Subject Computer Science, Botany.
Specific subject area Computer vision, Image classification, Image processing, Machine learning.
Type of data Plant and leaf Images.
How the data were acquired Images were captured using a 64 megapixel smartphone camera on the Mi 11 Lite 5G Ne .
Number of Plant Species 20.
Total Images 1072.
Data format PNG.
Data Resolution 4640*3472.
Description of data collection Collected the images directly using a smartphone with different configurations as described in the “How the data were acquired” row. This dataset has 1072 images from twenty classes.
Data source location Rural areas of Tangail and Faridpur, Bangladesh.

#### Value of the dataset

This dataset includes the identification of the following Bangladeshi plant species, namely,
'01_Aegle marmelos', '02_Aquilaria malaccensis', '03_Artocarpus heterophyllus', '04_Azadirachta indica', '05_Cestrum nocturnum', '06_Citrus grandis', '07_Coccinia grandis', '08_Codiaeum variegatum', '09_Delbergia sissoo', '10_Epipremnum aureum', '11_Ficus benjamina', '12_Gmelina arborea', '13_Hibiscus rosa sinensis', '14_Lablab purpureus', '15_Monoon longifolium', '16_Phoenix dactylifera', '17_Pithecellobium dulce', '18_Psidium guajava', '19_Rosa canina', '20_Ziziphus mauritiana'], index = ['01_Aegle marmelos', '02_Aquilaria malaccensis', '03_Artocarpus heterophyllus', '04_Azadirachta indica', '05_Cestrum nocturnum', '06_Citrus grandis', '07_Coccinia grandis', '08_Codiaeum variegatum', '09_Delbergia sissoo', '10_Epipremnum aureum', '11_Ficus benjamina', '12_Gmelina arborea', '13_Hibiscus rosa sinensis', '14_Lablab purpureus', '15_Monoon longifolium', '16_Phoenix dactylifera', '17_Pithecellobium dulce', '18_Psidium guajava', '19_Rosa canina', '20_Ziziphus mauritiana'.

• The information gathered is of high quality and valuable, intending to serve as content for data analysis.

• The dataset may prove helpful in testing image recognition classifiers for the identification of various medicinal plants.

Using the dataset's images of medicinal plant leaves, classification algorithms can be trained, tested, and validated.

• The data can be used for a variety of machine-learning applications, including image classification and image detection.

This dataset is beneficial for developing methods for plant identification, plant classification, plant growth monitoring, leave disease diagnosis, etc.

Crop diseases are a major threat to food security, but their rapid identification remains difficult in many parts of the world due to the lack of the necessary infrastructure. The combination of increasing global smartphone penetration and recent advances in computer vision made possible by deep learning has paved the way for smartphone-assisted disease diagnosis.

Purpose of dataset: Using Deep Learning for Image-Based Plant Disease Detection

How to approach: Deep convolutional neural networks (CNN) for the classification problem

THIS RESOURCE IS NO LONGER IN SERVICE. Documented on August 26, 2019.

Database of leaf senescence to collect SAGs, mutants, phenotypes and literature references. Leaf senescence has been recognized as the last phase of plant development, a highly ordered process regulated by genes called SAGs. By integrating the data from mutant studies and transgenic analysis, they collected many SAGs related to regulation of the leaf senescence in various species. Additionally, they have categorized SAGs according to their functions in regulation of leaf senescence and used standard criteria to describe senescence associated phenotypes for mutants. Users are welcome to submit the new SAGs.

Dataset

This dataset was created by majorproject24

Released under CC0: Public Domain

Contents

This data set provides carbon (C), nitrogen (N), phosphorus (P), potassium (K), calcium (Ca), and magnesium (Mg) concentrations in green and senesced leaves. Vegetation characteristics reported include species growth habit, leaf area, mass, and mass loss with senescence. The data were compiled from 86 selected studies in 31 countries, and resulted in approximately 1,000 data points for both green and senesced leaves from woody and non-woody vegetation as described in Vergutz et al (2012). The studies were conducted from 1970-2009. There are two comma-delimited data files with this data set.

Computer vision can predominantly be focused to design the strategies for the conservation of the plants. Previous decade’s trends and the current prevailing incidents with respect to global warming, forest fires, and famines act as potential indicators of how much nature is destroyed by human activities. Plants are vitally used in foodstuff, medicine, industry and as well for environmental protection. However, due to lack of resources and knowledge, it is difficult to recognize different plant species, plant diseases, etc. Nowadays modern equipment’s are being designed to address these issues. So considering the challenges, demands, we have constructed a database of different plants. The plants taken for study are the native plants of the Kashmir region of India. The climate of Kashmir remains chilling for a few months and pleasant for the rest of the year. Eight different plants namely Apple, Apricot, Cherry, Cranberry, Grapes, Peach, Pear, and Walnut are selected for the study based on their commercial and medicinal usage. The leaf is the primary object of reference taken for making the database, as they grow much earlier than fruits as well as the other plant parts. For each plant two types of leaves are selected, one healthy and the other diseased. Considering the natural conditions under which the farmers or the agriculturists have to work, the images are captured in broad daylight under the auto mode with the Nikon D-SLR digital camera with an ISO Speed = 100, Aperture = F/5.6, Flash = Not Fired, Shutter Speed = 1/640. All the images are captured by an 18-55 mm lens and are in .JPG format. The leaves are divided into two major classes A and B respectively. The two major classes were then divided into 16 sub classes i.e., eight healthy and eight diseased. The symbol “h” e.g., plant-name_h001 in the images represent healthy images and “d” i.e., plant-name_d001 represents the diseased images. The images are labeled, resized and classified into different classes. The class of healthy images comprises of a total of 1201 images and the diseased images constitute of a total of 935 images. Thus a total of 2136 images were selected from the captured images to sew up this database. Every little step towards a positive perspective marks the beginning of the era of growth with kindness.

This data set provides global leaf area index (LAI) values for woody species. The data are a compilation of field-observed data from 1,216 locations obtained from 554 literature sources published between 1932 and 2011. Only site-specific maximum LAI values were included from the sources; values affected by significant artificial treatments (e.g. continuous fertilization and/or irrigation) and LAI values that were low due to drought or disturbance (e.g. intensive thinning, wildfire, or disease), or because vegetation was immature or old/declining, were excluded (Lio et al., 2014). To maximize the generic applicability of the data, original LAI values from source literature and values standardized using the definition of half of total surface area (HSA) are included. Supporting information, such as geographical coordinates of plot, altitude, stand age, name of dominant species, plant functional types, and climate data are also provided in the data file. There is one data file in comma-separated (.csv) format with this data set and one companion file which provides the data sources.

This dataset contains images of five different types of leaves: Mango, Guava, Jackfruit, Neem, and Banana. Each category includes 500 high-quality images, totaling 2,500 images. The dataset is designed for machine learning tasks such as image classification, plant species identification, and computer vision projects related to agriculture or botany. All images are organized into folders by leaf type, making it easy to use for training and evaluation in supervised learning models.

Poster at the "The Society for Mathematical Biology Annual Meeting and Conference", Knoxville,TN (USA), July 25-28, 2012

Manually collected image dataset of sugarcane leaf disease. It has mainly five categories in it. Healthy, Mosaic, Redrot, Rust and Yellow disease. The dataset has been captured with smart phones of various configuration to maintain the diversity. It contains total 2569 images including all categories. This database has been collected in Maharashtra, India. The database is balanced and contains good variety. The image sizes are not constant as it originates form various capturing devices. All images are in RGB format.

There are two datasets and one table uploaded in this platform under the title "MED117_Medicinal Plant Leaf Dataset & Name Table". A folder is created with title "MED 117 Leaf Species". Inside this two sub folders with titles " Raw leaf image set of medicinal plants_v2" and "Segmented leaf set using UNET segmentation" are created. Raw leaf image set consists of leaf images of 117 medicinal plants found in Assam. All the samples are collected by visiting different (Govt, Public and Private) medicinal gardens situated in different places of Assam and some other general places where they are mostly found. Videos of 10 to 15 seconds duration were taken for two to three leaves of every species on a white background and video recording was done using a SLR Canon Camera. Individual videos were segregated into image frames and thus were able to get around 77,700 jpg image frames from the videos. The Raw leaf image set consists of folders with scientific name and common name within bracket. Second folder with title "Segmented leaf set using UNET segmentation" consists of 115 medicinal plant species with their segmented leaf image samples using UNET segmentation technique. Here two species are excluded from the original dataset due to small unpredictable size of the samples, so total 115 subfolders inside the segmented folder is achieved. Thirdly a table in doc format with title "Medicinal Plant Name Table" is uploaded and it includes Scientific name, Common name and Assamese name of the plants listed in the folders in the same sequence. The whole contribution is absolutely original and new, collected from different sources then processed for segmentation and prepared the table by discussing with taxonomy experts from Botany department of Gauhati University, Guwahati, Assam. India.

This dataset contains a curated subset of plant leaf images labeled as either healthy or diseased, intended for binary image classification tasks in plant pathology and agricultural health monitoring.The dataset is derived from the original publicly available dataset titled "A Database of Leaf Images: Practice towards Plant Conservation with Plant Pathology", published by Chouhan et al. on Mendeley Data in 2019 under a Creative Commons Attribution 4.0 (CC BY 4.0) license. The original dataset is available here.For this version, images of healthy and diseased leaves were extracted from the original dataset and organized into two separate folders: healthy/ and diseased/, to support binary classification tasks in machine learning workflows. No additional augmentation or transformation has been applied to the images in this upload.

The Global Spectra-Trait Initiative (GSTI) aims to generate generalizable spectra trait models using reflectance data to predict leaf traits associated with the photosynthesis capacity of leaves. It comprises a synthesized dataset of leaf trait data, input datasets and code. Leaf traits include the maximum carboxylation rate of rubisco (Vcmax), the maximum electron transport rate (Jmax), the dark respiration, as well as the prediction of leaf nitrogen, leaf mass per area (LMA), and leaf water content (LWC). The dataset comprises >7500 paired observations from around 400 species from a broad range of biomes. This dataset comprises a zip file of the GSTI GitHub repository (https://github.com/plantphys/gsti), the synthesized database (.csv) and database metadata files.

1) Data Introduction • The Potato Disease Leaf Dataset(PLD) is a computer vision dataset designed for multi-class image classification to diagnose diseases on potato leaves. It consists of three classes: healthy, early blight, and late blight.

2) Data Utilization (1) Characteristics of the Potato Disease Leaf Dataset(PLD): • The dataset consists of potato leaf images collected from real agricultural environments, enabling the development of practical models that take into account regional disease variations and environmental factors.

(2) Applications of the Potato Disease Leaf Dataset(PLD): • Development of disease classification models: This dataset can be used to train deep learning classifiers that automatically distinguish between early blight and late blight on potato leaves. • Research on agricultural diagnostics and smart farming systems: The dataset can support the development of smart agriculture solutions, including early disease detection systems, disease spread prediction tools, and automated alert systems aimed at improving crop yield.

The LT-Brazil data set contains observations of leaf mass per area, leaf N and P concentration per unit mass, and leaf N:P ratio from native woody species across the Brazilian territory, encompassing information of biome, vegetation, taxonomic data, geographical coordinates, climatic parameters, as well as soil properties. We compiled data from several geographical coordinates in native vegetation distributed across all biomes (i.e., Amazônia, Caatinga, Cerrado, Mata Atlântica, Pampa, and Pantanal) found in Brazil. Our compilation was focused on native woody plants (i.e., trees, shrubs, subshrubs, and lianas), excluding monocots, palm trees, herbs, and hemiparasitic plants. The compiled data set covers c. 9% of woody angiosperm species of Brazil. Unidentified or mixed species were also considered when met our eligibility criteria. Contributions to expand this database can be performed through our repository at GitHub (https://github.com/emariano-git/lt-brazil.git). Major versions of the LT-Brazil data set will also be made available via the TRY Plant Trait Database (https://www.try-db.org).

This global data set of photosynthetic rates and leaf nutrient traits was compiled from a comprehensive literature review. It includes estimates of Vcmax (maximum rate of carboxylation), Jmax (maximum rate of electron transport), leaf nitrogen content (N), leaf phosphorus content (P), and specific leaf area (SLA) data from both experimental and ambient field conditions, for a total of 325 species and treatment combinations. Both the original published Vcmax and Jmax values as well as estimates at standard temperature are reported. The maximum rate of carboxylation (Vcmax) and the maximum rate of electron transport (Jmax) are primary determinants of photosynthetic rates in plants, and modeled carbon fluxes are highly sensitive to these parameters. Previous studies have shown that Vcmax and Jmax correlate with leaf nitrogen across species and regions, and locally across species with leaf phosphorus and specific leaf area, yet no universal relationship suitable for global-scale models is currently available. These data are suitable for exploring the general relationships of Vcmax and Jmax with each other and with leaf N, P and SLA. This data set contains one *.csv file.

1) Data Introduction • The MangoLeaf Dataset is a computer vision dataset consisting of mango leaf images, categorized into eight classes: seven disease types and one healthy class.

2) Data Utilization (1) Characteristics of the MangoLeaf Dataset: • The images were collected from real-world mango orchards rather than controlled lab environments, making this dataset highly suitable for training models with real-world applicability.

(2) Applications of the MangoLeaf Dataset: • Development of mango leaf disease classification models: This dataset can be used to train deep learning models for both multi-class classification (seven diseases + healthy) and binary classification (healthy vs. diseased). • Development of AI solutions for agriculture: It serves as foundational data for building agricultural AI applications such as disease diagnosis apps, mango cultivation monitoring systems, and smart farming alert platforms.

For the latest version of the LT-Brazil data set, please visit our permanent repository at Zenodo (https://doi.org/10.5281/zenodo.4574445).

This dataset was created for my Master's thesis "Data Augmentation for Plant Species Recognition". The original images were captured by me and my sister, in a local park within a 1 km2 area. They were taken with different levels of light, at different orientations, zoom levels and with different backgrounds. They were then compressed to a smaller size and run through an algorithm that uses image processing methods to create 1023 new images from the original one. The original version of the dataset can be found here. The following image processing methods were used, - Rotate and flip - Adding noise - Changing brightness and contrast - Warping image and perspective

Data for publication in Data in Brief.