Accurate rice grain counting in natural morphology: A method based on image classification and object detection

Abstract

• The study introduces a solution which achieves accurate extraction of the grain count per rice on their natural form. The findings promise significant advancements in rice grain counting methodologies. • The proposed method counting the natural form grains after rice panicles classification based on the looseness of nature form, which proved to be great improvement in counting accuracy. Offering a novel solution for detecting and counting dense objects. • Randomly capturing and analyzing single-sided images of rice panicles is more practical for obtaining the number of grains in their natural form. The findings offer a possibility of real-time counting of rice grains in the field. Accurately counting the number of grains per panicle is crucial for evaluating rice yield and selecting superior germplasm resources. Traditional measurement methods are labor-intensive, time-consuming, and prone to errors. To address this challenge, computer vision-based methods have emerged as a promising approach for seed counting. However, achieving precise grain counting is particularly challenging due to their natural morphology, which involves occlusion and substantial variations in size, shape and orientation. This often requires additional steps, such as manual shaping or threshing. Therefore, we propose an innovative approach for precisely counting rice grains in their natural form by integrating object detection, image classification and regression equations. Initially, we trained the Yolov7-tiny model for grain counting. Subsequently, we introduced a classification system based on the variability in the natural morphology of rice panicles using the EfficientNetV2 network, enabling the classification of rice panicles into five distinct classes. Furthermore, we devised a set of univariate linear regression equations for the different classes of rice panicles, utilizing data from 2920 diverse rice germplasm to establish correlations predicted and actual values. Experimental findings demonstrated a counting accuracy of 92.60% with an average absolute percentage error of 7.69%. And, this study revealed that utilizing two-sided images of panicles did not significantly improve counting accuracy. This study represents a successful endeavor in achieving precise and efficient counting of rice panicles within their natural morphology, offering a novel solution for detecting and counting dense objects.

Topics & Concepts