Dr. Bingxin Liu Profile

Dr. Bingxin Liu

at Dalian Maritime Univ

SPIE Involvement:

Author

Publications (3)

Proceedings Article | 14 August 2019 Paper

Oil slick extraction from hyperspectral images using a modified stacked auto-encoder network

Wen Chang, Bingxin Liu, Qiang Zhang

Proceedings Volume 11179, 1117931 (2019) https://doi.org/10.1117/12.2539664

KEYWORDS: Hyperspectral imaging, Remote sensing, Machine learning, Neural networks, Image classification

Read Abstract +

SPIE Journal Paper | 20 November 2015

Oil spill detection method using X-band marine radar imagery

Xueyuan Zhu, Ying Li, Haiyang Feng, Bingxin Liu, Jin Xu

JARS, Vol. 9, Issue 01, 095985, (November 2015) https://doi.org/10.1117/12.10.1117/1.JRS.9.095985

KEYWORDS: Radar, Ocean optics, X band, Digital filtering, Image filtering, Image processing, Antennas, Signal to noise ratio, Signal attenuation, Image segmentation

Read Abstract +

Proceedings Article | 18 November 2014 Paper

Assessing the capabilities of hyperspectral remote sensing to map oil films on waters

Bingxin Liu, Ying Li, Xueyuan Zhu

Proceedings Volume 9299, 929902 (2014) https://doi.org/10.1117/12.2066293

KEYWORDS: Sensors, Reflectivity, Remote sensing, Environmental sensing, Signal to noise ratio, Interference (communication), Target detection, Ocean optics, Image sensors, Light scattering

Read Abstract +

The harm of oil spills has caused extensive public concern. Remote sensing technology has become one of the most effective means of monitoring oil spill. However, how to evaluate the information extraction capabilities of various sensors and choose the most effective one has become an important issue. The current evaluation of sensors to detect oil films was mainly using in-situ measured spectra as a reference to determine the favorable band, but ignoring the effects of environmental noise and spectral response function. To understand the precision and accuracy of environment variables acquired from remote sensing, it is important to evaluate the target detection sensitivity of the entire sensor-air-target system corresponding to the change of reflectivity. The measurement data associated with the evaluation is environmental noise equivalent reflectance difference (NEΔRE ), which depends on the instrument signal to noise ratio(SNR) and other image data noise (such as atmospheric variables, scattered sky light scattering and direct sunlight, etc.). Hyperion remote sensing data is taken as an example for evaluation of its oil spill detection capabilities with the prerequisite that the impact of the spatial resolution is ignored. In order to evaluate the sensor’s sensitivity of the film of water, the reflectance spectral data of light diesel and crude oil film were used. To obtain Hyperion reflectance data, we used FLAASH to do the atmospheric correction. The spectral response functions of Hyperion sensor was used for filtering the measured reflectance of the oil films to the theoretic spectral response. Then, these spectral response spectra were normalized to NEΔRE, according to which, the sensitivity of the sensor in oil film detecting could be evaluated. For crude oil, the range for Hyperion sensor to identify the film is within the wavelength from 518nm to 610nm (Band 17 to Band 26 of Hyperion sensors), within which the thin film and thick film can also be distinguished. For light diesel oil film, the range for Hyperion sensor to identify the film is within the wavelength from 468nm to 752nm (Band 12 to Band 40 of Hyperion sensors).

View contact details

UPDATE YOUR PROFILE

Is this your profile? Update it now.

Sign into your SPIE.org account

Don’t have a profile and want one?

Create an account on SPIE.org

Keywords/Phrases

Search In:

Publication Years