New DEKR-Sprior model revolutionizes high-throughput. newswise

Newswise – A research team developed the DEKR-Sprior model to improve high-throughput phenotyping of soybean pods and seeds. This model, which enhances feature discrimination through a new SPrior module, significantly reduces the average absolute error in pod phenotyping compared to existing models. DEKR-SPrior’s ability to accurately count and locate densely packed pods and seeds promises to streamline soybean breeding processes, enhancing crop yield prediction and advancing agricultural research. Provides a valuable tool.

Soybean is an agriculturally important legume rich in protein and oil, with breeders aiming to increase yields through traits such as seed weight, size and number of pods. Current research leverages deep learning (DL) for high-throughput phenotyping, yet traditional methods are laborious and error-prone. Segmentation-based and detection-based DL methods face challenges with dense, overlapping pods. To address these issues, attention has focused on the exploration of point-based detection methods, such as P2PNet, to accurately phenotype soybean pods and seeds in situ.

A study published in Plant Phenomics on 27 June 2024 (DOI: 10.34133/plantphenomics.0198) proposes the DEKR-Sprior model incorporating structural prior knowledge to improve the accuracy of soybean pod phenotyping.

In this study, the performance of the DEKR-Sprior model was compared with four other bottom-up models—Lightweight-OpenPose, OpenPose, HigherHRNet, and the original DEKR, on a high-resolution subimage dataset containing images of 205 cropped soybean plants . DEKR-SPrior achieved better accuracy with AP, AP50, AP(1-seeded), AP(2-seeded), AP(3-seeded), and AP(4-seeded) values ​​of 72.4%, 91.4%, 71.7%. Exhibited. , 80.9%, 85.6%, and 83.6%, respectively. Compared to the original DEKR, DEKR-Sprior showed significant improvements in all metrics, with particularly significant gains in AP for 2-seeded and 3-seeded pods. Precision-recall (PR) curves indicated that DEKR-Sprior maintained high precision at given recall rates, effectively reducing misses and misidentifications. Visualization of the results revealed accurate identification and connection of seed positions, even in densely packed beans. Ablation analysis confirmed the enhancement provided by the SPrior module, with optimal performance achieved at a specific hyperparameter value. DEKR-SPrior also outperformed other models in full-size image tests, achieving low mean absolute errors (MAE) and high Pearson correlation coefficients (PCC) for both seeds and pods, underscoring its efficacy in soybean phenotyping. Does.

According to Jingjing He, lead researcher of the study, “This paper demonstrated the great potential of DEKR-SPrior for plant phenotyping, and we hope that DEKR-SPrior will help in future plant phenotyping.”

In summary, the DEKR-SPrior model achieved high precision and recall rates, demonstrating its effectiveness in accurately detecting and counting soybean pods and seeds. Looking to the future, DEKR-SPrior holds great potential to advance agricultural research and breeding programs by providing a more accurate and efficient method for phenotyping crop traits. This model can be further refined and adapted to other crops, thereby enhancing yield prediction and contributing to food security.

,

Reference

DOI

10.34133/plantphenomics.0198

original source url

https://spj.science.org/doi/10.34133/plantphenomics.0198

Author

Jingjing He1*, Lin Weng1, Xiaogang Xu2, Ruochen Chen1, Bo Peng1, Nannan Li1, Zhengchao Xie1, Lijian Sun1, Qiang Han1, Pengfei He1, Fangfang Wang1, Hui Yu3, Javed Akhtar Bhat1, and Jianzhong fang3

Affiliation

1Zhejiang Laboratory, Hangzhou 311100, Zhejiang, China.

2School of Computer Science and Technology, Zhejiang Gongshang University, Hangzhou 310012, Zhejiang, China.

3 Key Laboratory of Soybean Molecular Design Breeding, Northeast Institute of Geography and Agroecology, Chinese Academy of Sciences, Changchun 130102, Jilin, China.

Funding Information

This work was partially supported by the National Key Research and Development Program of China (2023YFD1202600), the National Natural Science Foundation of China (62103380), the Research and Development Project of the Department of Science and Technology of Zhejiang Province (2023C01042). Soybean Intelligent Computational Breeding and Application of Zhejiang Lab (2021PE0AC04), Intelligent Technology and Platform Development for Rice Breeding of Zhejiang Lab (2021PE0AC05), and Fine-Grained Semantic Modeling and Cross-Modal Encoding-Decoding for Multilingual Visual Text Extraction (2022M722911 ) ).

About Plant Phenomics

Plant Fenomics is an open access journal published in collaboration with the State Key Laboratory of Crop Genetics and Germplasm Enhancement, Nanjing Agricultural University (NAU) and published by the American Association for the Advancement of Science (AAAS). Like all partners participating in the Science Partner Journal Programme, Plant Fenomics is editorially independent from the Science family of journals. Editorial decisions and scientific activities adopted by the Journal’s Editorial Board are carried out independently, on the basis of scientific merit and adhering to the highest standards for the accurate and ethical promotion of science. These decisions and activities are in no way influenced by the financial support of NAU, the NAU administration, or any other institutions and sponsors. The Editorial Board is solely responsible for all content published in the magazine. To learn more about the Science Partner Journal Program, visit the SPJ program homepage.