Data from: Beech leaf disease symptom detection using deep learning and computer vision tools
This image dataset consists of symptomatic and non-symptomatic American beech (Fagus grandifolia) leaves collected for the purpose of training a convolutional neural network to distinguish between leaves affected by Beech Leaf Disease (BLD) and those without visible symptoms. Images were captured in Montgomery County and Prince George’s County, Maryland, during May and June of 2025.Two types of image acquisition settings were used to ensure variation in lighting, background, and presentation. Leaves were photographed outdoors on trees. Additional leaves were imaged under natural field conditions and in a controlled laboratory environment. The dataset includes images captured using an iPhone 15 and a Nikon D600, representing a range of resolutions and optical characteristics.The combined dataset provides diverse examples of symptomatic and non-symptomatic leaves across varying environmental and photographic conditions, supporting robust training and evaluation of machine learning models for BLD detection.
Complete Metadata
| @type | dcat:Dataset |
|---|---|
| accessLevel | public |
| accrualPeriodicity | irregular |
| bureauCode |
[
"005:18",
"005:96"
]
|
| contactPoint |
{
"fn": "Waldo, Benjamin, D.",
"hasEmail": "mailto:Benjamin.Waldo@usda.gov"
}
|
| description | <p dir="ltr">This image dataset consists of symptomatic and non-symptomatic American beech (<i>Fagus grandifolia</i>) leaves collected for the purpose of training a convolutional neural network to distinguish between leaves affected by Beech Leaf Disease (BLD) and those without visible symptoms. Images were captured in Montgomery County and Prince George’s County, Maryland, during May and June of 2025.</p><p dir="ltr">Two types of image acquisition settings were used to ensure variation in lighting, background, and presentation. Leaves were photographed outdoors on trees. Additional leaves were imaged under natural field conditions and in a controlled laboratory environment. The dataset includes images captured using an iPhone 15 and a Nikon D600, representing a range of resolutions and optical characteristics.</p><p dir="ltr">The combined dataset provides diverse examples of symptomatic and non-symptomatic leaves across varying environmental and photographic conditions, supporting robust training and evaluation of machine learning models for BLD detection.</p> |
| distribution |
[
{
"@type": "dcat:Distribution",
"title": "BLD_dataset_1-6-26.zip",
"format": "zip",
"mediaType": "application/zip",
"downloadURL": "https://ndownloader.figshare.com/files/60816103"
}
]
|
| identifier | 10.15482/USDA.ADC/30850214.v1 |
| keyword |
[
"Litylenchus crenatae mccannii",
"beech leaf disease",
"computer vision",
"machine learning",
"training data"
]
|
| license | https://creativecommons.org/publicdomain/zero/1.0/ |
| modified | 2026-03-16 |
| programCode |
[
"005:040"
]
|
| publisher |
{
"name": "Agricultural Research Service",
"@type": "org:Organization"
}
|
| temporal | 2025-05-15/2025-06-13 |
| title | Data from: Beech leaf disease symptom detection using deep learning and computer vision tools |
