Search results

Filters

  • Journals
  • Authors
  • Keywords
  • Date
  • Type

Search results

Number of results: 2
items per page: 25 50 75
Sort by:
Download PDF Download RIS Download Bibtex

Abstract

This paper discusses the identification of garbage using the YOLO algorithm. In the rivers, it is usually difficult to distinguish between garbage and plants, especially when it is done in real-time and at the time of too much light. Therefore, there is a need of an appropriate method. The HSV and SIFT methods were used as preliminary tests. The tests were quite successful even in close condition, however, there were still many problems faced in using this method since it is only based on pixel and shape readings. Meanwhile, YOLO algorithm was able to identify garbage and water hyacinth even though they were closed to each other.
Go to article

Bibliography

[1] J. Gasperi, R. Dris, T. Bonin, V. Rocher, and B. Tassin. “Assessment of floating plastic debris in surface water along the Seine River,” Environ. Pollut, 2014;195, p. 163–166.
[2] C. Su, W. Dongxing, L. Tiansong, R. Weichong, and Z. Yachao. “An autonomous ship for cleaning the garbage floating on a lake,” In: 2009 Second International Conference on Intelligent Computation Technology and Automation, 2009, vol. 3, p. 471–474.
[3] A. S. A. Kader, M. K. M. Saleh, M. R. Jalal, O. O. Sulaiman, and W. N. W. Shamsuri. “Design of Rubbish Collecting System for Inland Waterways,” J. Transp. Syst. Eng, 2015; 2, no. 2, p. 1–13.
[4] Y. Liu, K.-C. Fung, W. Ding, H. Guo, T. Qu, and C. Xiao. “Novel Smart Waste Sorting System based on Image Processing Algorithms: SURF-BoW and Multi-class SVM. Comput,” Inf. Sci. 2018; 11, no. 3, p. 35.
[5] B. M. Chinnathurai, R. Sivakumar, S. Sadagopan, and J. M. Conrad. “Design and implementation of a semi-autonomous waste segregation robot,” in SoutheastCon 2016. p. 1–6.
[6] A. Torres-García, O. Rodea-Aragón, O. Longoria-Gandara, F. Sánchez-García, and L. E. González-Jiménez. “Intelligent waste separator,” Comput. y Sist. 2015; 19, no. 3. p. 487–500.
[7] T. P. Deepa and S. Roka, “Estimation of garbage coverage area in water terrain,” in 2017, International Conference On Smart Technologies For Smart Nation (SmartTechCon), 2017. p. 347–352.
[8] A. Krizhevsky, I. Sutskever, and G. E. Hinton. “Imagenet classification with deep convolutional neural networks,” in Advances in neural information processing systems. 2012, p. 1097–1105.
[9] R. Girshick, J. Donahue, T. Darrell, and J. Malik, „Rich feature hierarchies for accurate object detection and semantic segmentation,” in Proceedings of the IEEE conference on computer vision and pattern recognition. 2014, p. 580–587.
[10] R. Girshick. “Fast r-cnn,” in Proceedings of the IEEE international conference on computer vision. 2015, p. 1440–1448.
[11] S. Ren, K. He, R. Girshick, and J. Sun. “Faster r-cnn: Towards real-time object detection with region proposal networks,” in Advances in neural information processing systems. 2015, p. 91–99.
[12] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi. “You only look once: Unified, real-time object detection,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2016, p. 779–788.
[13] G. Mittal, K. B. Yagnik, M. Garg, and N. C. Krishnan. “Spotgarbage: smartphone app to detect garbage using deep learning,” in Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing. 2016, p. 940–945.
[14] M. S. Rad et al. “A computer vision system to localize and classify wastes on the streets,” in International Conference on Computer Vision Systems. 2017, p. 195–204.
[15] J. Redmon and A. Farhadi.”YOLO9000: better, faster, stronger,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017, p. 7263–7271.
[16] D. Mu, Y. Zhao, G. Wang, Y. Fan, and Y. Bai. “USV model identification and course control,” in 2016 Sixth International Conference on Information Science and Technology (ICIST). 2016, p. 263–267.
[17] Q. Zhu. “Design of control system of USV based on double propellers,” in IEEE Reg. 10 Annu. Int. Conf. Proceedings/TENCON. 2013.
[18] R. Yan, S. Pang, H. Sun, and Y. Pang. “Development and missions of unmanned surface vehicle,” J. Mar. Sci. Appl. 2010; 9, no. 4, p. 451–457.
[19] P. H. Heins, B. L. Jones, and D. J. Taunton. “Design and validation of an unmanned surface vehicle simulation model,” Appl. Math. Model. 2017; 48, p. 749–774.
Go to article

Authors and Affiliations

Bhakti Yudho Suprapto
1
Kelvin
1
Muhammad Arief Kurniawan
1
Muhammad Kevin Ardela
1
Hera Hikmarika
1
Zainal Husin
1
Suci Dwijayanti
1

  1. Department of Electrical Engineering, Faculty of Engineering, Universitas Sriwijaya, Indonesia
Download PDF Download RIS Download Bibtex

Abstract

Geospatial data obtained using Unmanned Aerial Vehicles (UAVs) and Unmanned Surface Vehicles (USVs) are increasingly used to model the terrain in the coastal zone, in particular in shallow waterbodies (with a depth of up to 1 m). In order to generate a terrain relief, it is important to choose a method for modelling that will allow it to be accurately projected. Therefore, the aim of this article is to present a method for accuracy assessment of topo-bathymetric surface models based on geospatial data recorded by UAV and USV vehicles. Bathymetric and photogrammetric measurements were carried out on the waterbody adjacent to the public beach in Gdynia (Poland) in 2022 using a DJI Phantom 4 RTK UAV and an AutoDron USV. The geospatial data integration process was performed in the Surfer software. As a result, Digital Terrain Models (DTMs) in the coastal zone were developed using the following terrain modelling methods: Inverse Distance to a Power (IDP), Inverse Distance Weighted (IDW), kriging, the Modified Shepard’s Method (MSM) and Natural Neighbour Interpolation (NNI). The conducted study does not clearly indicate any of the methods, as the selection of the method is also affected by the visualization of the generated model. However, having compared the accuracy measures of the charts and models obtained, it was concluded that for this type of data, the kriging (linear model) method was the best. Very good results were also obtained for the NNI method. The lowest value of the Root Mean Square Error (RMSE) (0.030 m) and the lowest value of the Mean Absolute Error (MAE) (0.011 m) were noted for the GRID model interpolated with the kriging (linear model) method. Moreover, the NNI and kriging (linear model) methods obtained the highest coefficient of determination value (0.999). The NNI method has the lowest value of the R68 measure (0.009 m), while the lowest value of the R95 measure (0.033 m) was noted for the kriging (linear model) method.
Go to article

Authors and Affiliations

Oktawia Lewicka
1 2

  1. Department of Geodesy and Oceanography, Gdynia Maritime University, ul. Morska 81-87, 81-225 Gdynia, Poland
  2. Marine Technology Ltd., ul. Wiktora Roszczynialskiego 4-6, 81-521 Gdynia, Poland

This page uses 'cookies'. Learn more