Detection and annotation of plant organs from digitised herbarium scans using deep learning

Younis, Sohaib; Schmidt, Marco; Weiland, Claus; Dressler, Steffan; Seeger, Bernhard; Hickler, Thomas

doi:10.3897/bdj.8.e57090

Cited by 25 publications

(17 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Their study trained a Faster R-CNN model and reported an accuracy of 95% on a sub-set of 61 test images. Similarly, Younis et al [ 30 ] proposed a Faster R-CNN model to detect and annotate different plant organs from digitized herbarium specimens. The authors manually annotated hundreds of images and used a subset of 498 images to train the model to detect different organs including flowers, leaf, fruit, seed, root, and stem.…”

Section: Related Workmentioning

confidence: 99%

Automated Extraction of Phenotypic Leaf Traits of Individual Intact Herbarium Leaves from Herbarium Specimen Images Using Deep Learning Based Semantic Segmentation

Malik

Ong

Slik

2021

Sensors

View full text Add to dashboard Cite

With the increase in the digitization efforts of herbarium collections worldwide, dataset repositories such as iDigBio and GBIF now have hundreds of thousands of herbarium sheet images ready for exploration. Although this serves as a new source of plant leaves data, herbarium datasets have an inherent challenge to deal with the sheets containing other non-plant objects such as color charts, barcodes, and labels. Even for the plant part itself, a combination of different overlapping, damaged, and intact individual leaves exist together with other plant organs such as stems and fruits, which increases the complexity of leaf trait extraction and analysis. Focusing on segmentation and trait extraction on individual intact herbarium leaves, this study proposes a pipeline consisting of deep learning semantic segmentation model (DeepLabv3+), connected component analysis, and a single-leaf classifier trained on binary images to automate the extraction of an intact individual leaf with phenotypic traits. The proposed method achieved a higher F1-score for both the in-house dataset (96%) and on a publicly available herbarium dataset (93%) compared to object detection-based approaches including Faster R-CNN and YOLOv5. Furthermore, using the proposed approach, the phenotypic measurements extracted from the segmented individual leaves were closer to the ground truth measurements, which suggests the importance of the segmentation process in handling background noise. Compared to the object detection-based approaches, the proposed method showed a promising direction toward an autonomous tool for the extraction of individual leaves together with their trait data directly from herbarium specimen images.

show abstract

Section: Related Workmentioning

confidence: 99%

Automated Extraction of Phenotypic Leaf Traits of Individual Intact Herbarium Leaves from Herbarium Specimen Images Using Deep Learning Based Semantic Segmentation

Malik

Ong

Slik

2021

Sensors

View full text Add to dashboard Cite

show abstract

“…This means related datasets either focus on detecting entire plants or individual organs, such as leaves, from a singular plant species. Recently, a few small-scale plant organ detection approaches and datasets using more than one species have been developed, each using at most a few hundred herbarium sheets with the intended use of phenological information gathering [36,22,32]. In the context of species identification, these approaches are unable to collect data on present biodiversity and the geographical distribution of different plant species, an advantage that crowdsourcing initiatives possess.…”

Section: Related Workmentioning

confidence: 99%

Fine-Grained Visual Classification of Plant Species In The Wild: Object Detection as A Reinforced Means of Attention

Keaton,

Zaveri,

Kovur

et al. 2021

Preprint

View full text Add to dashboard Cite

Plant species identification in the wild is a difficult problem in part due to the high variability of the input data, but also because of complications induced by the long-tail effects of the datasets distribution. Inspired by the most recent fine-grained visual classification approaches which are based on attention to mitigate the effects of data variability, we explore the idea of using object detection as a form of attention. We introduce a bottom-up approach based on detecting plant organs and fusing the predictions of a variable number of organ-based species classifiers. We also curate a new dataset with a long-tail distribution for evaluating plant organ detection and organ-based species identification, which is publicly available 1 .

show abstract

“…Afonso et al., 2020; Buddha et al., 2019), detection of leaves and other plant organs on herbarium specimens (e.g. Ott et al., 2020; Weaver et al., 2020; Younis et al., 2020), stomata counting using microscopic leaf images (e.g. Fetter et al., 2019), animal counting using camera traps (Norouzzadeh et al., 2021) and many more.…”

Section: Introductionmentioning

confidence: 99%

GinJinn2: Object detection and segmentation for ecology and evolution

Ott

Lautenschlager

2022

Methods Ecol Evol

View full text Add to dashboard Cite

Leveraging image data for ecological and evolutionary/systematic research typically requires substantial effort for data collection and preparation. The ability to automate time-consuming steps of this process, possibly along with further downstream analyses, for example, using programming languages like Python or R, can not only increase productivity, but also allow otherwise infeasible large-scale analyses. Recent advances in machine learning (ML), both on the soft-and hardware side, make it even possible to automate tasks that are difficult to solve by means of classically designed algorithms. Computer vision, in particular, has largely profited from deep

show abstract

Detection and annotation of plant organs from digitised herbarium scans using deep learning

Cited by 25 publications

References 36 publications

Automated Extraction of Phenotypic Leaf Traits of Individual Intact Herbarium Leaves from Herbarium Specimen Images Using Deep Learning Based Semantic Segmentation

Automated Extraction of Phenotypic Leaf Traits of Individual Intact Herbarium Leaves from Herbarium Specimen Images Using Deep Learning Based Semantic Segmentation

Fine-Grained Visual Classification of Plant Species In The Wild: Object Detection as A Reinforced Means of Attention

GinJinn2: Object detection and segmentation for ecology and evolution

Contact Info

Product

Resources

About