Fourier transform infrared spectroscopy microscopic imaging classification based on multifractal methods

Lian Liu; Xiukun Yang; Xiaojun Jing

doi:10.1364/AO.56.001689

Applied Optics
Vol. 56,
Issue 6,
pp. 1689-1700
(2017)
•https://doi.org/10.1364/AO.56.001689

Fourier transform infrared spectroscopy microscopic imaging classification based on multifractal methods

Lian Liu, Xiukun Yang, and Xiaojun Jing

Not Accessible

Your library or personal account may give you access

Get PDF
Email
Share
Get Citation
Copy Citation Text
Lian Liu, Xiukun Yang, and Xiaojun Jing, "Fourier transform infrared spectroscopy microscopic imaging classification based on multifractal methods," Appl. Opt. 56, 1689-1700 (2017)

Export Citation
- BibTex
- Endnote (RIS)
- HTML
- Plain Text
Citation alert
Save article

Related Topics
Table of Contents Category
- Imaging Systems
Optics & Photonics Topics
?

The topics in this list come from the Optics and Photonics Topics applied to this article.

About this Article
History
- Original Manuscript: October 12, 2016
- Revised Manuscript: January 18, 2017
- Manuscript Accepted: January 23, 2017
- Published: February 16, 2017
Virtual Issues
Virtual Journal for Biomedical Optics Vol. 12, Iss. 4

Abstract

Multifractal analysis (MFA) based on generalized concepts of fractals has been applied to biological tissues composed of complex structures. In this paper, a new MFA methodology based on the neighborhood spatial correlation (NSC) is proposed for an extracting texture feature. NSC is used to extract spatial features, and the obtained spatial features are combined with spectral features of characteristic absorption peaks (CAPs) to promote more feature information. This spatial-spectral structure is used as a feature to differentiate cholesterol from Fourier transform infrared spectroscopy microscopic imaging of a rabbit artery by a support vector machine classifier. The dataset was collected between 4000 and $720 {cm}^{- 1}$ on rabbit arteries as research objects. The experimental results show that the accuracy of the proposed spatial-spectral structure is higher than that of other multivariate analysis methods (PCA and 2DPCA). The NSC method, compared to the bottom interface method, new bottom interface method, variance method multi-weight method, and neighborhood spatial correlation method, could effectively reduce the influence of speckle noise, and the convergence rate of the weight factor $q$ is not increased.

Full Article | PDF Article

More Like This

Optical image encryption based on multifractional Fourier transforms

Banghe Zhu, Shutian Liu, and Qiwen Ran
Opt. Lett. 25(16) 1159-1161 (2000)

Demodulation of multimodulation artifacts in Fourier-transform infrared spectroscopy

Mathias Schilling and Jürgen Stohner
Appl. Opt. 56(14) 4076-4084 (2017)

Spectral-spatial feature-based neural network method for acute lymphoblastic leukemia cell identification via microscopic hyperspectral imaging technology

Qian Wang, Jianbiao Wang, Mei Zhou, Qingli Li, and Yiting Wang
Biomed. Opt. Express 8(6) 3017-3028 (2017)

Previous Article Next Article

Cited By

You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Figures (18)

You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Tables (6)

You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

Equations (27)

You do not have subscription access to this journal. Equations are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.

Contact your librarian or system administrator
or
Login to access Optica Member Subscription

$ω$ ( ${cm}^{- 1}$ )	BIM	NBIM	VM	MWM	NSC
1056	12.0755	7.5159	9.9613	2.8104	2.0435
1464	12.0678	7.4492	9.1269	2.5476	2.0140
2868	12.0974	7.4877	11.0784	2.6669	1.7880
2924	12.0692	7.4668	6.5276	2.5901	1.8436
Average	12.0774	7.4799	9.1735	2.6537	1.9222

Method	Wavenumber ( ${cm}^{- 1}$ )	$α_{\max}$	$α_{\min}$	$Δ α$	$\max f (α)$	$\min f (α)$	$f (α_{\max})$	$f (α_{\min})$	$Δ f$	$q$ Cutoff Value
BIM	1056	2.7455	0	2.7455	2.2482	0.3290	0.3290	2.2264	1.8974	2
	1464	2.8354	0	2.8354	2.2482	0.0448	0.0448	2.2264	2.1816	2
	2868	2.8607	0	2.8607	2.2482	0.0075	0.0075	2.2264	2.1289	2
	2924	2.8265	0	2.8265	2.2482	0.0535	0.0535	2.2264	2.1729	2
NBIM	1056	2.1627	0	2.1627	1.9972	0.3294	0.3294	1.8846	1.5552	2
	1464	2.2528	0	2.2528	1.9972	0.0449	0.0449	1.8846	1.8397	2
	2868	2.2781	0	2.2781	1.9972	0.0076	0.0076	1.8846	1.8770	2
	2924	2.2365	0	2.2365	1.9972	0.0452	0.0452	1.8846	1.8394	2
VM	1056	4.5553	1.5468	3.0085	1.9972	0	0	0.4704	0.4704	10
	1464	4.6565	1.8049	2.8516	1.9972	0.0792	0.0792	1.0776	0.9984	10
	2868	4.3641	1.7810	2.5831	1.9972	0	0	0.9535	0.9535	10
	2924	3.9518	1.7870	2.1648	1.9972	0.2005	0.2005	1.0756	0.8751	10
MWM	1056	1.4289	0	1.4289	1.9972	0.1743	0.1743	1.8882	1.7139	2
	1464	1.4769	0	1.4769	1.9972	0.0660	0.0660	1.8882	1.8882	2
	2868	1.2479	0	1.2479	1.9972	0.2416	0.2416	1.8882	1.6466	2
	2924	1.5283	0	1.5283	1.9972	0.0320	0.0320	1.8882	1.8562	2
NSC	1056	1.1468	0	1.1468	1.9972	0.3914	0.3914	1.8846	1.4932	2
	1464	1.2999	0	1.2999	1.9972	0.1490	0.1490	1.8846	1.7356	2
	2868	1.6514	0	1.6514	1.9972	0.0059	0.0059	1.8846	1.8787	2
	2924	1.6514	0	1.6514	1.9972	0.0059	0.0059	1.8846	1.8787	2

Method		$s = 5$	$s = 10$	$s = 15$	$s = 20$
PCA	CA	$58.58 \pm 0.31$	$49.71 \pm 0.27$	$88.01 \pm 0.11$	$99.22 \pm 0.024$
	BA	$21.92 \pm 0.12$	$99.66 \pm 0.04$	$96.72 \pm 0.04$	$90.87 \pm 0.03$
	OA	$23.26 \pm 0.12$	$98.81 \pm 0.02$	$96.57 \pm 2.88$	$91.01 \pm 3.81$
SDPCA	CA	$89.03 \pm 0.08$	$86.73 \pm 0.11$	$92.43 \pm 1.42$	$70.63 \pm 0.06$
	BA	$81.56 \pm 0.06$	$99.05 \pm 0.05$	$99.25 \pm 0.75$	$99.44 \pm 0.05$
	OA	$81.64 \pm 0.74$	$98.84 \pm 0.25$	$99.13 \pm 0.07$	$98.94 \pm 0.05$
2DPCA	CA	$48.34 \pm 0.12$	$61.17 \pm 2.32$	$94.95 \pm 0.10$	$66.02 \pm 1.90$
	BA	$30.14 \pm 0.68$	$99.67 \pm 0.64$	$99.12 \pm 0.01$	$99.55 \pm 1.61$
	OA	$31.34 \pm 0.42$	$98.96 \pm 0.06$	$99.05 \pm 0.01$	$98.97 \pm 0.11$
${(2 D)}^{2} PCA$	CA	$54.76 \pm 0.19$	$74.37 \pm 0.38$	$84.66 \pm 0.51$	$84.27 \pm 0.77$
	BA	$99.49 \pm 0.49$	$99.43 \pm 0.89$	$99.34 \pm 0.41$	$99.35 \pm 0.62$
	OA	$98.72 \pm 0.09$	$98.99 \pm 0.47$	$99.34 \pm 0.40$	$99.09 \pm 0.37$
CAPI	CA	$98.66 \pm 0.19$	$98.80 \pm 0.38$	$99.72 \pm 0.77$	$99.95 \pm 0.95$
	BA	$88.99 \pm 0.02$	$91.69 \pm 0.05$	$93.37 \pm 0.03$	$79.87 \pm 0.12$
	OA	$89.18 \pm 2.52$	$91.83 \pm 2.99$	$93.48 \pm 3.22$	$80.22 \pm 2.40$
$CAPI + NSC$	CA	$99.70 \pm 0.13$	$99.41 \pm 0.73$	$97.73 \pm 0.73$	$99.61 \pm 0.02$
	BA	$98.23 \pm 0.63$	$99.72 \pm 0.05$	$96.32 \pm 0.05$	$97.54 \pm 0.47$
	OA	$98.25 \pm 0.62$	$99.71 \pm 0.33$	$96.34 \pm 0.04$	$97.57 \pm 0.46$

Method		$ε = 3$	$ε = 5$	$ε = 7$	$ε = 9$
BIM	CA	$89.57 \pm 0.68$	$95.08 \pm 0.25$	$97.35 \pm 0.73$	$96.05 \pm 2.14$
	BA	$97.55 \pm 0.19$	$99.43 \pm 0.05$	$99.51 \pm 0.04$	$99.40 \pm 1.26$
	OA	$97.41 \pm 0.18$	$99.36 \pm 0.04$	$99.47 \pm 0.05$	$99.35 \pm 0.07$
NBIM	CA	$90.29 \pm 0.51$	$95.47 \pm 0.71$	$96.31 \pm 1.55$	$96.05 \pm 1.42$
	BA	$97.58 \pm 0.14$	$99.33 \pm 0.15$	$99.57 \pm 0.08$	$99.74 \pm 0.08$
	OA	$97.46 \pm 0.13$	$99.26 \pm 0.14$	$99.51 \pm 0.05$	$99.37 \pm 0.06$
VM	CA	$95.86 \pm 0.11$	$92.88 \pm 0.29$	$93.39 \pm 0.39$	$94.37 \pm 0.10$
	BA	$98.45 \pm 0.02$	$99.31 \pm 0.06$	$99.15 \pm 1.61$	$98.93 \pm 0.12$
	OA	$98.41 \pm 0.07$	$99.19 \pm 0.06$	$99.05 \pm 0.08$	$98.85 \pm 0.01$
MWM	CA	$90.87 \pm 0.19$	$80.58 \pm 0.34$	$73.27 \pm 0.45$	$72.62 \pm 0.70$
	BA	$97.28 \pm 0.09$	$98.34 \pm 0.08$	$98.53 \pm 0.36$	$97.49 \pm 0.62$
	OA	$97.17 \pm 0.04$	$98.03 \pm 0.08$	$98.09 \pm 0.35$	$97.07 \pm 0.62$
NSC	CA	$96.57 \pm 0.16$	$97.42 \pm 0.19$	$99.41 \pm 0.73$	$85.11 \pm 1.29$
	BA	$98.73 \pm 0.25$	$99.71 \pm 0.18$	$99.72 \pm 0.05$	$99.60 \pm 0.03$
	OA	$98.70 \pm 0.25$	$99.42 \pm 0.17$	$99.71 \pm 0.33$	$99.35 \pm 0.05$

Method	BIM	NBIM	VM	MWM	NSC
Time(s)	0.7373	0.6453	0.4317	0.6942	1.0237

Method	BIM- SVM	NBIM- SVM	VM- SVM	MWM- SVM	NSC- SVM
Time(s)	54.38	54.72	53.48	55.69	71.43

Abstract

Cited By

Figures (18)

Tables (6)

Equations (27)

Applied Optics