Ying-Ke Lei

Discriminant sparse neighborhood preserving embedding for face recognition

Sparse subspace learning has drawn more and more attentions recently. However, most of the sparse subspace learning methods are unsupervised and unsuitable for classification tasks. In this paper, a new sparse subspace learning algorithm called discriminant sparse neighborhood preserving embedding (DSNPE) is proposed by adding the discriminant information into sparse neighborhood preserving embedding (SNPE). DSNPE not only preserves the sparse reconstructive relationship of SNPE, but also sufficiently utilizes the global discriminant structures from the following two aspects: (1) maximum margin criterion (MMC) is added into the objective function of DSNPE; (2) only the training samples with the same label as the current sample are used to compute the sparse reconstructive relationship. Extensive experiments on three face image datasets (Yale, Extended Yale B and AR) demonstrate the effectiveness of the proposed DSNPE method.

Face recognition via Weighted Sparse Representation

Face recognition using Sparse Representation based Classification (SRC) is a new hot technique in recent years. SRC can be regarded as a generalization of Nearest Neighbor and Nearest Feature Subspace. This paper first reviews the Nearest Feature Classifiers (NFCs), including Nearest Neighbor (NN), Nearest Feature Line (NFL), Nearest Feature Plane (NFP) and Nearest Feature Subspace (NFS), and formulates them as general optimization problems, which provides a new perspective for understanding NFCs and SRC. Then a locality Weighted Sparse Representation based Classification (WSRC) method is proposed. WSRC utilizes both data locality and linearity; it can be regarded as extensions of SRC, but the coding is local. Experimental results on the Extended Yale B, AR databases and several data sets from the UCI repository show that WSRC is more effective than SRC.

Using manifold embedding for assessing and predicting protein interactions from high-throughput experimental data

MOTIVATION High-throughput protein interaction data, with ever-increasing volume, are becoming the foundation of many biological discoveries, and thus high-quality protein-protein interaction (PPI) maps are critical for a deeper understanding of cellular processes. However, the unreliability and paucity of current available PPI data are key obstacles to the subsequent quantitative studies. It is therefore highly desirable to develop an approach to deal with these issues from the computational perspective. Most previous works for assessing and predicting protein interactions either need supporting evidences from multiple information resources or are severely impacted by the sparseness of PPI networks. RESULTS We developed a robust manifold embedding technique for assessing the reliability of interactions and predicting new interactions, which purely utilizes the topological information of PPI networks and can work on a sparse input protein interactome without requiring additional information types. After transforming a given PPI network into a low-dimensional metric space using manifold embedding based on isometric feature mapping (ISOMAP), the problem of assessing and predicting protein interactions is recasted into the form of measuring similarity between points of its metric space. Then a reliability index, a likelihood indicating the interaction of two proteins, is assigned to each protein pair in the PPI networks based on the similarity between the points in the embedded space. Validation of the proposed method is performed with extensive experiments on densely connected and sparse PPI network of yeast, respectively. Results demonstrate that the interactions ranked top by our method have high-functional homogeneity and localization coherence, especially our method is very efficient for large sparse PPI network with which the traditional algorithms fail. Therefore, the proposed algorithm is a much more promising method to detect both false positive and false negative interactions in PPI networks. AVAILABILITY MATLAB code implementing the algorithm is available from the web site http://home.ustc.edu.cn/∼yzh33108/Manifold.htm.

Modified locally linear discriminant embedding for plant leaf recognition

Based on locally linear embedding (LLE) and modified maximizing margin criterion (MMMC), a modified locally linear discriminant embedding (MLLDE) algorithm is proposed for plant leaf recognition in this paper. By MLLDE, the plant leaf images are mapped into a leaf subspace for analysis, which can detect the essential leaf manifold structure. Furthermore, the unwanted variations resulting from changes in period, location, and illumination can be eliminated or reduced. Different from principal component analysis (PCA) and linear discriminant analysis (LDA), which can only deal with flat Euclidean structures of plant leaf space, MLLDE not only inherits the advantages of locally linear embedding (LLE), but makes full use of class information to improve discriminant power by introducing translation and rescaling models. The experimental results on real plant leaf database show that the MLLDE is effective for plant leaf recognition.

Maximum margin criterion with tensor representation

In this paper, we propose tensor based Maximum Margin Criterion algorithm (TMMC) for supervised dimensionality reduction. In TMMC, an image object is encoded as an nth-order tensor, and its 2-D representation is directly treated as matrix. Meanwhile, the k-mode optimization approach is exploited to iteratively learn multiple interrelated discriminative subspaces for dimensionality reduction of the higher order tensor. TMMC generalizes the traditional MMC based on vector data to the one based on matrix and tensor data, which completes the MMC family in terms of data representation. The results of experiments conducted on four databases show that the accurate recognition rate of TMMC is better than that of the method of Concurrent Subspaces Analysis (CSA), and is comparable with the method of Multilinear Discriminant Analysis (MDA). The experimental results also show that the accurate recognition rate of the tensor/matrix-based methods may not always be better than that of vector-based methods. Reasonable discussions about this phenomenon have been given in this paper.

Increasing reliability of protein interactome by fast manifold embedding

Over the last decade, the development of high-throughput techniques has resulted in a rapid accumulation of protein-protein interaction (PPI) data. However, the high-throughput experimental interaction data is prone to exhibit high level of false-positive rates. It is therefore highly desirable to develop an approach to deal with these issues from the computational perspective. In this paper, we develop a robust computational technique for assessing the reliability of interactions by fast manifold embedding algorithm. A fast isometric feature mapping (fast-ISOMAP) is proposed to transform a PPI network into a low dimensional metric space, which recasts the problem of assessing protein interactions into the form of measuring similarity between points of its metric space. Then a reliability index (RI), a likelihood indicating the interaction of two proteins, is assigned to each protein pair in the PPI networks based on the similarity between the points in the embedding space. Validation of the proposed method is performed with extensive experiments on PPI networks of yeast. Results demonstrate that the interactions ranked top by our method have high functional homogeneity and localization coherence. Therefore, the proposed algorithm is a much more promising method to detect false positive interactions in PPI networks.

Newborn footprint recognition using orientation feature

Newborn and infant personal authentication is a critical issue for hospital, birthing centers, and other institutions where multiple births occur, which has not been well studied in the past. In this paper, we propose a novel online newborn personal authentication system for this issue based on footprint recognition. Compared with traditional offline footprinting scheme, the proposed system can capture digital footprint images with high quality. We also develop a preprocessing method for orientation and scale normalization. In this way, a coordinate system is defined to align the images, and a region of interest (ROI) is cropped. In recognition stage, four orientation feature-based approaches, Ordinal Code, BOCV, Competitive Code, and Robust Line Orientation Code, are exploited for recognition. A newborn footprint database is established to examine the performance of the proposed system, and promising experimental results demonstrate the effectiveness of the proposed system.

Orthogonal locally discriminant spline embedding for plant leaf recognition

Based on local spline embedding (LSE) and maximum margin criterion (MMC), two orthogonal locally discriminant spline embedding techniques (OLDSE-I and OLDSE-II) are proposed for plant leaf recognition in this paper. By OLDSE-I or OLDSE-II, the plant leaf images are mapped into a leaf subspace for analysis, which can detect the essential leaf manifold structure. Different from principal component analysis (PCA) and linear discriminant analysis (LDA) which can only deal with flat Euclidean structures of plant leaf space, OLDSE-I and OLDSE-II not only inherit the advantages of local spline embedding (LSE), but makes full use of class information to improve discriminant power by introducing translation and rescaling models. The proposed OLDSE-I and OLDSE-II methods are applied to recognize the plant leaf and are examined using the ICL-PlantLeaf and Swedish plant leaf image databases. The numerical results show compared with MMC, LDA, SLPP, and LDSE, the proposed OLDSE-I and OLDSE-II methods can achieve higher recognition rate.

Modified orthogonal discriminant projection for classification

From the perspective of manifold learning, the weight between two nodes of graph plays an indispensable role, which provides the similarity between pairwise nodes, and can effectively reveal the intrinsic relationship between data classes. In the original Locality Preserving Projections (LPP), Unsupervised Discriminant Projection (UDP), Orthogonal LPP (OLPP), and other spectral mapping methods, the weight between two points is usually defined as a heat kernel or simply 0-1 weight, which cannot effectively reflect the sample class information. In Orthogonal Discriminant Projection (ODP), the weight between two points was defined based on their local information and class information, but it is not a monotonically decreasing with the increase of the distance between two nodes, so it is not very sound. In this paper, we first analyze the defect of the weight in ODP, then propose a novel weight measure between two nodes of a graph by combining their label information and local information, finally present a modified ODP algorithm following the ODP technique. The modified ODP algorithm can explore the intrinsic structure of original data and enhance the classification ability. The experimental results show that the modified ODP algorithm is effective and feasible.

Semi-supervised orthogonal discriminant projection for plant leaf classification

Abstract Plant classification based on the leaf images is an important and tough task. For leaf classification problem, in this paper, a new weight measure is presented, and then a dimensional reduction algorithm, named semi-supervised orthogonal discriminant projection (SSODP), is proposed. SSODP makes full use of both the labeled and unlabeled data to construct the weight by incorporating the reliability information, the local neighborhood structure and the class information of the data. The experimental results on the two public plant leaf databases demonstrate that SSODP is more effective in terms of plant leaf classification rate.