Automatic Image Classification Using Neural Networks Increases Accuracy for Allergenic Pollen Monitoring

Abstract

Monitoring of airborne pollen concentrations provides an important source of information for the globally increasing number of hay fever patients. Airborne pollen are traditionally counted under the microscope, but with the latest developments in image recognition methods, automating this process has become feasible. A challenge that persists, however, is that many pollen grains cannot be distinguished beyond the genus or family level using a microscope. Here, we assess the use of a Convolutional Neural Network (CNN) to increase taxonomic accuracy for airborne pollen. As a case study we use the nettle family (Urticaceae), which contains two main genera (Urtica and Parietaria) common in European landscapes which pollen cannot be separated by trained specialists. While pollen from Urtica species have very low allergenic relevance, those from several species of Parietaria are severely allergenic. We collect pollen from both fresh as well as from herbarium specimens and use these to train the CNN model VGG16. The model shows that Urticaceae pollen can be distinguished with 98.3% accuracy. We then apply our model on Urticaceae pollen collected from aerobiological samples and show that the genera can be confidently distinguished, despite the more challenging input images that are often overlain by debris. Our method can also be applied to other pollen families in the future and will thus help to make allergenic pollen monitoring more specific.

Automatic 3D Pollen Recognition Based on Convolutional Neural Network

“…e experimental results are compared to SKLP [18] descriptors, LDP descriptors [14], faster CNN [33], and multi-CNNs [7] to verify the validity of the proposed algorithm on Confocal-E dataset. It can be seen from the three-dimensional pollen images that different textures cover the external wall of various pollen grains, such as thorn, tumor, rod, cave, and net, which are more obvious than those in the twodimensional pollen images.…”

Weakly Supervised Collaborative Learning for Airborne Pollen Segmentation and Classification from SEM Images

“…SEM images can be used to improve the performance of pollen identification in practice [ 20 , 21 , 22 , 23 , 24 , 25 ], since the high resolution of SEM unlocks the potential to provide the obvious differentiation of the pollen grains for the experts (as shown in Table 1 ). In our study, we obtained 1324 SEM pollen images in total (805 Cupressaceae, 248 Fraxinus and 271 Ginkgo) from Beijing Meteorological Center, which served as the data for our automatic pollen identification task.…”

“…Therefore, SEM images have exhibited significant potential to enhance the ability to accurately identify airborne pollens in an automatic manner for the palynology community [ 19 ]. Currently, some preliminary research works on SEM-based automatic pollen identification have been reported [ 20 , 21 , 22 , 23 , 24 , 25 ]. An early example of an SEM-image-based method was [ 20 ], in which the authors extracted the texture features to train a Fisher linear discriminant classifier.…”

“…To subdivide Artemisia pollen into three types, hierarchical clustering analysis on the pollen’s morphological features was carried out by Lu et al [ 23 ]. Polling et al [ 24 ] trained VGG16 to increase the taxonomic accuracy for Urtica and Parietaria. Based on the statistical features and histogram coefficients, MLP was trained in [ 25 ] for SEM pollen image classification.…”