Physiological and Biochemical Analysis Revealing the Key Factors Influencing 2-Phenylethanol and Benzyl Alcohol Production in Crabapple Flowers.

Floral scent (FS) plays a crucial role in the ecological functions and industrial applications of plants. However, the physiological and metabolic mechanisms underlying FS formation remain inadequately explored. Our investigation focused on elucidating the differential formation mechanisms of 2-phenylethanol (2-PE) and benzyl alcohol (BA) by examining seven related enzyme concentrations and the content of soluble sugar, soluble proteins, carbon (C) and nitrogen (N), as well as the C/N ratio. The findings revealed that the peak content of 2-PE in M. 'Praire Rose' and BA in M. 'Lollipop' occurred during the end flowering stage (S4) and flowering stage (S3) periods, respectively. The enzyme concentration change trends of phenylpyruvate decarboxylase (PDL), phenylacetaldehyde reductase (PAR), soluble protein, C, N, and C/N ratio changes during the S3-S4 period in M. 'Praire Rose' and M. 'Lollipop' were entirely opposite. Correlation and PCA analysis demonstrated that the content of CYP79D73 (a P450) and N, and the C/N ratio were key factors in 2-PE production in M. 'Praire Rose'. The production of BA in M. 'Lollipop' was more influenced by the content of phenylacetaldehyde synthase (PAAS), CYP79D73, and soluble sugar. As CYP79D73 exits oppositely in correlation to 2-PE (M. 'Praire Rose') and BA (M. 'Lollipop'), it is hypothesized that CYP79D73 was postulated as the primary factor contributing to the observed differences of 2-PE (M. 'Praire Rose') and BA (M. 'Lollipop') formation. These results carry significant implications for crabapple aromatic flower breeding and the essential oil industry etc.

Semi-supervised Learning for Weed and Crop Segmentation Using UAV Imagery.

Weed control has received great attention due to its significant influence on crop yield and food production. Accurate mapping of crop and weed is a prerequisite for the development of an automatic weed management system. In this paper, we propose a weed and crop segmentation method, SemiWeedNet, to accurately identify the weed with varying size in complex environment, where semi-supervised learning is employed to reduce the requirement of a large amount of labelled data. SemiWeedNet takes the labelled and unlabelled images into account when generating a unified semi-supervised architecture based on semantic segmentation model. A multiscale enhancement module is created by integrating the encoded feature with the selective kernel attention, to highlight the significant features of the weed and crop while alleviating the influence of complex background. To address the problem caused by the similarity and overlapping between crop and weed, an online hard example mining (OHEM) is introduced to refine the labelled data training. This forces the model to focus more on pixels that are not easily distinguished, and thus effectively improve the image segmentation. To further exploit the meaningful information of unlabelled data, consistency regularisation is introduced by maintaining the context consistency during training, making the representations robust to the varying environment. Comparative experiments are conducted on a publicly available dataset. The results show the SemiWeedNet outperforms the state-of-the-art methods, and its components have promising potential in improving segmentation.