Plant Species Classification Based on Hyperspectral Imaging via a Lightweight Convolutional Neural Network Model

Abstract

In recent years, many image-based approaches have been proposed to classify plant species. Most methods utilized red green blue (RGB) imaging materials and designed custom features to classify the plant images using machine learning algorithms. Those works primarily focused on analyzing single-leaf images instead of live-crown images. Without considering the additional features of the leaves’ color and spatial pattern, they failed to handle cases that contained leaves similar in appearance due to the limited spectral information of RGB imaging. To tackle this dilemma, this study proposes a novel framework that combines hyperspectral imaging (HSI) and deep learning techniques for plant image classification. We built a plant image dataset containing 1,500 images of 30 different plant species taken by a 470–900 nm hyperspectral camera and designed a lightweight conventional neural network (CNN) model (LtCNN) to perform image classification. Several state-of-art CNN classifiers are chosen for comparison. The impact of using different band combinations as the network input is also investigated. Results show that using simulated RGB images achieves a kappa coefficient of nearly 0.90 while using the combination of 3-band RGB and 3-band near-infrared images can improve to 0.95. It is also found that the proposed LtCNN can obtain a satisfactory performance of plant classification (kappa = 0.95) using critical spectral features of the green edge (591 nm), red-edge (682 nm), and near-infrared (762 nm) bands. This study also demonstrates the excellent adaptability of the LtCNN model in recognizing leaf features of plant live-crown images while using a relatively smaller number of training samples than complex CNN models such as AlexNet, GoogLeNet, and VGGNet.

Leaf-based species classification of hybrid cherry tomato plants by using hyperspectral imaging

“…13 Previous studies showed that leaf-based HSI technology exhibited excellent performance in the classification of plant species. 14,15 But most of these works used the whole leaf as the region of interest (ROI) to extract reflectance spectra for the classification models. The different effects of the mesophyll region (MR) and vein region (VR) of leaves are not distinguished in modeling although differences occur between the optical behaviors of the MR and VR.…”

Transformer attention network and unmanned aerial vehicle hyperspectral remote sensing for grassland rodent pest monitoring research

“…Convolutional neural network (CNN), as one of the representative algorithms of deep learning, is widely used in HSI. 52 However, because CNN uses a fixedsize convolutional kernel to extract features from its images, it is limited by the acceptance field of view. Therefore, CNN can only extract local features and cannot focus on global features, making it difficult to break the bottleneck in classification performance.…”

“…Also, deep learning has a large number of learnable parameters, which makes the classification performance better and more robust. Convolutional neural network (CNN), as one of the representative algorithms of deep learning, is widely used in HSI 52 . However, because CNN uses a fixed-size convolutional kernel to extract features from its images, it is limited by the acceptance field of view.…”

Classification of Degraded Species in Desert Grasslands Based on Multi-Feature Fusion and Unmanned Aerial Vehicle Hyperspectral

“…Currently, a convolutional neural network (CNN) is a widely used hyperspectral image classification method in hyperspectral image classification (Liu et al, 2022;Xu et al, 2021). CNN is divided into a 1D convolutional neural network (1DCNN), a 2D convolutional neural network (2DCNN), and a 3D convolutional neural network (3DCNN).…”