Reconstruction of damaged herbarium leaves using deep learning techniques for improving classification accuracy

Malik, Owais Ahmed; Ong, Wee-Hong; Slik, Johan Willem Frederik

doi:10.1016/j.ecoinf.2021.101243

Cited by 13 publications

(12 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Leaf architecture has been an important topic in taxonomy (Bucksch 2014), especially encountering flowering periodization (Jones 1986;Sack et al 2008), and leaves as identification tools in the flowering plants are bound to be very useful in the future. The predicted leaf architecture is then reconstructed using computational modelling (Hussein et al 2021) to identify severely degraded herbarium specimens (Little et al 2014). The petiole anatomy has proven to be useful as an identification tool in Shorea spp.…”

Section: Leaf Character Variationmentioning

confidence: 99%

Short communication: Leaf architecture of 35 species of Dipterocarpaceae cultivated in Forest Area with Special Purposes in Carita, Banten, Indonesia

et al. 2021

View full text Add to dashboard Cite

Abstract. MeinataA, Na’iem M, Adriyanti DT, Syahbudin A. 2021. Short communication: Leaf architecture of 35 species of Dipterocarpaceae cultivated in Forest Area with Special Purposes in Carita, Banten, Indonesia. Biodiversitas 22: 2952-2960. Dipterocarpaceae is a major commercial timber characterized by high unbranched bole, paired stipules, and winged fruit. The identification process in the family becomes problematic, in cases where the generative organ is absent. Therefore, a new approach needs to be established to address any misidentification leading to improper utilization. This study aims to determine the leaf architecture in 35 species of Dipterocarpaceae cultivated in Forest Area with Special Purposes (KHDTK) Carita, Banten, Indonesia. The ten leaf samples collected were the third and fourth leaves from terminal shoots of a single tree of each species. Subsequently, the data were observed and measured based on the 17 morphological characters. Each character was then scored and analyzed using multivariate analyses cluster to determine the relationship between species. Dipterocarpaceae generally has pinnate leaf category, geniculate petiole, and entire margin. Furthermore, the phenon line in the dendrogram is cut at 0.695 similarity level to establish meaningful interpretation. The 35 species observed were grouped into 4 major clusters. Small leaf group, inconsistent tertiary vein pattern group, symmetrical leaves with cordate base leaf group, and other groups that do not resemble mentioned characters. The key determination comprised 34 couplets with three characters repeated due to limited descriptors.

show abstract

Section: Leaf Character Variationmentioning

confidence: 99%

Short communication: Leaf architecture of 35 species of Dipterocarpaceae cultivated in Forest Area with Special Purposes in Carita, Banten, Indonesia

et al. 2021

View full text Add to dashboard Cite

show abstract

“…This is due to the upward looping of the secondary vein with a weak joint between its upper venation. In terms of species delineation, the 2 o vein has a strong significance in several studies such as the results obtained from Sapotaceae and Malvaceae (Hussein et al 2021). This has been proven to separate two different genera in Dipterocarpaceae, i.e.…”

Section: Along With Genetic Factors (De Kort Et Al 2021)mentioning

confidence: 96%

Short communication: Variations in leaf morphological characters of Shorea leprosula in progeny trial stand of a logging concession in Kalimantan, Indonesia

Meinata

Adriyanti

2021

Biodiversitas

View full text Add to dashboard Cite

Indonesia has the highest distribution of Dipterocarpaceae members, and Shorea leprosula is one of the species that has a natural hybrid in its habitat. Furthermore, the members have intermediate morphological character with the neighboring species, Shorea curtisii. This study aimed to investigate the morphological variations of Shorea leprosula Miq. Progeny trial in PT Sari Bumi Kusuma. The morphological level of 72 Shorea leprosula in PT Sari Bumi Kusuma was identified through sampling. Furthermore, macroscopic and microscopic observations were conducted, and the measurement data were analyzed using cluster and principal component analyses to explain the morphological variable contribution. The results showed leaf architecture variations in laminar shape, apex shape, base shape, and midrib thickness category. The cluster analyses classified the samples into four cluster groups and they consist of a notophyll leaf size category with a rounded base. Meanwhile, the second group consists of an ellipse laminar shape with medium midrib thickness, and the third has an oblong laminar with an obtuse apex shape. The fourth group has a notophyll leaf size category with an obtuse base. The principal component analysis showed that the base shape has the highest contribution to diversity in the samples.

show abstract

“…These frameworks are extremely flexible, well‐supported, and surprisingly approachable. As a result, many recent projects have also coalesced around these two frameworks with great success, including efforts to segment leaves (Younis et al, 2020; Triki et al, 2020, 2021; Guo et al, 2021; Hussein et al, 2021b; Gu et al, 2022; Ott and Lautenschlager, 2022), segment plant tissue (Love et al, 2021; Goëau et al, 2022; Milleville et al, 2023), isolate plant organs (Davis et al, 2020; Pearson et al, 2020; Triki et al, 2020; Ott and Lautenschlager, 2022), extract specimen label data (Milleville et al, 2023), isolate diseased or damaged leaf tissue (Kaur et al, 2022; Mu et al, 2022; Kavitha Lakshmi and Savarimuthu, 2023), measure bird skeletons (Weeks et al, 2023), isolate preserved snakes (Curlis et al, 2022), segment fossils (Panigrahi et al, 2022), or remotely monitor phenology (Mann et al, 2022). However, rather than relying on a single machine learning architecture to extract trait and archival data from specimens, we developed a modular framework of seven different machine learning algorithms that work in tandem to comprehensively process each image (Table 2, Figure 1).…”

Section: Term Definitionmentioning

confidence: 99%

From leaves to labels: Building modular machine learning networks for rapid herbarium specimen analysis with LeafMachine2

Weaver,

Smith

2023

Appl Plant Sci

View full text Add to dashboard Cite

PremiseQuantitative plant traits play a crucial role in biological research. However, traditional methods for measuring plant morphology are time consuming and have limited scalability. We present LeafMachine2, a suite of modular machine learning and computer vision tools that can automatically extract a base set of leaf traits from digital plant data sets.MethodsLeafMachine2 was trained on 494,766 manually prepared annotations from 5648 herbarium images obtained from 288 institutions and representing 2663 species; it employs a set of plant component detection and segmentation algorithms to isolate individual leaves, petioles, fruits, flowers, wood samples, buds, and roots. Our landmarking network automatically identifies and measures nine pseudo‐landmarks that occur on most broadleaf taxa. Text labels and barcodes are automatically identified by an archival component detector and are prepared for optical character recognition methods or natural language processing algorithms.ResultsLeafMachine2 can extract trait data from at least 245 angiosperm families and calculate pixel‐to‐metric conversion factors for 26 commonly used ruler types.DiscussionLeafMachine2 is a highly efficient tool for generating large quantities of plant trait data, even from occluded or overlapping leaves, field images, and non‐archival data sets. Our project, along with similar initiatives, has made significant progress in removing the bottleneck in plant trait data acquisition from herbarium specimens and shifted the focus toward the crucial task of data revision and quality control.

show abstract

Reconstruction of damaged herbarium leaves using deep learning techniques for improving classification accuracy

Cited by 13 publications

References 17 publications

Short communication: Leaf architecture of 35 species of Dipterocarpaceae cultivated in Forest Area with Special Purposes in Carita, Banten, Indonesia

Short communication: Leaf architecture of 35 species of Dipterocarpaceae cultivated in Forest Area with Special Purposes in Carita, Banten, Indonesia

Short communication: Variations in leaf morphological characters of Shorea leprosula in progeny trial stand of a logging concession in Kalimantan, Indonesia

From leaves to labels: Building modular machine learning networks for rapid herbarium specimen analysis with LeafMachine2

Contact Info

Product

Resources

About