Development of a Graphical User Interface for a handy type Hyperspectral Imaging System to Investigate the Alternate Bearing Phenomenon of Tree crops

Abstract

This paper describes the development of a graphical user interface using Matlab for the processing and analyses of hyperspectral images. The hyperspectral sensor used was a ground based handy type line sensor ImSpector (V10_12_102) mounted on a battery powered movable base of a tripod. This sensor had a nominal spectral resolution of 1.5 to 2 nm and a wavelength range of 360 – 1010 nm. The motivation in the development of the interface is to conduct a study regarding the alternate bearing phenomenon of tree crops such as Citrus Unshu and Acorn trees. This is done by investigating the relationship between the hyperspectral image of the tree crops and their yield. Since the spectral images acquired has a large amount of data and is multi-dimensional, the interface would aid in the processing and analysis. The interface can generate images from the raw spectral data of audio-video interleave format. It can handle other types of image formats such as TIFF or BMP files. It can also display an image at a chosen wavelength, generate false RGB images, display the spectral characteristics by point and click on the image, and display normalized vegetation index such as normalized difference vegetation index (NDVI) and other vegetation indices. Useful image analysis algorithms are also included, such as image thresholding and pixel count of a binarize image. Image analysis capability can be expanded to include other types of analysis techniques. The interface was developed in a MatlabR12 environment, and existing libraries in the Matlab software can be easily incorporated to provide more capability of the graphical user interface.