Development of Remote Sensing Software Based on Hyperspectral Imaging Framework

Abstract

The purpose of this study is to develop application software for various hyperspectral imaging analyses in agricultural sensing and ecosystem observation. As previously reported, we constructed a software framework for hyperspectral imaging. A number of common hyperspectral image processing algorithms are available in this framework. Custom analytical software for specialized sensing tasks can be efficiently developed using the framework. Spectral processing algorithms can be switched through polymorphism, a standard object-oriented programming mechanism. Analytical software developers need only write the specialized portion of the program code that implements their spectral processing algorithm; the framework handles the common tasks. In this study, six spectral processors (for waveband extraction, false color generation, data normalization, segmentation between plant and soil, SPAD estimation, and for plant classification) were experimentally developed as examples of spectral processor development. In addition, data sampling software that can be used to obtain waveband images and pixel spectral data was developed.