Leaf Analyzer: A fully automated and open-source tool for high-throughput leaf trait measurement

What is it about?

Leaf Analyzer is an open-source, fully automated computer vision tool for high-throughput measurement of leaf morphological and phenotypic traits, including leaf area, dimensions, perimeter, count, Green Leaf Index (GLI), and percentage damage. It is designed to make leaf analysis simple and accessible: users place leaves or plants on a specially designed reference pattern, capture an image, and Leaf Analyzer automatically segments the leaves and extracts quantitative traits. The workflow requires no manual outlining of leaves and no adjustment of segmentation parameters during runtime. Unlike many image-analysis approaches that rely on strong foreground–background contrast or tightly controlled imaging conditions, Leaf Analyzer uses an unsupervised approach together with a novel Leaf Background Separation (LBS) feature designed to accommodate variations in leaf colour, illumination, shadows, and background conditions. It supports both individual and batch image processing and can be applied to a wide range of applications, including destructive and non-destructive leaf measurements, seed germination monitoring, grain counting, root phenotyping, and other image-based phenotyping experiments. This makes Leaf Analyzer a versatile tool for researchers working across plant science, agriculture, and environmental research. Key features Fully automated workflow – Users simply select an image or a folder containing images, choose the desired traits, and click **Run**. Leaf Analyzer automatically processes the images and exports the measurements to a spreadsheet, with no parameter tuning required during runtime. Automatic leaf segmentation without training – The proposed approach eliminates the need for training datasets and provides robust automatic segmentation across leaves of different shapes and colours, except white leaves, even under challenging lighting conditions. Support for destructive and non-destructive measurements – Leaf Analyzer supports both destructive sampling, where leaves are removed for analysis, and non-destructive measurements, where leaves remain attached to the plant, providing flexibility for different experimental designs. Indoor and outdoor compatibility – Leaf Analyzer can be used in both controlled indoor environments and outdoor field conditions, making it suitable for a wide range of research applications. High-throughput processing – Large image datasets can be batch processed directly from a folder, with measurements from all images automatically saved to a single spreadsheet file. Comprehensive trait analysis – Leaf Analyzer measures leaf area, perimeter, length, width, leaf count, Green Leaf Index, percentage damage, and other derived traits. Strong performance compared with existing tools – In comparative evaluations with popular tools such as Petiole Pro and LeafByte, Leaf Analyzer achieved substantially lower measurement error. Cross-platform availability – Leaf Analyzer is available for Windows, Linux, macOS on Intel processors, and macOS on Apple silicon. You can download Leaf Analyzer from GitHub: https://github.com/squashking/Leaf-Analyzer

Featured Image

Photo by Chris Lawton on Unsplash

Photo by Chris Lawton on Unsplash

Why is it important?

Accurate and efficient measurement of leaf traits is essential for understanding plant growth, health, stress responses, and performance under different environmental and experimental conditions. However, conventional leaf measurement methods can be time-consuming, labour-intensive, destructive, or dependent on specialised imaging setups and manual parameter adjustment. Leaf Analyzer helps address these challenges by providing a fast, automated, and accessible way to extract quantitative leaf traits from ordinary digital images. By reducing manual effort and supporting high-throughput and non-destructive measurements, it enables researchers to analyse larger datasets, monitor plants repeatedly over time, and obtain more consistent and reproducible measurements, ultimately supporting more efficient research in plant science, agriculture, breeding, and environmental studies.

Perspectives