A synthetic datasets based instance segmentation network for High-throughput soybean pods phenotype investigation

作者：

Highlights：

• Designed an Expert System for high throughput soybean pods phenotype investigation.

• A novel method for synthesizing training datasets automatically.

• A hybrid sim/real dataset designed for soybean pods instance segmentation.

• Significant improvement on accuracy of phenotype by segmented-mask based method.

摘要

•Designed an Expert System for high throughput soybean pods phenotype investigation.•A novel method for synthesizing training datasets automatically.•A hybrid sim/real dataset designed for soybean pods instance segmentation.•Significant improvement on accuracy of phenotype by segmented-mask based method.

论文关键词：Deep learning,Synthetic dataset,Instance segmentation,Plant phenotyping,High throughput

论文评审过程：Received 3 March 2021, Revised 13 November 2021, Accepted 11 December 2021, Available online 21 December 2021, Version of Record 4 January 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2021.116403