Estimating 3D Chlorophyll Content Distribution of Trees Using an Image Fusion Method Between 2D Camera and 3D Portable Scanning Lidar

Fumiki Hosoi,Kuangting Kuo,Sho Umeyama

doi:10.3390/rs11182134

Abstract

An image fusion method has been proposed for plant images taken using a two-dimensional (2D) camera and three-dimensional (3D) portable lidar for obtaining a 3D distribution of physiological and biochemical plant properties. In this method, a 2D multispectral camera with five bands (475–840 nm) and a 3D high-resolution portable scanning lidar were applied to three sets of sample trees. After producing vegetation index (VI) images from multispectral images, 3D point cloud lidar data were projected onto the 2D plane based on perspective projection, keeping the depth information of each of the lidar points. The VI images were 2D registered to the lidar projected image based on the projective transformation and VI 3D point cloud images were reconstructed based on the depth information. Based on the relationship between the VI values and chlorophyll contents taken by a soil and plant analysis development (SPAD)-502 plus chlorophyll meter, 3D distribution images of the chlorophyll contents were produced. Similarly, a thermal 3D image for a sample was also produced. The resultant chlorophyll distribution images offered vertical and horizontal distributions, and those for each orientation for each sample, showing the spatial variability of the distribution and the difference between the samples.

Highlights

Accurate measurements of plant physiological and biochemical properties are necessary to understand plant functions
A portable scanning lidar that can provide a colored point cloud was used for estimating the chlorophyll contents within trees [16]
In the methods adopted by [24,25], the texture mapping technique [28,29] was applied for fusion, in which a lidar-derived 3D point cloud image was converted to a polygonal surface mesh image

Summary

Introduction

Accurate measurements of plant physiological and biochemical properties are necessary to understand plant functions. Structural information has been extracted using lidars, their ability to provide a detailed 3D reconstruction of plants can be utilized to obtain 3D distributions of physiological and biochemical plant properties. Such applications have been reported in several studies. Chlorophyll (abbreviated as Chl) and nitrogen contents were estimated by a portable scanning lidar with a green laser source [17,18] These methods can provide chlorophyll contents, which relate to biochemical processes, associated with the structural properties. To effectively obtain plant physiological and biochemical information together with the structural features, the fusion method should be applicable to different types of 2D cameras and lidars

Methods

Results

Conclusion