A method of cotton root segmentation based on edge devices

Abstract

The root is an important organ for plants to absorb water and nutrients. In situ root research method is an intuitive method to explore root phenotype and its change dynamics. At present, in situ root research, roots can be accurately extracted from in situ root images, but there are still problems such as low analysis efficiency, high acquisition cost, and difficult deployment of image acquisition devices outdoors. Therefore, this study designed a precise extraction method of in situ roots based on semantic segmentation model and edge device deployment. It initially proposes two data expansion methods, pixel by pixel and equal proportion, expand 100 original images to 1600 and 53193 respectively. It then presents an improved DeeplabV3+ root segmentation model based on CBAM and ASPP in series is designed, and the segmentation accuracy is 93.01%. The root phenotype parameters were verified through the Rhizo Vision Explorers platform, and the root length error was 0.669%, and the root diameter error was 1.003%. It afterwards designs a time-saving Fast prediction strategy. Compared with the Normal prediction strategy, the time consumption is reduced by 22.71% on GPU and 36.85% in raspberry pie. It ultimately deploys the model to Raspberry Pie, realizing the low-cost and portable root image acquisition and segmentation, which is conducive to outdoor deployment. In addition, the cost accounting is only $247. It takes 8 hours to perform image acquisition and segmentation tasks, and the power consumption is as low as 0.051kWh. In conclusion, the method proposed in this study has good performance in model accuracy, economic cost, energy consumption, etc. This paper realizes low-cost and high-precision segmentation of in-situ root based on edge equipment, which provides new insights for high-throughput field research and application of in-situ root.

Application of Improved UNet and EnglightenGAN for Segmentation and Reconstruction of In Situ Roots

“…This study is based on previous studies, the most widely used networks for in situ root image segmentation are SegNet [ 25 ], DeeplabV3+ [ 36 – 38 ], and UNet [ 28 , 44 ]. Therefore, this study compares these 3 networks.…”

“…The RhizoPot used to collect plant root images and the corresponding control terminal are placed in a laboratory with constant temperature and light. The image acquisition is carried out through the automatic acquisition scheme based on the same interval time proposed in the previous research [ 38 ] so that the time and space of image acquisition are consistent. Given the influence of root hair interference and labeling errors, the experimental group will employ professional labelers to manually relabel the root and root hairs to minimize the impact of labeling results on reconstruction.…”

“…The RhizoPot used to collect plant root images and the corresponding control terminal are placed in a laboratory with constant temperature and light. The image acquisition is carried out through the automatic acquisition scheme based on the same interval time proposed in the previous research [ 38 ] so that the time and space of image acquisition are consistent.…”

“…Based on the improvement of DeeplabV3+, our research team proved the possibility of accurately segmenting the root in the early stage, but there is a problem the root diameter and surface area are far from the actual value in the root phenotype analysis results [ 36 ]. In the follow-up, the research team conducted continuous research on the accurate segmentation method of roots [ 37 , 38 ], which improved the accuracy of in situ root segmentation in the soil background, but the small pieces covered by the soil cannot be identified. The purpose of this article is to repair the root covered by the soil and solve the phenomenon that the deep learning segmentation of the in situ root cannot be fully fitted.…”

A Review of Deep Learning in The Field of Plant Root Segmentation

“…Although the current research has achieved good results in the accurate acquisition of roots, the requirements for equipment are becoming higher and higher, resulting in increasing costs. Yu et al [53] used DeepLabV3+ semantic segmentation model to design a time-saving fast prediction strategy to achieve low-cost and portable root image acquisition and segmentation. Although the time and cost were reduced, the extraction accuracy would also be reduced.…”