Maize-PAS: Automated Maize Phenotyping Analysis Software using Deep Learning

Abstract

Background: Maize (Zea mays L.) is one of the most important food sources in the world and has been one of the main targets of plant genetics and phenotypic research for centuries. Observation and analysis of various morphological phenotypic traits during maize growth are essential for genetic and breeding study. The generally huge number of samples produce an enormous amount of high-resolution image data. While high throughput plant phenotyping platforms are increasingly used in maize breeding trials, there is a reasonable need for software tools that can automatically identify visual phenotypic features of maize plants and implement batch processing on image datasets.Results: On the boundary between computer vision and plant science, we utilize advanced deep learning methods based on convolutional neural networks to empower the workflow of maize phenotyping analysis. This paper presents Maize-PAS ( Maize Phenotyping Analysis Software), an integrated application supporting one-click analysis of maize phenotype, embedding multiple functions: I. Projection, II. Color Analysis, III. Internode length, IV. Height, V. Stem Diameter and VI. Leaves Counting. Taking the RGB image of maize as input, the software provides a user-friendly graphical interaction interface and rapid calculation of multiple important phenotypic characteristics, including leaf sheath points detection and leaves segmentation. In function Leaves Counting, the mean and standard deviation of difference between prediction and ground truth are 1.60 and 1.625.Conclusion: The Maize-PAS is easy-to-use and demands neither professional knowledge of computer vision nor deep learning. All functions for batch processing are incorporated, enabling automated and labor-reduced tasks of recording, measurement and quantitative analysis of maize growth traits on a large dataset. We prove the efficiency and potential capability of our techniques and software to image-based plant research, which also demonstrates the feasibility and capability of AI technology implemented in agriculture and plant science.

Automation of leaf counting in maize and sorghum using deep learning

“…However, leaf overlap and self occlusion will be an inherent problem whenever a 3D plant is represented as a 2D image. A similar object detection model was recently employed to estimate leaf number by detecting whole leaves in maize images (Zhou et al., 2020). However, this approach also suffers from issues with occlusion.…”

“…Leaf counting has been shown to be practical in sorghum when using depth cameras to reconstruct 3D models of individual plants (McCormick et al., 2016; Xiang et al., 2019). However, only limited work has been conducted in leaf counting from 2D images of grain crops such as maize and sorghum (Pound et al., 2017; Zhou et al., 2020), potentially because of the lack of large, annotated, public training datasets for them. Unlike rosette plants, counting leaves from top‐down images is not practical in these species (Figure 1d).…”

Automation of Leaf Counting in Maize and Sorghum Using Deep Learning

“…However, leaf overlap and self occlusion will be an inherent problem whenever a 3D plant is represented as a 2D image. A similar object detection model was recently employed to estimate leaf numbers by detecting whole leaves in maize images (Zhou et al, 2020). However, this approach also suffers from issues with occlusion.…”

“…These crop species also provide roughly half of all calories consumed by humans around the globe (Tester and Langridge, 2010). Leaf counting has been shown to be practical in sorghum when using depth cameras to reconstruct 3D models of individual plants (McCormick et al, 2016;Xiang et al, 2019) However, only limited work has been conducted in leaf counting from 2D images of grain crops such as maize and sorghum (Pound et al, 2017;Zhou et al, 2020), potentially because of the lack of large, annotated, public training datasets for them. Unlike rosette plants, counting leaves from topdown images is not practical in these species ( Figure 1D).…”