An initial study of deep learning for mangrove classification

2nd Nommensen International Conference on Technology and Engineering

Faza, S.^1 ; Nababan, E.B.^2 ; Efendi, S.^2 ; Basyuni, M.^3 ; Rahmat, R.F.^2

Stud. in Faculty of Computer Science and Information Technology, Universitas Sumatera Utara, Medan, Indonesia^1

Faculty of Computer Science and Information Technology, Universitas Sumatera Utara, Medan, Indonesia^2

Faculty of Forestry, Universitas Sumatera Utara, Medan, Indonesia^3

关键词: Hyperspectral satellite;    Morphological data;    Plant classification;    Real values;    Testing accuracy;    Training errors;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/420/1/012093/pdf
DOI  :  10.1088/1757-899X/420/1/012093

来源: IOP

【 摘 要 】

Deep Learning is a new breakthrough in the area of neural network. One of it methodology is Deep Neural Network. Usually, deep neural network is a method that can solve problems such as classification or prediction. Here in this study, we collect data for mangrove classification, and we see an opportunity to classify the type of mangrove based on its features using deep neural network. Research on mangrove plants related to classification is more widely used in mangrove dispersal spread through spectral and hyperspectral satellite images than using real value data such as morphological data, thus providing blank space for researchers to use mangrove morphological data. This initial study is to build a deep neural network architecture and analyze it in term of mangrove sprout plant classification. The result from our architectural methodology of this research is reaching the lowest training error of 0.1345 and the highest testing accuracy value of 98%.