How experts of plants recognize plant growth?

Growing plants is challenging. Imagine a future where one snapshot of a plant with your smartphone reveals its growth rate and environmental factors instantly.

I've studying prediction of the growth rate of plants (such as trees and duckweed) from images of their leaves using AI technology. Plant leaves have distinct visual characteristics. Each plant's leaves have distinct visual traits—some shared across many species, others unique to specific ones. For instance, duckweed grows in clusters called colonies, whereas some plants extend their leaves individually. These nuanced differences are typically known only to plant experts. By analyzing these visual traits, my goal is to accurately assess growth rates and environmental factors from just one image.

Keywords

Plant Leaves
Image Recognition
Deep Learning
Computer Vision

Which colony do you think will increase the most?

Plant experts focus on specific features within a colony: leaf area, color, and roundness. Based on these features, they estimate that the colony ( ) will grow the most, while the colony ( ) will grow the second most.

This estimation is challenging for non-experts. The automation of this estimation will encourage non-experts plant cultivation.

Feature Fusion for Leaf Image Classification

we tackle the problem of classifying tree species from leaf images with deep learning. First, we classify leaf images with deep learning, focusing on a single leaf feature: only the whole-leaf feature, only the leaf-shape feature, or only the leaf-vein feature. The leaf feature that contributes to the classification accuracy differs depending on the species. Second, we classify leaf images by combining the whole-leaf, the leaf-shape, and the leaf-vein feature. We define combining these features as feature fusion. After feature fusion, the classification performance improves to 92.07%. It is important for people to teach which leaf features to focus on.

Moeri Okuda, and Hiroaki Ohshima: “Feature Fusion for Leaf Image Classification”, In Proceedings of the 2022 IEEE International Conference on Big Data and Smart Computing（BigComp 2022） , January, 2022.

Estimation of Surface Area and Number of Leaves on Duckweed Focused on Leaf-Shape and Leaf-Color Features

we propose a method to estimate the frond area and frond number of a floating macrophyte, duckweed (Lemna minor) using photographs as input. Duckweed is a fast-growing plant that holds potential for water purification and biomass production . Accurate recognition of plant growth is essential for its effective utilization. The frond area and the frond number of duckweed are used as indicators of duckweed growth. The proposed method consists of three parts: (1) colony segmentation, (2) colony health estimation, and (3) leaf number estimation within the colony. To facilitate training and testing, we cultivated duckweed and created a dataset through photo collection and annotation.

Moeri Okuda, Hidehiro Ishizawa, and Hiroaki Ohshima: “Estimation of Surface Area and Number of Leaves on Duckweed Focused on Leaf-Shape and Leaf-Color Features”, IECE Transactions D, Jan, 2024.