3D Imaging of Greenhouse Plants with an Inexpensive Binocular Stereo Vision System

Abstract

Nowadays, 3D imaging of plants not only contributes to monitoring and managing plant growth, but is also becoming an essential part of high-throughput plant phenotyping. In this paper, an inexpensive (less than 70 USD) and portable platform with binocular stereo vision is established, which can be controlled by a laptop. In the stereo matching step, an efficient cost calculating measure-AD-Census-is integrated with the adaptive support-weight (ASW) approach to improve the ASW's performance on real plant images. In the quantitative assessment, our stereo algorithm reaches an average error rate of 6.63% on the Middlebury datasets, which is lower than the error rates of the original ASW approach and several other popular algorithms. The imaging experiments using the proposed stereo system are carried out in three different environments including an indoor lab, an open field with grass, and a multi-span glass greenhouse. Six types of greenhouse plants are used in experiments; half of them are ornamentals and the others are greenhouse crops. The imaging accuracy of the proposed method at different baseline settings is investigated, and the results show that the optimal length of the baseline (distance between the two cameras of the stereo system) is around 80 mm for reaching a good trade-off between the depth accuracy and the mismatch rate for a plant that is placed within 1 m of the cameras. Error analysis from both theoretical and experimental sides show that for an object that is approximately 800 mm away from the stereo platform, the measured depth error of a single point is no higher than 5 mm, which is tolerable considering the dimensions of greenhouse plants. By applying disparity refinement, the proposed methodology generates dense and accurate point clouds of crops in different environments including an indoor lab, an outdoor field, and a greenhouse. Our approach also shows invariance against changing illumination in a real greenhouse, as well as the capability of recovering 3D surfaces of highlighted leaf regions. The method not only works on a binocular stereo system, but is also potentially applicable to a SFM-MVS (structure-from-motion and multiple-view stereo) system or any multi-view imaging system that uses stereo matching.

Genetics of barley tiller and leaf development

“…Laser sensors: This system is based on light detection and ranging (LiDAR), where laser beams are projected onto plants. The projected laser beams (scattered energy from the plant or the surface) can then be measured using triangulation and a dense 3D map of point clouds is constructed (Kjaer and Ottosen ; Li et al ). This laser sensor approach can measure the distance between sensor and the object, based on the elapsed time between the emission and return of laser pulse from the sensor (the ToF method), or based on trigonometry (Omasa et al ).…”

Imaging Wheat Canopy Through Stereo Vision: Overcoming the Challenges of the Laboratory to Field Transition for Morphological Features Extraction

“…For this study, a baseline of 50 mm was used. For the same distance between the cameras and the plants (1 m), Li et al (2017) have shown that a baseline of 80 mm provides the best compromise between depth accuracy and mismatch rate. However, their test took place in indoor conditions.…”

“…It means that disparities differing from one pixel contributed to the error, with the same weight as a more important error. In comparison with the literature (Scharstein and Szeliski, 2002) and more particularly with an optimized stereo matching algorithm leading to errors of 2.9 % and 7 % for cones and teddy images respectively (Li et al, 2017), the performances of the SGBM algorithm on the Middlebury dataset were considered as sufficient.…”

“…Jimenez-berni et al (2018) measure height with a RMSE of 0.017 m. Leaf angle measurements could be obtained from LiDAR point clouds coupled with RGB image just as for stereo vision point clouds. Li et al (2017) measure green area index with a RMSE of 0.22. However, even if LiDAR is more and more affordable, it remains costly and has to be coupled with a RGB camera to provide both morphological and color information.…”

Novel Low Cost 3D Surface Model Reconstruction System for Plant Phenotyping

“…While the stereo-vision system is insensitive to movement and has color information, it also has poor depth resolution and is susceptible to sunlight. A low-cost and portable stereo-vision system was established to realize 3D imaging of plants under greenhouse cultivation [8]. A growth chamber system for 3D imaging by using a 3D-calibrated camera array was successfully used to quantify leaf morphological traits, 2D, and 3D leaf areas, and diurnal cycles of leaf movement; however, the limitation of this system is that it can only acquire half-shelf models, which lack detailed information about overlapping leaves [9].…”