Application of 3D reconstruction system based on close-range photogrammetry method for plant growth estimation

Abstract

Abstract Plant growth monitoring is an important aspect of precision agriculture implementation. The monitoring can be performed by estimating the volume by the result of Three-dimensional (3D) reconstruction by using Close-range Photogrammetry. However, to present the functionality of the system for plant growth behavior, it is necessary to evaluate its accuracy and performance. In this study, a plant growth estimation system based on non-contact measurement using the Close-Range Photogrammetry (CRP) method for volume estimation of Chinese cabbage was developed to measure the growth. The objective of this study was to apply a 3D reconstruction system using the CRP method for validating volume variation and estimate the rate of growth of Chinese cabbage. This system consists of Canon 700D’s DSLR camera and camera stabilizer. The stage of image processing using 3DF’s Zephyr Pro photogrammetric software for generating 3D models. For the validation purposes and its functionality for modelling and estimating volumetric objects, Chinese cabbage with four size variations at different ages was used (14, 21, 28, and 35 Days After Transplant). As the result, the developed system could observe and generate the plant in a three-dimensional manner resemble the actual plant model. Farther the volumetric validation could be obtained with the result of R2 of 0,9991, Root Mean Square Error (RMSE) = 26,08 cm3 and Mean Absolute Percentage Error (MAPE) = 10,43%. From the result, the system could be used for generating plants into 3D objects and its accuracy of measurement is quite good. Further improvements in accuracy need to be made for precise measurements as well as validation for other crop types.