Line Detection, Rough Sea Conditions, Candidate Region of Interest, Color Clustering, Least Squares Regression.
SHAN X., ZHAO D., PAN M., WANG D. and ZHAO L. Sea-sky line and its nearby ships detection based on the motion attitude of visible light sensors. Sensors (Switzerland), 2019, 19(18): 1–23.
TODOROVIC S. and NECHYBA M.C. A vision system for intelligent mission profiles of micro air vehicles. IEEE Transactions on Vehicular Technology., 2004, 53(6): 1713–1725. https://doi.org/10.1109/TVT.2004.834880
MCGEE T. G., SENGUPTA R., and HEDRICK K. Obstacle Detection for Small Autonomous Aircraft Using Sky Segmentation. Proceedings of the 2005 IEEE International Conference on Robotics and Automation, Barcelona, Spain, 2005: 4679-4684, https://doi.org/10.1109/ROBOT.2005.1570842.
DEL PIZZO S., GAGLIONE S., ANGRISANO A., SALVI G., and TROISI S. Reliable vessel attitude estimation by wide angle camera. Measurement: Journal of the International Measurement Confederation, 2018, 127, (January): 314–324, https://doi.org/10.1016/j.measurement.2018.05.104
GLADSTONE R., MOSHE Y., BAREL A., and SHENHAV E. Distance estimation for marine vehicles using a monocular video camera. 24th European Signal Processing Conference (EUSIPCO), Budapest, 2016: 2405-2409, https://doi.org/10.1109/EUSIPCO.2016.7760680.
ETTINGER S.M., NECHYBA M.C., IFJU P.G., and WASZAK M. Vision-Guided Flight Stability and Control for Micro Air Vehicles. IEEE/RSJ International Conference on Intelligent Robots and Systems, Lausanne, Switzerland, 2002, 3: 2134-2140, https://doi.org/10.1109/IRDS.2002.1041582.
FEFILATYEV S., GOLDGOF D., SHREVE M., and LEMBKE C. Detection and tracking of ships in open sea with rapidly moving buoy-mounted camera system. Ocean Engineering, 2012, 54: 1–12, https://doi.org/10.1016/j.oceaneng.2012.06.028
HASHMANI M. A., UMAIR M., HUSSAIN RIZVI S.S., and REHMAN GILAL A. A Survey on Edge Detection based recent Marine Horizon Line Detection Methods and their Applications. 3rd International Conference on Computing, Mathematics and Engineering Technologies (iCoMET), Sukkur, Pakistan, 2020: 1-5, https://doi.org/10.1109/iCoMET48670.2020.9073895.–5.
JEONG C.Y., YANG H.S., and MOON K.D. Fast horizon detection in maritime images using region-of-interest. International Journal of Distributed Sensor Networks, 2018, 14(7), https://doi.org/10.1177/1550147718790753
LAING A. K. et al., Guide to Wave Analysis and Forecasting. 1998.
STOFFREGEN T. Force 10 on the Beaufort Scale - Balance research ship. 2016. https://www.youtube.com/watch?v=1BXCMoQhnBU.
UNITED STATES NATIONAL WEATHER SERVICE. Sea State Photographs for Determining Wind Speed. in U.S. Government Posters, Book 59, 1987.
GERSHIKOV E. Horizon Line Detection in Marine Images: Which Method to Choose? International Journal on Advances in Intelligent Systems, 2013, 6(1–2): 79–88. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.682.4145&rep=rep1&type=pdf
PRASAD D.K., RAJAN D., RACHMAWATI L., RAJABALLY E., and QUEK C. MuSCoWERT: multi-scale consistence of weighted edge Radon transform for horizon detection in maritime images. Journal of the Optical Society of America A, 2016, 33(12): 2491-2500, https://doi.org/10.1364/JOSAA.33.002491
MARAGOS P., Morphological Filtering, 1st ed. Elsevier, 2009.
- There are currently no refbacks.