Clustering-based initialization for non-negative matrix factorization

Xue, Yun; Tong, Chong Sze; Chen, Ying; Chen, Wen Sheng

doi:10.1016/j.amc.2008.05.106

Cited by 56 publications

(33 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The convergence rate and quality of the solution is totally dependent on the initialisation of W and H. Inappropriate initialisation may lead to slow and poor results. Some of the initialisation methods are the spherical k-means clustering method (Wild et al 2004), SVD-based method (Boutsidis and Gallopoulos 2008), relaxed kmeans clustering method (Xue et al 2008), Gabor wavelet method (Zheng et al 2007) and population-based methods (Janecek and Tan 2011). But all these initialisation methods needed complex pre-processing and might stuck in relatively poor local solution.…”

Section: Non-negative Matrix Factorisationmentioning

confidence: 98%

Colour normalisation of histopathological images

Saraswat

Arya

2013

Computer Methods in Biomechanics and Biomedical Engineering: Im

View full text Add to dashboard Cite

Colour transfer is a prime area of research in image processing in recent years. In many real-life image applications, colour transfer of the image is required. A few methods have been developed to alter the colour appearance of the images as per the colour information of the reference image. Histopathological image analysis is one of the application areas where colour transfer method can be used to reduce the effect of varying staining and illuminance. This paper proposes a hybrid method of colour normalisation using modified non-negative matrix factorisation (NMF) and novel colour transfer method. The NMF method has been initialised using Lab colour space which provides fast and reliable results. Uncorrelated color space has been used to transfer the colour of target image to source image. The results show that the proposed method outperforms the considered existing methods and can successfully be applied for the colour normalisation of histopathological images.

show abstract

Section: Non-negative Matrix Factorisationmentioning

confidence: 98%

Colour normalisation of histopathological images

Saraswat

Arya

2013

Computer Methods in Biomechanics and Biomedical Engineering: Im

View full text Add to dashboard Cite

show abstract

“…So this model is not completely automatic. A method based on sampler selection is proposed in [27], whose drawback is that it needs to try all the possible values of model order and then to choose a best one according to a certain convergent condition. Obviously, this method is not impressive enough for unsupervised learning.…”

Section: Related Workmentioning

confidence: 99%

Non-informative hierarchical Bayesian inference for non-negative matrix factorization

Sun

et al. 2015

Signal Processing

View full text Add to dashboard Cite

“…Kim and Tidor [14] proposed two data folding methods. Xue et al [15] proposed a clustering-based initialization for NMF.…”

Section: Related Workmentioning

confidence: 99%

In vivo snapshot hyperspectral image analysis of age-related macular degeneration

Lee

Wielaard

Fawzi

et al. 2010

2010 Annual International Conference of the IEEE Engineering in Medicine and Biology

View full text Add to dashboard Cite

Drusen, the hallmark lesions of age related macular degeneration (AMD), are biochemically heterogeneous and the identification of their biochemical distribution is key to the understanding of AMD. Yet the challenges are to develop imaging technology and analytics, which respect the physical generation of the hyperspectral signal in the presence of noise, artifacts, and multiple mixed sources while maximally exploiting the full data dimensionality to uncover clinically relevant spectral signatures. This paper reports on the statistical analysis of hyperspectral signatures of drusen and anatomical regions of interest using snapshot hyperspectral imaging and non-negative matrix factorization (NMF). We propose physical meaningful priors as initialization schemes to NMF for finding low-rank decompositions that capture the underlying physiology of drusen and the macular pigment. Preliminary results show that snapshot hyperspectral imaging in combination with NMF is able to detect biochemically meaningful components of drusen and the macular pigment. To our knowledge, this is the first reported demonstration in vivo of the separate absorbance peaks for lutein and zeaxanthin in macular pigment.

show abstract

Clustering-based initialization for non-negative matrix factorization

Cited by 56 publications

References 18 publications

Colour normalisation of histopathological images

Colour normalisation of histopathological images

Non-informative hierarchical Bayesian inference for non-negative matrix factorization

In vivo snapshot hyperspectral image analysis of age-related macular degeneration

Contact Info

Product

Resources

About