An application of multispectral semantic-image space for global farming analysis and crop condition comparisons

The global environmental analysis system is a new platform to analyze environmental multimedia data that acquired from nature resources. This system aims to realize and interpret environmental phenomena and changes occurring that happening in world wide scope. Semantic computing is important and promising approach to multispectral semantic-image analysis for various environmental aspects and contexts in physical world. In the previous study, we proposed a new system of agricultural monitoring and analysis based on semantic computing concept that it realizes the interpretation of agricultural health condition as human-level interpretation. In this paper, we propose a new analytical method for agriculture global comparisons to realize and recognize crop condition with several places in global scale. Multispectral semantic-image space for agricultural analysis can be utilized for global crop health monitoring by comparing crop conditions among different places. Our method applies semantic distance calculation to measure similarity among multispectral image data to realize the crop health condition as a ranking. According to our new proposed analytical method, we demonstrate a prototype implementation in the case of rye farm in Latvia and Finland. This prototype implementation shows an analysis in the case that image data have same crop type and conditions.