Dataset Information

BDMediLeaves: A leaf images dataset for Bangladeshi medicinal plants identification.

ABSTRACT: This paper introduces a newly curated dataset named "BDMediLeaves" that includes a diverse collection of leaf images of ten distinct medicinal plants from various regions in Dhaka, Bangladesh. The ten distinct categories are Phyllanthus emblica, Terminalia arjuna, Kalanchoe pinnata, Centella asiatica, Justicia adhatoda, Mikania micrantha, Azadirachta indica, Hibiscus rosa-sinensis, Ocimum tenuiflorum, and Calotropis gigantea. The dataset contains a total of 2,029 original leaf images, along with an additional 38,606 augmented images. Each original image was meticulously captured under natural lighting conditions with an appropriate background. Experts provided accurate labeling for each image, ensuring its seamless integration into various machine learning (ML) and deep learning (DL) models. This comprehensive dataset holds immense potential for researchers in utilizing various ML and DL methods to make significant advancements in the healthcare and pharmaceutical sectors. It serves as a valuable resource for future investigations, laying the foundation for crucial developments in these domains.

SUBMITTER: Islam S

PROVIDER: S-EPMC10450835 | biostudies-literature | 2023 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

BDMediLeaves: A leaf images dataset for Bangladeshi medicinal plants identification.

Islam Saiful S Ahmed Md Rayhan MR Islam Siful S Rishad Md Mahfuzul Alam MMA Ahmed Sayem S Utshow Toyabur Rahman TR Siam Minhajul Islam MI

Data in brief 20230811

This paper introduces a newly curated dataset named "BDMediLeaves" that includes a diverse collection of leaf images of ten distinct medicinal plants from various regions in Dhaka, Bangladesh. The ten distinct categories are Phyllanthus emblica, Terminalia arjuna, Kalanchoe pinnata, Centella asiatica, Justicia adhatoda, Mikania micrantha, Azadirachta indica, Hibiscus rosa-sinensis, Ocimum tenuiflorum, and Calotropis gigantea. The dataset contains a total of 2,029 original leaf images, along with ...[more]

PMID: 37636130

Similar Datasets

Project description:Medicinal plants have been used to treat diseases since ancient times. Plants used as raw materials for herbal medicine are known as medicinal plants [2]. The U. S. Forest Service estimates that 40% of pharmaceutical drugs in the Western world are derived from plants [1]. Seven thousand medical compounds are derived from plants in the modern pharmacopeia. Herbal medicine combines traditional empirical knowledge with modern science [2]. A medicinal plant is considered an important source of prevention against various diseases [2]. The essential medicine component is extracted from different parts of the plants [8]. In underdeveloped countries, people use medicinal plants as a substitute for medicine. There are various species of plants in the world. Herbs are one of them, which are of different shapes, colors, and leaves [5]. It is difficult for ordinary people to recognize these species of herbs. People use more than 50000 plants in the world for medicinal purposes. There are 8000 medicinal plants in India with evidence of medicinal properties [7]. Automatic classification of these plant species is important because it requires intensive domain knowledge to manually classify the proper species. Machine learning techniques are extensively used in classifying medicinal plant species from photographs, which is challenging but intriguing to academics. Artificial Neural Network classifiers' effective performance depends on the quality of the image dataset [4]. This article represents a medicinal plant dataset: an image dataset of ten different Bangladeshi plant species. Images of medicinal plant leaves were from various gardens, including the Pharmacy Garden at Khwaja Yunus Ali University and the Khwaja Yunus Ali Medical College & Hospital in Sirajganj, Bangladesh. Images were collected by taking pictures with high-resolution mobile phone cameras. Ten medicinal species, 500 images per species are included in the data set, namely, Nayantara (Catharanthus roseus), Pathor kuchi (Kalanchoe pinnata), Gynura procumbens (Longevity spinach), Bohera (Terminalia bellirica), Haritaki (Terminalia chebula), Thankuni (Centella asiatica), Neem (Azadirachta indica), Tulsi (Ocimum tenniflorum), Lemon grass (Cymbopogon citratus), and Devil backbone (Euphorbia tithymaloides). This dataset will benefit researchers applying machine learning and computer vision algorithms in several ways. For example, training and evaluation of machine learning models with this well-curated high-quality dataset, development of new computer vision algorithms, automatic medicinal plant identification in the field of botany and pharmacology for drug discovery and conservation, and data augmentation. Overall, this medicinal plant image dataset can provide researchers in the field of machine learning and computer vision with a valuable resource to develop and evaluate algorithms for plant phenotyping, disease detection, plant identification, drug development, and other tasks related to medicinal plants.

Project description:BackgroundMedicinal plant trade is important for local livelihoods. However, many medicinal plants are difficult to identify when they are sold as roots, powders or bark. DNA barcoding involves using a short, agreed-upon region of a genome as a unique identifier for species- ideally, as a global standard.Research questionWhat is the functionality, efficacy and accuracy of the use of barcoding for identifying root material, using medicinal plant roots sold by herbalists in Marrakech, Morocco, as a test dataset.MethodologyIn total, 111 root samples were sequenced for four proposed barcode regions rpoC1, psbA-trnH, matK and ITS. Sequences were searched against a tailored reference database of Moroccan medicinal plants and their closest relatives using BLAST and Blastclust, and through inference of RAxML phylograms of the aligned market and reference samples.Principal findingsSequencing success was high for rpoC1, psbA-trnH, and ITS, but low for matK. Searches using rpoC1 alone resulted in a number of ambiguous identifications, indicating insufficient DNA variation for accurate species-level identification. Combining rpoC1, psbA-trnH and ITS allowed the majority of the market samples to be identified to genus level. For a minority of the market samples, the barcoding identification differed significantly from previous hypotheses based on the vernacular names.Conclusions/significanceEndemic plant species are commercialized in Marrakech. Adulteration is common and this may indicate that the products are becoming locally endangered. Nevertheless the majority of the traded roots belong to species that are common and not known to be endangered. A significant conclusion from our results is that unknown samples are more difficult to identify than earlier suggested, especially if the reference sequences were obtained from different populations. A global barcoding database should therefore contain sequences from different populations of the same species to assure the reference sequences characterize the species throughout its distributional range.

Dataset Information

BDMediLeaves: A leaf images dataset for Bangladeshi medicinal plants identification.

Publications

BDMediLeaves: A leaf images dataset for Bangladeshi medicinal plants identification.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets