Face Recognition Using Sparse Representation-Based Classification on K-Nearest Subspace

Mi, Jian-Xun; Liu, Jin-Xing

doi:10.1371/journal.pone.0059430

Cited by 32 publications

(17 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Twenty six images divided in two sessions were captured from each person. The same experiment protocol of Mi et al [53] , [54] was used and for each individual 7 images from the first session without any occlusion were used for training, while the 7 correspondent images from the second session was used for testing. Table 8 shows that the CANet achieves the highest recognition rate in comparison with the methods presented by Mi et al [53] , [54] such as Linear Regression-based Classification (LRC), Robust Linear Regression-based Classification (CLRC) and Sparse Representation-based Classification on K-Nearest Subspace (SRC-KNS).…”

Section: Methodsmentioning

confidence: 99%

Constructive Autoassociative Neural Network for Facial Recognition

2014

View full text Add to dashboard Cite

Autoassociative artificial neural networks have been used in many different computer vision applications. However, it is difficult to define the most suitable neural network architecture because this definition is based on previous knowledge and depends on the problem domain. To address this problem, we propose a constructive autoassociative neural network called CANet (Constructive Autoassociative Neural Network). CANet integrates the concepts of receptive fields and autoassociative memory in a dynamic architecture that changes the configuration of the receptive fields by adding new neurons in the hidden layer, while a pruning algorithm removes neurons from the output layer. Neurons in the CANet output layer present lateral inhibitory connections that improve the recognition rate. Experiments in face recognition and facial expression recognition show that the CANet outperforms other methods presented in the literature.

show abstract

Section: Methodsmentioning

confidence: 99%

Constructive Autoassociative Neural Network for Facial Recognition

2014

View full text Add to dashboard Cite

show abstract

“…For SC+RC, SDC+RC and SSGL+RC, we use the results from SC, SDC and SSGL as the initial values forĝ 0 , respectively. CDR is obtained after solving the problems in (14) and (15) iteratively.…”

Section: A Cup-to-disc Ratio Computationmentioning

confidence: 99%

“…A decimal score is obtained for cataract grading by solving the problems in (14) and (15) iteratively. A ceiling operation is further applied to get the integral grading score.…”

Section: B Cataract Gradingmentioning

confidence: 99%

See 1 more Smart Citation

Sparse Range-Constrained Learning and Its Application for Medical Image Grading

Cheng

2018

IEEE Trans. Med. Imaging

View full text Add to dashboard Cite

Sparse learning has been shown to be effective in solving many real-world problems. Finding sparse representations is a fundamentally important topic in many fields of science including signal processing, computer vision, genome study, and medical imaging. One important issue in applying sparse representation is to find the basis to represent the data, especially in computer vision and medical imaging where the data are not necessary incoherent. In medical imaging, clinicians often grade the severity or measure the risk score of a disease based on images. This process is referred to as medical image grading. Manual grading of the disease severity or risk score is often used. However, it is tedious, subjective, and expensive. Sparse learning has been used for automatic grading of medical images for different diseases. In the grading, we usually begin with one step to find a sparse representation of the testing image using a set of reference images or atoms from the dictionary. Then in the second step, the selected atoms are used as references to compute the grades of the testing images. Since the two steps are conducted sequentially, the objective function in the first step is not necessarily optimized for the second step. In this paper, we propose a novel sparse range-constrained learning (SRCL) algorithm for medical image grading. Different from most of existing sparse learning algorithms, SRCL integrates the objective of finding a sparse representation and that of grading the image into one function. It aims to find a sparse representation of the testing image based on atoms that are most similar in both the data or feature representation and the medical grading scores. We apply the new proposed SRCL to two different applications, namely, cup-to-disc ratio (CDR) computation from retinal fundus images and cataract grading from slit-lamp lens images. Experimental results show that the proposed method is able to improve the accuracy in CDR computation and cataract grading.

show abstract

“…However, according to [7], the authors argued that for the linear regressionbased classifiers, such as sparse-representation based classification (SRC), features extracted linearly, including the above mentioned ones, have little difference among them. In other words, even we use very simple feature extraction method as downsampling, it is highly possible that the recognition results are equal to the cases that very complicated linear features are used [8,9]. Therefore some studies suggested that for face recognition issue more attention should be paid to design high robust classifier since simple linear features are competent [10][11][12].…”

Section: Introductionmentioning

confidence: 99%