A deep learning approach to track Arabidopsis seedlings’ circumnutation from time-lapse videos

Abstract

Background Circumnutation (Darwin et al., Sci Rep 10(1):1–13, 2000) is the side-to-side movement common among growing plant appendages but the purpose of circumnutation is not always clear. Accurately tracking and quantifying circumnutation can help researchers to better study its underlying purpose. Results In this paper, a deep learning-based model is proposed to track the circumnutating flowering apices in the plant Arabidopsis thaliana from time-lapse videos. By utilizing U-Net to segment the apex, and combining it with the model update mechanism, pre- and post- processing steps, the proposed model significantly improves the tracking time and accuracy over other baseline tracking methods. Additionally, we evaluate the computational complexity of the proposed model and further develop a method to accelerate the inference speed of the model. The fast algorithm can track the apices in real-time on a computer without a dedicated GPU. Conclusion We demonstrate that the accuracy of tracking the flowering apices in the plant Arabidopsis thaliana can be improved with our proposed deep learning-based model in terms of both the racking success rate and the tracking error. We also show that the improvement in the tracking accuracy is statistically significant. The time-lapse video dataset of Arabidopsis is also provided which can be used for future studies on Arabidopsis in various takes.

The Development of a Stereo Vision System to Study the Nutation Movement of Climbing Plants

“…In the inertial tracker proposed by Geldhof et al [15], the errors are expressed as the precision of pitch and roll angles, being under 0.5 • . The recent method by Mao et al [13] using deep learning is able to achieve an average error of 1.02 mm in the location of the apex of the plant, in videos of 640 × 480 px of resolution; however, only one video is analyzed per plant, so the 3D locations are not extracted.…”

“…U-net, Linknet, FPN, PSPNet, and a 34-layer ResNet architecture were used for comparison, obtaining a maximum area under the curve (AUC) for the validation set over 0.92. Another recent method based on deep learning was presented by Mao et al [13], which proposed a U-net network architecture to segment the plant. This method is an improvement of their free software "Plant Tracer", which analyzed plant movement using the classic computer vision techniques of object tracking and blob detection.…”

The Early Dodder Gets the Host: Decoding the Coiling Patterns ofCuscuta campestriswith Automated Image Processing

“…In a digital data-dominated era, open-sourced high-throughput plant phenotyping (HTPP) pipelines are crucial to gain better and more profound insights from imagebased data (Araus and Cairns 2014;Fahlgren, et al 2015). For example, 2D image-processing HTTP pipelines can automatically track size changes in Arabidopsis leaves (Swartz, et al 2023), detect plant-pathogen interactions in excavated maize roots (Pierz, et al 2023), characterize leaf venation patterns for grasses (Robil, et al 2021), and even trace circumnutation movements made by Arabidopsis stems (Mao, et al 2023). However, adapting existing HTPP pipelines for C. campestris phenotyping presents challenges due to several unique features of the organism.…”