A Simple and Low-Cost Method for Leaf Surface Dimension Estimation Based on Digital Images

• Karla Gabriele Florentino da Silva UFJF / IFMG
• Jonas Magalhães Moreira IFMG
• Gabriel Barreto Calixto UFJF
• Luiz Maurílio da Silva Maciel UFJF / IFMG
• Márcio Assis Miranda IFMG
• Leandro Elias Morais IFMG

Resumo

The leaf is the organ of the plant body that performs photosynthesis and its area is one of the morphological parameters that most respond to droughts, climate changes, and attack of pathogens, associated with the accumulation of biomass and agricultural productivity. In addition, leaf area and other surface data (for example, width and length) are widely used in studies of plant anatomy and physiology. The methods of measuring these leaf surface parameters are often complicated and costly. In this context, this work aims to develop a simple and low-cost method capable of accurately measuring the leaf surface size of plant species with significant agricultural interest. Our method extract the information through images of leaves accompanied by a scale pattern whose real area is known, captured by a simple camera. To evaluate our method, we performed experiments with images of 118 leaves of 6 species. We compared the results to the ImageJ software, which is widely used to estimate leaf dimensions from images. The results showed our method present performance similar to ImageJ. However, unlike ImageJ, our method does not require user interaction during the dimensions estimation.