.png){kind=icon}
返回
Plant leaf identification based on shape and convolutional features
.png){kind=icon}
PRE
AI
DOI：10.1016/j.eswa.2023.119626.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Plant leaf identification is an important and challenging issue in botany and image analysis. This is because leaves of the same class exhibit very large differences, and leaves of different classes have very small changes. In this article, we propose a novel plant identification method by combining shape and convolutional characteristics. First, we present an effective shape descriptor named improved multiscale triangle descriptor (IMTD) to capture the shape properties of a leaf. Then, we analyze different levels of convolutional features for plant leaf identification. Finally, we combine complementary shape and convolutional features for the task of plant leaf recognition. The proposed method has been tested on nine benchmark leaf datasets, including the Flavia, CVIP100, Swedish, MEW2012, ICL Compound, LeafSnap, Mulberry, Corn leaf disease, and Turkey-Plant dataset. Our method achieves good recognition results on nine benchmark leaf datasets for the unsupervised leaf image retrieval and supervised leaf image classification. The recognition performance of our approach is better than prior state-of-the-art plant identification approaches.
Keyword:
Plant leaf recognition
Improved multiscale triangle descriptor
Shape feature
Convolutional feature
期刊
IF:
7.5
论文数:
3.0W
被引数:
10.2W
机构
引用论文
分享收藏Multi-Organ Plant Classification Based on Convolutional and Recurrent Neural Networks基于卷积和循环神经网络的多器官植物分类
OAAI
OAAI分享收藏Plant Leaves Classification: A Few-Shot Learning Method Based on Siamese Network.png)
IEEE ACCESSOAAI
IEEE ACCESSIF3.6
OAAI分享收藏Multiple attentional pyramid networks for Chinese herbal recognition用于中药识别的多注意金字塔网络PATTERN RECOGNITIONOAAI
PATTERN RECOGNITIONIF7.6
OAAI分享收藏
查看更多内容