Pollen grain classification has a remarkable role in many fields from medicine to biology and agronomy. Indeed, automatic pollen grain classification is an important task for all related applications and areas. This work presents the first large-scale pollen grain image dataset, including more than 13 thousands objects. After an introduction to the problem of pollen grain classification and its motivations, the paper focuses on the employed data acquisition steps, which include aerobiological sampling, microscope image acquisition, object detection, segmentation and labelling. Furthermore, a baseline experimental assessment for the task of pollen classification on the built dataset, together with discussion on the achieved results, is presented.
Impedance flow cytometry (IFC) is a versatile lab-on-chip technology which enables fast and label-free analysis of pollen grains in various plant species, promising new research possibilities in agriculture and plant breeding. Hazelnut is a monoecious, anemophilous species, exhibiting sporophytic self-incompatibility. Its pollen is dispersed by wind in midwinter when temperatures are still low and relative humidity is usually high. Previous research found that hazelnut can be characterized by high degrees of pollen sterility following a reciprocal chromosome translocation occurring in some cultivated genotypes. In this study, IFC was used for the first time to characterize hazelnut pollen biology. IFC was validated via dye exclusion in microscopy and employed to (i) follow pollen hydration over time to define the best pre-hydration treatment for pollen viability evaluation; (ii) test hazelnut pollen viability and sterility on 33 cultivars grown in a collection field located in central Italy, and two wild hazelnuts. The accessions were also characterized by their amount and distribution of catkins in the tree canopy. Pollen sterility rate greatly varied among hazelnut accessions, with one main group of highly sterile cultivars and a second group, comprising wild genotypes and the remaining cultivars, producing good quality pollen. The results support the hypothesis of recurring reciprocal translocation events in Corylus avellana cultivars, leading to the observed gametic semi-sterility. The measured hazelnut pollen viability was also strongly influenced by pollen hydration (Radj2 = 0.83, P ≤ 0.0001) and reached its maximum at around 6 h of pre-hydration in humid chambers. Viable and dead pollen were best discriminated at around the same time of pollen pre-hydration, suggesting that high humidity levels are required for hazelnut pollen to maintain its functionality. Altogether, our results detail the value of impedance flow cytometry for high throughput phenotyping of hazelnut pollen. Further research is required to clarify the causes of pollen sterility in hazelnut, to confirm the role of reciprocal chromosome translocations and to investigate its effects on plant productivity.
High-quality pollen is a prerequisite for plant reproductive success. Pollen viability and sterility can be routinely assessed using common stains and manual microscope examination, but with low overall statistical power. Current automated methods are primarily directed towards the analysis of pollen sterility, and high throughput solutions for both pollen viability and sterility evaluation are needed that will be consistent with emerging biotechnological strategies for crop improvement. Our goal is to refine established labelling procedures for pollen, based on the combination of fluorescein (FDA) and propidium iodide (PI), and to develop automated solutions for accurately assessing pollen grain images and classifying them for quality. We used open-source software programs (CellProfiler, CellProfiler Analyst, Fiji and R) for analysis of images collected from 10 pollen taxa labelled using FDA/PI. After correcting for image background noise, pollen grain images were examined for quality employing thresholding and segmentation. Supervised and unsupervised classification of per-object features was employed for the identification of viable, dead and sterile pollen. The combination of FDA and PI dyes was able to differentiate between viable, dead and sterile pollen in all the analysed taxa. Automated image analysis and classification significantly increased the statistical power of the pollen viability assay, identifying more than 75,000 pollen grains with high accuracy (R2 = 0.99) when compared to classical manual counting. Overall, we provide a comprehensive set of methodologies as baseline for the automated assessment of pollen viability using fluorescence microscopy, which can be combined with manual and mechanized imaging systems in fundamental and applied research on plant biology. We also supply the complete set of pollen images (the FDA/PI pollen dataset) to the scientific community for future research.
Woody species encroachment on grassland ecosystems is occurring worldwide with both negative and positive consequences for biodiversity conservation and ecosystem services. Remote sensing and image analysis represent useful tools for the monitoring of this process. In this paper, we aimed at evaluating quantitatively the potential of using high-resolution UAV imagery to monitor the encroachment process during its early development and at comparing the performance of manual and semi-automatic classification methods. The RGB images of an abandoned subalpine grassland on the Western Italian Alps were acquired by drone and then classified through manual photo-interpretation, with both pixel- and object-based semi-automatic models, using machine-learning algorithms. The classification techniques were applied at different resolution levels and tested for their accuracy against reference data including measurements of tree dimensions collected in the field. Results showed that the most accurate method was the photo-interpretation (≈99%), followed by the pixel-based approach (≈86%) that was faster than the manual technique and more accurate than the object-based one (≈78%). The dimensional threshold for juvenile tree detection was lower for the photo-interpretation but comparable to the pixel-based one. Therefore, for the encroachment mapping at its early stages, the pixel-based approach proved to be a promising and pragmatic choice.
Hazelnut cultivation is rapidly expanding in regions outside its native range. New hazelnut plantations in South Africa are facing adverse environmental conditions which threaten the pollination process and hamper nut yield. Artificial pollination can increase fruit yield and fill pollination gaps in some fruit crops, but its application on hazelnut is still not well explored. This study investigated biological factors and technical procedures in the first artificial pollination experiment on hazelnut in South Africa. A suspension media for artificial application composed of 10% sucrose, 0,1% agar and 0,02% boric acid was used. In addition, alternative low-cost suspension media containing other forms of sugar were also evaluated. Moreover, a novel and practical approach to assess pollen conditions in liquid solutions was designed. Pollen viability was tested in the different suspension media. This varied greatly among the examined cultivars. Wild hazelnut produced the pollen with the highest viability, while Tonda Gentile delle Langhe (TGL) and Barcelona the lowest. Sterile grains were very abundant, especially in cultivated varieties (up to 65% in TGL). Preliminary data on nut yield were also collected. Altogether, this study indicated that artificial pollination is a promising approach for hazelnut cultivation in increasing the final yield. Further research is needed to develop integrated pollination strategies, especially in areas where the environmental conditions are adverse to the pollination process.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.