Plant identification using leaf shapes—A pattern counting approach

Zhao, Cong; Chan, Sharon S.; Cham, Wai-Kuen; Chu, L.M.

doi:10.1016/j.patcog.2015.04.004

Cited by 121 publications

(79 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Hall et al [8] proposed using Hand-Crafted Shape (HCS) and Histogram of Curvature over Scale (HoCS) [7] to analyse leaves. Zhao et al [22] proposed a new counting-based shape descriptor, namely independent-IDSC(I-IDSC) features, to recognize simple and compound leaves. Apart from studying the whole shape contour of the leaf, some studies [9,23] analysed leaf margins for species classification.…”

Section: Related Studiesmentioning

confidence: 99%

How deep learning extracts and learns leaf features for plant classification

Lee

Chan

Mayo

et al. 2017

Pattern Recognition

460

172

View full text Add to dashboard Cite

Highlights• A deep learning approach to quantify discriminatory leaf is proposed.• Shape is not a dominant feature for leaf but rather the different orders of venation.• Deep learning reveals transformation of leaf features from general to specific types.• Findings archived fit with the hierarchical botanical definitions of leaf characters• Features learned using deep learning can improve plant recognition performance. AbstractPlant identification systems developed by computer vision researchers have helped botanists to recognize and identify unknown plant species more rapidly. Hitherto, numerous studies have focused on procedures or algorithms that maximize the use of leaf databases for plant predictive modeling, but this results in leaf features which are liable to change with different leaf data and feature extraction techniques. In this paper, we learn useful leaf features directly from the raw representations of input data using Convolutional Neural Networks (CNN), and gain intuition of the chosen features based on a Deconvolutional Network (DN) approach. We report somewhat unexpected results: (1) different orders of venation are the best representative features compared to those of outline shape, and (2) we observe multi-level representation in leaf data, demonstrating the hierarchical transformation of features from lower-level to higher-level abstraction, corresponding to species classes. We show that these findings fit with the hierarchical botanical definitions of leaf characters. Through these findings, we gained insights into the design of new hybrid feature extraction models which are able to further improve the discriminative power of plant classification systems. The source code and models are available at: https://github.com/cs-chan/Deep-Plant.

show abstract

Section: Related Studiesmentioning

confidence: 99%

How deep learning extracts and learns leaf features for plant classification

Lee

Chan

Mayo

et al. 2017

Pattern Recognition

460

172

View full text Add to dashboard Cite

show abstract

“…Leaf contours contain rich discriminative information about leaf margins, apexes and bases [8] which are usually taken as the main cues for species recognition. Thus, contour based descriptors have been developed for plant species recognition.…”

Section: Related Workmentioning

confidence: 99%

“…Recently, Zhao et al [8] proposed a pattern counting method for the classification of both simple leaves and compound leaves. A novel feature, independent-IDSC, is developed for better capturing global and local information of a leaf, such as the overall shape, margin type of a simple leaf, and leaflet details of a compound leaf.…”

Section: Related Workmentioning

confidence: 99%

From species to cultivar: Soybean cultivar recognition using joint leaf image patterns by multiscale sliding chord matching

Wang

Gao

Yuan

et al. 2020

Biosystems Engineering

View full text Add to dashboard Cite

“…It is noteworthy that our goal is to solve the leaf disease identification problem, which is different from the leaf species identification problem [19]. It is necessary to model the leaf shapes for leaf species identification [19,20]. However, for the leaf disease identification, it is crucial to find the location of the leaf disease to extract discriminative features from background and spot areas separately.…”

Section: Related Workmentioning

confidence: 99%

Apple Leaf Disease Identification Through Region-of-Interest-Aware Deep Convolutional Neural Network

Son

Lee³

2020

jist

View full text Add to dashboard Cite

A new method of recognizing apple leaf diseases through region-ofinterest-aware deep convolutional neural network is proposed in this paper. The primary idea is that leaf disease symptoms appear in the leaf area whereas the background region contains no useful information regarding leaf diseases. To realize this idea, two subnetworks are first designed. One is for the division of the input image into three areas: background, leaf area, and spot area indicating the leaf diseases, which is the region of interest (ROI), and the other is for the classification of leaf diseases. The two subnetworks exhibit the architecture types of an encoder-decoder network and VGG network, respectively; subsequently, they are trained separately through transfer learning with a new training set containing class information, according to the types of leaf diseases and the ground truth images where the background, leaf area, and spot area are separated. Next, to connect these subnetworks and subsequently train the connected whole network in an end-to-end manner, the predicted ROI feature map is stacked on the top of the input image through a fusion layer, and subsequently fed into the subnetwork used for the leaf disease identification. The experimental results indicate that correct recognition accuracy can be increased using the predicted ROI feature map. It is also shown that the proposed method obtains better performance than the conventional state-of-the-art methods: transfer-learning-based methods, bilinear model, and multiscale-based deep feature extraction and pooling approach.

show abstract

Plant identification using leaf shapes—A pattern counting approach

Cited by 121 publications

References 11 publications

How deep learning extracts and learns leaf features for plant classification

How deep learning extracts and learns leaf features for plant classification

From species to cultivar: Soybean cultivar recognition using joint leaf image patterns by multiscale sliding chord matching

Apple Leaf Disease Identification Through Region-of-Interest-Aware Deep Convolutional Neural Network

Contact Info

Product

Resources

About