Study on Counting Method of on Panicle Rice Grains Based on Deep Learning
Journal Title: Journal of Shenyang Agricultural University - Year 2025, Vol 56, Issue 1
Abstract
[Objective]Rice grain counting is a crucial step in rice seed testing. To address the inefficiencies and errors associated with traditional manual counting of rice grains, this study constructed an in-situ counting model for rice grains. In-situ counting method can not destroy the original topological structure of rice panicles, and then it can be further applied to obtain other phenotypic parameters. [Methods]The model employs ResNet as its backbone network and predicts the probability density distribution of rice grains by leveraging the feature correlation between image and rice grain exemplars. Subsequently, the number of rice grains is the proposed model is reduced by 2.2%. When compared to the SAM (Segment Anything Model) based on instance segmentation, the MRE of the proposed model decreases by 12.2%, and compared to the T-Rex2 model, it is reduced by 6.5%. [Conclusion]This research method is based on the deep learning model, which can automatically identify and count rice grains in the image and improve the counting efficiency. At the same time, compared with other deep learning models, this research model has stronger learning ability with few samples. The method presented in this paper can be effectively applied to the task of rice grain counting in panicles, and the research provides valuable insights for obtaining phenotypic parameters of rice panicles. obtained by summing the density maps. An image dataset of rice panicles was collected, and a loss function tailored for rice grain counting on panicles was defined. This function takes into account both the consistency between the predicted density map and the actual rice grain distribution, as well as the relevant constraints of the exemplar labeling box. The performance of the model was evaluated using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Relative Error (MRE). Experimental [Results]Utilizing ResNet50 as the model's backbone network achieves impressive accuracy, with MAE, RMSE, and MRE values of 10.937, 19.286, and 13.4%, respectively. This method exhibits superior counting performance. Compared to YOLOv8-seg, the MRE of
Authors and Affiliations
ZHOU Yuncheng, ZHANG Yu, LIU Zeyu, LI Ruiyang
Keywords
Related Articles
- EP ID EP764961
- DOI DOI:10.3969/j.issn.1000-1700.2025.01.009
- Views 5
- Downloads 0
Big Data Platform for Comprehensive Supervision of Laying Hens Industry: Framework, Technology, and Application
[Objective]The high-quality development of the laying hens industry plays an important role in the construction of a modern industrial system and the comprehensive realization of rural revitalization in China. In respon...
Landscape Ecological Risk Assessment for the Black Soil Region in Northeast China Based on Land Use Change
[Objective]As one of the four major black soil regions in the world, the Northeast black soil region is home to precious black soil resources and plays an important role in safeguarding national security. [Methods]Based...
Exploring QTL for Early Heading of Weedy Rice Based on High-density Genetic Map
[Objective]The heading date of rice (Oryza sativa L.) is a quantitative trait regulated by multiple genes, which affects the yield and quality of rice, and determines the planting area and distribution range, heading mar...
Spatial-temporal Evolution and Influencing Factors of Green Utilization Efficiency of Cultivated Land in Lower Liaohe Plain
[Objective]This research examines the laws governing the evolution of green land-use efficiency in cultivated areas of the Liaohe Plain, both temporally and spatially, as well as the underlying factors. It aims to provi...
Design and Testing of Soil Crushing and Compacting Machine Based on Corn Strip Tillage Technology
[Objective]At present, the soil breaking in the domestic corn strip ploughing operation has the problem that the soil is finely broken after rotary ploughing, and the moisture loss is large, which is easy to cause dust...