Abstract:<ul class="unordered-list"><li>Leaf vein network geometry can predict levels of resource transport, defence, and mechanical support that operate at different spatial scales. However, it is challenging to quantify network architecture across scales, due to the difficulties both in segmenting networks from images, and in extracting multi‐scale statistics from subsequent network graph representations.</li><li>Here we develop deep learning algorithms using convolutional neural networks (CNNs) to automatically segment leaf vein networks. Thirty‐eight CNNs were trained on subsets of manually‐defined ground‐truth regions from >700 leaves representing 50 southeast Asian plant families. Ensembles of 6 independently trained CNNs were used to segment networks from larger leaf regions (~100 mm<sup>2</sup>). Segmented networks were analysed using hierarchical loop decomposition to extract a range of statistics describing scale transitions in vein and areole geometry. </li><li>The CNN approach gave a precision‐recall harmonic mean of 94.5% ± 6%, outperforming other current network extraction methods, and accurately described the widths, angles, and connectivity of veins. Multi‐scale statistics then enabled identification of previously undescribed variation in network architecture across species.</li><li>We provide a LeafVeinCNN software package to enable multi‐scale quantification of leaf vein networks, facilitating comparison across species and exploration of the functional significance of different leaf vein architectures.</li></ul>

Extraction of Leaf Vein in Plant Leaf Image Based on Cellular Neural Network

Extraction of Leaf Vein Features Based on Artificial Neural Network--Studies on the Living Plant Identification I

The Extraction Method of Leaf Image Based on the Active Contours

A Leaf Vein Extraction Method Based On Snakes Technique

Automated and accurate segmentation of leaf venation networks via deep learning

Extracting vein of leaf image based on K-means clustering

Leaf Vein Extraction and Angle Measurement Using Hue Information and Line Detection

Feature Extraction and Automatic Recognition of Plant Leaf Using Artificial Neural Network

Deep Learning for Plant Species Classification Using Leaf Vein Morphometric

Leaf Recognition Based on PCNN.

Feature Extraction of Plant Leaf Using Deep Learning

Plant Leaf Classification and Recognition Algorithm Based on Neural Network Technology

Plant Identification Using New Architecture Convolutional Neural Networks Combine with Replacing the Red of Color Channel Image by Vein Morphology Leaf

Automatic Leaf Extraction from Outdoor Images

Comparative Study on Different PCNN Models in Plant Leaf Classification

Study on Tree Species Identification Based on Convolution Neural Network and Leaf Texture Image

Recognition of Plants by Leaves Digital Image and Neural Network

Plant Leaf Recognition Based On Small Datasets Using Deep Learning Algorithm

Plant Leaf Identification Via A Growing Convolution Neural Network With Progressive Sample Learning

Research on Feature Extraction about Soybean Leaf

A Leaf Recognition Algorithm for Plant Classification Using Probabilistic Neural Network