Ji-Xiang Du

Classification of plant leaf images with complicated background

Classifying plant leaves has so far been an important and difficult task, especially for leaves with complicated background where some interferents and overlapping phenomena may exist. In this paper, an efficient classification framework for leaf images with complicated background is proposed. First, a so-called automatic marker-controlled watershed segmentation method combined with pre-segmentation and morphological operation is introduced to segment leaf images with complicated background based on the prior shape information. Then, seven Hu geometric moments and sixteen Zernike moments are extracted as shape features from segmented binary images after leafstalk removal. In addition, a moving center hypersphere (MCH) classifier which can efficiently compress feature data is designed to address obtained mass high-dimensional shape features. Finally, experimental results on some practical plant leaves show that proposed classification framework works well while classifying leaf images with complicated background. There are twenty classes of practical plant leaves successfully classified and the average correct classification rate is up to 92.6%.

Computer-Aided Plant Species Identification (CAPSI) Based on Leaf Shape Matching Technique

In this paper, an efficient computer-aided plant species identification (CAPSI) approach is proposed, which is based on plant leaf images using a shape matching technique. Firstly, a Douglas - Peucker approximation algorithm is adopted to the original leaf shapes and a new shape representation is used to form the sequence of invariant attributes. Then a modified dynamic programming (MDP) algorithm for shape matching is proposed for the plant leaf recognition. Finally, the superiority of our proposed method over traditional approaches to plant species identification is demonstrated by experiment. The experimental result showed that our proposed algorithm for leaf shape matching is very suitable for the recognition of not only intact but also partial, distorted and overlapped plant leaves due to its robustness.

Leaf recognition based on the combination of wavelet transform and gaussian interpolation

In this paper, a new approach for leaf recognition using the result of segmentation of leafs skeleton based on the combination of wavelet transform (WT) and Gaussian interpolation is proposed. And then the classifiers, a nearest neighbor classifier (1-NN), a K-nearest neighbor classifier (k-NN) and a radial basis probabilistic neural network (RBPNN) are used, based on run-length features (RF) extracted from the skeleton to recognize the leaves. Finally, the effectiveness and efficiency of the proposed method is demonstrated by several experiments. The results show that the skeleton can be successfully and obviously extracted from the whole leaf, and the recognition rates of leaves based on their skeleton can be greatly improved.

Recognition of leaf images based on shape features using a hypersphere classifier

Recognizing plant leaves has so far been an important and difficult task. This paper introduces a method of recognizing leaf images based on shape features using a hypersphere classifier. Firstly, we apply image segmentation to the leaf images. Then we extract eight geometric features including rectangularity, circularity, eccentricity, etc, and seven moment invariants for classification. Finally we propose using a moving center hypersphere classifier to address these shape features. As a result there are more than 20 classes of plant leaves successfully classified. The average correct recognition rate is up to 92.2 percent.

Bark texture feature extraction based on statistical texture analysis

This paper quantitatively describes and discusses the usefulness of texture analysis methods for the recognition of bark. Comparative studies of bark texture feature extraction are performed for the four texture analysis methods such as the gray level run-length method (RLM), co-occurrence matrices method (COMM) and histogram method (HM) as well as auto-correlation method (ACM). Specifically, we use three classifiers of nearest neighbor (l-NN), k-nearest neighbor (k-NN) and moving median centers (MMC) hypersphere classifiers to verify the validity of the extracted bark texture features. To gain good results we added to color information that proved very efficient. Moreover, the experimental results also demonstrate that from the viewpoint of the recognition accuracy and computational complexity, the COMM method is superior to the other three methods.

A novel full structure optimization algorithm for radial basis probabilistic neural networks

Abstract In this paper, a novel full structure optimization algorithm for radial basis probabilistic neural networks (RBPNN) is proposed. Firstly, a minimum volume covering hyperspheres (MVCH) algorithm is proposed to heuristically select the initial hidden layer centers of the RBPNN, and then the recursive orthogonal least square (ROLS) algorithm combined with the particle swarm optimization (PSO) algorithm is adopted to further optimize the initial structure of the RBPNN. Finally, the effectiveness and efficiency of our proposed algorithm are evaluated through a plant species identification task involving 50 plant species.

Recognition of leaf image based on ring projection wavelet fractal feature

Recognizing plant leaves has been an important and difficult task. This paper introduces a method of recognizing leaf images based on Ring Projection Wavelet Fractal Feature. Firstly, we apply pre-processing to leaf images, and extract from leaves around the border area by the white pixels and all pixel black background binary contour map. Secondly, we get one-dimensional feature of leaves by using Ring Projection to reduce the dimension of two-dimensional pattern. Then, the one-dimensional is decomposed with Daubechies discrete wavelet transform to obtain sub pattern. Finally, we seek the fractal dimension of each sub model. Leaf shape features are extracted from pre-processed leaf images, which include fractal dimension of each sub model and seven Hu moment invariants. As a result there are 30 classes of plant leaves successfully classified.

Palmprint recognition using ICA based on winner-take-all network and radial basis probabilistic neural network

This paper proposes a novel method for recognizing palmprint using the winner-take-all (WTA) network based independent component analysis (ICA) algorithm and the radial basis probabilistic neural network (RBPNN) proposed by us. The WTA-ICA algorithm exploits the maximization of the sparse measure criterion as the cost function, and it extracts successfully palmprint features. The classification performance is implemented by the RBPNN. The RBPNN is trained by the orthogonal least square (OLS) algorithm and its structure is optimized by the recursive OLS (ROLS) algorithm. Experimental results show that the RBPNN achieves higher recognition rate and better classification efficiency with other usual classifiers.

A hypersphere method for plant leaves classification

The recognition of the tens of thousands of kinds of plants on Earth is an important and difficult task. If we use a nearest neighbour classifier to solve it, millions of training data are needed to obtain a high correct recognition rate. However, the corresponding classification process is quite time-consuming. We propose a new approach using a moving median centers hypersphere technique to recognize different plant leaves. With this new method, we can successfully decrease the classification time, and reduce the storage size without sacrificing the classification accuracy. Experimental results using real data show that this method is really efficient and effective in classifying different plant leaves.

Bark classification based on textural features using artificial neural networks

In this paper, a new method for bark classification based on textural and fractal dimension features using Artificial Neural Networks is presented. The approach involving the grey level co-occurrence matrices and fractal dimension is used for bark image analysis, which improves the accuracy of bark image classification by combining fractal dimension feature and structural texture features on bark image. Furthermore, we have investigated the relation between Artificial Neural Network (ANN) topologies and bark classification accuracy. Furthermore, the experimental results show the facts that this new approach can automaticly identify the Tplants categories and the classification accuracy of the new method is better than that of the method using the nearest neighbor classifier.