stakes and improved the data quality. Data source location The images were collected in the city of Ouro Branco, Minas Gerais, Latitude −20.535912, Longitude −43.711031, Brazil. Data accessibility Repository name: Leaf on Stem Image Dataset Beans (LSID-Beans) Data identification number: 10.17632/f42hwwrpgn.2 Direct URL to data: https://data.mendeley.com/datasets/f42hwwrpgn/2 1 Value of the Data • The dataset images are useful for developing deep learning methods for non-destructive leaf dimension estimation. We provide each leaf's known area, perimeter, width, and length, which can be used to train supervised machine learning algorithms. • Methods developed using the dataset can help to moni
Open resource ↗10.17632/f42hwwrpgn.2 · lines:1-50Unverified paper record
Bean leaf image dataset annotated with leaf dimensions, segmentation masks, and camera calibration.
Data in brief · 27 Jan 2025 · 10.1016/j.dib.2025.111328
Abstract
Leaf dimensioning is relevant for analyzing plant responses to several conditions such as soil fertility, availability of light, agricultural pesticide effect, and access to water in the soil or periods of drought. In this paper, we present a dataset composed of 6981 images of 612 common bean leaves ( Phaseolus vulgaris ). We captured the images of each leaf accompanied by a fiducial marker and annotated the known leaf dimensions (area, perimeter, length, and width). We provide annotations concerning image segmentation, known area uniformly distributed over the leaf region, real area of the marker region, marker pose, capture conditions, and camera calibration. This dataset can be useful for developing deep learning algorithms for leaf dimensioning and related problems. Therefore, there is a potential to contribute to computer vision and plant physiology researchers and specialists.
Plant phenotyping relevance
葉面積・周長・長さ・幅の画像ベース計測用データセットを提供し、セグメンテーション、マーカー姿勢、カメラ校正も含むため、植物表現型取得手法の基盤として中心的です。
abstractWe captured the images of each leaf accompanied by a fiducial marker and annotated the known leaf dimensions (area, perimeter, length, and width).
abstractWe provide annotations concerning image segmentation, known area uniformly distributed over the leaf region, real area of the marker region, marker pose, capture conditions, and camera calibration.
abstractThis dataset can be useful for developing deep learning algorithms for leaf dimensioning and related problems.
Code and data availability
The paper is itself a data descriptor for the LSID-Beans bean leaf image dataset (6981 images, 612 leaves, with leaf dimension annotations, segmentation masks, area maps, and camera calibration). The dataset is publicly deposited on Mendeley Data (DOI 10.17632/f42hwwrpgn.2), and the authors' data-processing scripts are
for that split. Section Cross-validation protocol definition details our proposed cross-validation protocol. 4 Experimental Design, Materials and Methods Fig. 3 shows the steps performed to build our dataset. We describe each step in the next sections. The source codes used to process the data are available in this repository: https://github.com/gcg-ufjf/LSID-Beans-Scripts . Fig. 3 Steps of the dataset construction. Fig 3 4.1 Plant cultivation We selected black bean seeds and carried out planting in April 2022. On average, 3 seeds were sown in each pit, made with the aid of a hoe, along 9 rows of 30 plants. The soil used had never been cultivated and had rejects of construction material on t
Open resource ↗github · lines:66-146This is an automatically classified, unverified record. Curator approval is required before any resource enters the Catalog.