Feature Extraction‐Based Bioaerosol Telemetry Identification Algorithm

Rong Yang; Jihui Dong; Bojia Su; Zhehou Yang; Yong Chen; Xiaofeng Li; Chunli Chen; Dingfu Zhou

doi:10.3788/CJL230847

Chinese Journal of Lasers, Volume. 51, Issue 5, 0510001(2024)

Feature Extraction‐Based Bioaerosol Telemetry Identification Algorithm

Rong Yang^1,2,3, Jihui Dong^1,2,3、*, Bojia Su^1,2,3, Zhehou Yang^1,2,3,4, Yong Chen^1,2,3,5, Xiaofeng Li^1,2,3,5, Chunli Chen^1,2,3, and Dingfu Zhou^1,2,3

Author Affiliations

¹Southwest Institute of Technical Physics, Chengdu 610041, Sichuan , China

²Sichuan Provincial Key Laboratory of National Defense Science and Technology of LiDAR and Device Technology, Chengdu 610041, Sichuan , China

³Key Laboratory of Laser Device Technology, China North Industries Group Corporation Limited, Chengdu 610041, Sichuan , China

⁴College of Physics, Beijing Institute of Technology, Beijing 100081, China

⁵College of Optoelectronics, Beijing Institute of Technology, Beijing 100081, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(9)

Fig. 1. Original fluorescence spectra of various biological substances obtained by 105 measurements under 355 nm excitation. (a) Rose pollen; (b) canola pollen; (c) pine pollen; (d) NADH; (e) riboflavin; (f) tryptophan

Download full size

Fig. 2. Schematic of decision tree algorithm

Download full size

Fig. 3. Fluorescence spectra of various biological substances after adding different noises under 355 nm excitation. (a1)‒(a4) Rose pollen; (b1)‒(b4) canola pollen; (c1)‒(c4) pine pollen; (d1)‒(d4) NADH; (e1)‒(e4) riboflavin; (f1)‒(f4) tryptophan

Download full size

Fig. 4. Established decision discriminant trees. (a) Decision discriminant tree for traditional algorithm; (b) decision discriminant tree for improved algorithm

Download full size

Fig. 5. Comparison diagrams of number of spectral point information used by algorithm before and after improvement. (a) Before improvement; (b) after improvement

Download full size

Fig. 6. $X (k)$ versus $k$ for fluorescence spectral signal of rose pollen after DCT

Download full size

Fig. 7. Recognition accuracy versus S_N of signal for two algorithms. (a) Improved algorithm; (b) traditional algorithm

Download full size

Table 1. Test results of traditional algorithm

View table

Table 1. Test results of traditional algorithm

Signal-to-noise ratio /dB	Average accuracy /%						Time /ms
Signal-to-noise ratio /dB	Rose	Canola	Pine	NADH	Riboflavin	Tryptophan	Time /ms
Original	98.52	95.41	97.20	99.60	99.39	96.00	84.12
20	95.79	95.67	96.57	95.01	96.57	92.53	116.60
10	87.26	59.58	67.63	71.51	96.22	68.09	393.6
5	72.09	36.13	49.39	47.15	83.19	44.15	509.1

Table 2. Test results of improved algorithm

View table

Table 2. Test results of improved algorithm

Signal-to-noise ratio /dB	Average accuracy /%						Time /ms
Signal-to-noise ratio /dB	Rose	Canola	Pine	NADH	Riboflavin	Tryptophan	Time /ms
Original	99.98	99.90	98.64	99.34	100.00	96.24	16.56
20	100.00	93.54	99.17	99.12	100.00	91.97	24.13
10	99.90	79.83	88.93	95.10	100.00	84.88	22.15
5	100.00	67.43	74.88	91.13	100.00	73.29	32.98

Tools

Get Citation

Copy Citation Text

Rong Yang, Jihui Dong, Bojia Su, Zhehou Yang, Yong Chen, Xiaofeng Li, Chunli Chen, Dingfu Zhou. Feature Extraction‐Based Bioaerosol Telemetry Identification Algorithm[J]. Chinese Journal of Lasers, 2024, 51(5): 0510001

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites

Paper Information

Category: remote sensing and sensor

Received: May. 18, 2023

Accepted: Aug. 7, 2023

Published Online: Feb. 19, 2024

The Author Email: Dong Jihui (j.h.dong@163.com)

DOI:10.3788/CJL230847

CSTR:32183.14.CJL230847

Topics

laser devices and laser physics

Lasers and Laser Optics

Laser physics

laser manufacturing

Instrumentation, Measurement and Metrology