Rice Productivity Prediction Model Design Based On Linear Regression of Spectral Value Using NDVI and LSWI Combination On Landsat-8 Imagery

3rd International Conference of Indonesia Society for Remote Sensing

地球科学;计算机科学

Prasetyo, Yudo^1 ; Sukmono, Abdi^1 ; Aziz, Kurnia Wisnu^1 ; Prakosta Santu Aji, Bernardinus Joko^1

Department of Geodetic Engineering, Faculty of Engineering, Diponegoro University Semarang, Indonesia^1

关键词: Agricultural remote sensing;    Agriculture crops;    Coefficient of determination;    Land surface water index;    Population growth;    Production models;    Satellite remote sensing;    Strategic contingencies;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/165/1/012002/pdf
DOI  :  10.1088/1755-1315/165/1/012002

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Rice is one of the most important agriculture crops in many countries, and it is a primary food source for more than three billion people worldwide. The global rice consumption is projected to be 873 million tonnes in 2030. Based on national socio-economic survey (susenas) data from Indonesia Statistic Agency (BPS), Indonesia's population of rice consumption 2016 reached 78,816 kg per capita per year. In the recent decades, two major issues like population growth (in particular in the major rice producing or consuming countries) and climate change put enormous pressure on the global food demand and its production. Almost all countries made the issue of food as a major issue. Now, Indonesian government wants to realize Indonesia became self-sufficient of rice in 2017. Some food security programs implemented to achieve it. One of them is predicting or estimation rice production and consumption. Prediction or estimation rice productivity before harvest is crucial, especially in regions characterised by climatic uncertainties. It is also enables governments to put in a good strategic contingency plans for the redistribution of food during times of famine. Satellite remote sensing has been widely applied and recognised as a powerful and effective tool for identifying agriculture crops. An important goal of agricultural remote sensing research is to spectrally estimate crop variables related to crop conditions, which can subsequently be entered into crop simulation and production models. In this research will be use a global and medium resolution of satellite remote sensing imagery (Landsat 8 ETM+) with 2016-2017 acquisition year in order to estimation of rice productivity. The rice productivity prediction modeling methods will be build in this research based on regression and spectral pattern analysis of NDVI (and LSWI (Land Surface Water Index in multi temporal satellite imagery. In addition, this prediction model will be tested against fact the level of productivity in the rice field. We will take a proportional field sampling test using cluster sampling test and random spectral value checking in order to measure accuracy level of rice productivity prediction model. The results of this research indicate that NDVI and LSWI algorithm is the best combination of linear regression in estimation of rice productivity level with coefficient of determination equal to 0.639. Validation of regression model equation to data of Agriculture Agency of Demak has difference or RMSE 8,394 Quintal/Ha