关键词: 麦穗, 籽粒表型, 深度学习, 坐标注意力

Abstract:

【Objective】 Wheat grain phenotype parameters were tested after grains only must been threshed by combine, this process was time-consuming, laborious and complicate. Therefore, a method to test morphological parameters of wheat grains on spike based on improved Mask R-CNN was proposed in this research.【Method】Two sides front images of three varieties wheat spikes, including Zhenmai 25, Ningmai 13 and Longmai 88 (Early maturity variety), were collected, and then the image enhancement data set was constructed by using Gaussian filter, salt and pepper noise, and vertical flip image enhancement method. A method combined with deep learning and morphological processing for testing phenotype parameters of wheat grains on spike was proposed. Firstly, the improved Mask R-CNN network model for segmenting spike glume was constructed, which was based on feature extraction networks of ResNet and FNP, and the innovative components Coordinate Attention (CA) module, Aggregation module, and Halfconv module were integrated into it. And the glumes on spike image were accurately detected, located, segmented and counted by the improved Mask R-CNN network model. Secondly, five phenotype parameters of the glumes on spike were extracted by using morphologic processing method from the segmented mask image of wheat spike glume, and the linear correlation equations between the phenotype parameters of the glumes and the phenotype parameters of grains were established. Finally, the linear correlation equations between the phenotype parameters of glume and the phenotypic parameters of grain were used to predict the phenotype parameters of grain.【Result】(1) F1 score, average precision (AP) and recall rate of the optimal model for separating spike glume based on the improved Mask R-CNN network model were 91.12%, 94.13% and 88.30%, respectively, and the average consuming time for detecting a single image was 97 milliseconds, so the improved Mask R-CNN network model could quickly and accurately identify the glumes on the single wheat spike. The root-mean-square error and average relative error for segmenting spike grain by the model were 0.94 piece and 0.65%, respectively, so this showed that spike glumes were segmented precisely by the model. (2) The established linear correlation equations for length, thickness, area, circumference, and length-thickness ratio between wheat spike glume and actual grain were y=0.7258x, y=0.5166x, y=0.3748x, y=0.6756x and y=1.4085x, respectively, and the determination coefficient (R²) of the linear correlation equations all were greater than 0.85. (3) The above correlation equations were verified and phenotype parameters of grain were predicted by using the extracted phenotype parameter data of wheat spike glume, the root-mean-square errors and average relative error for length, thickness, area, circumference and length-thickness ratio between predicted values and actual values of wheat grains were 0.17 mm, 0.08 mm, 0.46 mm², 0.33 mm, 0.12, and 0.02%, 0.02%, 0.02%, 0.03%, respectively. The determination coefficient (R²) for each phenotype parameter between the predicted data and the actual data was above 0.85, research results indicated that the proposed method in this study was feasible【Conclusion】 The method for testing phenotype parameters of wheat grains on spike based on improved Mask R-CNN was proposed in this research, and the phenotype parameters of wheat grain on spike could be predicted accurately and effectively by the phenotype parameters of wheat spike glume. This research provided a new method to extract rapidly and simply wheat grain phenotype parameters on spike.

Key words: wheat spike, grain phenotype, deep learning, Coordinate Attention