Wheat Ear Detection in Plots by Segmenting Mobile Laser Scanner Data

Abstract

ABSTRACT:The use of Light Detection and Ranging (LiDAR) to study agricultural crop traits is becoming popular. Wheat plant traits such as crop height, biomass fractions and plant population are of interest to agronomists and biologists for the assessment of a genotype's performance in the environment. Among these performance indicators, plant population in the field is still widely estimated through manual counting which is a tedious and labour intensive task. The goal of this study is to explore the suitability of LiDAR observations to automate the counting process by the individual detection of wheat ears in the agricultural field. However, this is a challenging task owing to the random cropping pattern and noisy returns present in the point cloud. The goal is achieved by first segmenting the 3D point cloud followed by the classification of segments into ears and non-ears. In this study, two segmentation techniques: a) voxelbased segmentation and b) mean shift segmentation were adapted to suit the segmentation of plant point clouds. An ear classification strategy was developed to distinguish the ear segments from leaves and stems. Finally, the ears extracted by the automatic methods were compared with reference ear segments prepared by manual segmentation. Both the methods had an average detection rate of 85%, aggregated over different flowering stages. The voxel-based approach performed well for late flowering stages (wheat crops aged 210 days or more) with a mean percentage accuracy of 94% and takes less than 20 seconds to process 50,000 points with an average point density of 16 points/cm 2 . Meanwhile, the mean shift approach showed comparatively better counting accuracy of 95% for early flowering stage (crops aged below 225 days) and takes approximately 4 minutes to process 50,000 points.

Yielding to the image: How phenotyping reproductive growth can assist crop improvement and production

“…In the field, the number of inflorescences (spikes, panicles) could be estimated from RGB images at flowering [32,34], as mentioned in section 2.2. The use of Light Detection and Ranging (LiDAR) is currently under development for the detection of wheat ears in the field, and recent studies report an average detection rate of 85%, aggregated over different flowering stages [98]. In most crops, the rest of the numerical components are best estimated or derived from measurements after harvest, some of them object of trivial image analysis, such as grain counting or grain dimensions, which can be carried out with flatbed scanners and free firmware.…”

Double-Threshold Segmentation of Panicle and Clustering Adaptive Density Estimation for Mature Rice Plants Based on 3D Point Cloud

“…Hosoi and Omasa [20] adopted the LiDAR to collect cloud data of rice canopy at different growth stages and established the relationship between rice canopy volume and density according to voxel estimation. Velumani et al [21] installed the signal LMS400 LiDAR to scan wheat field, and the voxel segmentation and mean-shift algorithm were used to segment the wheat head point cloud. Then the number of higher point-cloud clusters was calculated as the number of wheat panicles.…”

High-throughput phenotyping analysis of maize at the seedling stage using end-to-end segmentation network

“…Unmanned aerial vehicles (UAVs), manned ground vehicles (MGVs), and tractor-based high-throughput phenotyping platforms (HTPPs) can rapidly obtain high-resolution top-view images of crop canopies. Researchers can extract phenotypic parameters [8], such as plant size [9], shape [10], and color [11], from the acquired images. For some specific phenotypic parameters, these approaches can be substituted for traditional manual measurements, improving the efficiency of collecting plant phenotypic information.…”