ResearchBib Share Your Research, Maximize Your Social Impacts
Sign for Notice Everyday Sign up >> Login

Soil Properties Classification Using Support Vector Machine for Raver Tehsil

Journal: International Journal of Advanced Trends in Computer Science and Engineering (IJATCSE) (Vol.10, No. 6)

Publication Date:

Authors : ;

Page : 3154-3159

Keywords : ;

Source : Downloadexternal Find it from : Google Scholarexternal


Soil properties are dynamic in nature and different factors are affecting to the soil quality. It is directly consequence on soil productivity and soil fertility. The heavy use of fertilizers, heavy rain fall, various agricultural practices are responsible for soil quality degradation. The soil assessment is require to maintain the soil quality. The spectroscopic techniques using Remote sensing and GIS gives the fast and accurate results as compare to traditional soil testing methods. The present study is conducted for classification of soil physicochemical properties in pre monsoon and post monsoon season. Soil samples are collected where Organic, Chemical and Mixed fertilizers treatments were applied to banana and cotton crops sites from Raver tehsil of Jalgaon district. Total 220 soil specimens are collected in pre monsoon and post monsoon season for two year respectively. ASD FieldSpec4 spectroradiometer device were used for data acquisition in the controlled laboratory environment. Acquired spectral data were processed for conversion in numeric format then various statistical methods were used for quantitative analysis of the physiochemical soil properties. The support vector machine is used for classification of the collected soil samples in pre-monsoon and post-monsoon season and classification were performed on the basis of training and testing datasets. The soil samples are divide in pre-monsoon training, pre-monsoon testing and post –monsoon training and post-monsoon testing class with support vector. The hyper plane is used for separation of pre-monsoon and post-monsoon soil samples. Misclassification rate and Mean Squared Error were calculated in the SVM classification.

Last modified: 2021-12-16 23:54:24