Yinwei Zhang

Yinwei Zhang

(He/Him)

Talks and presentations

Investigation of Curing Process Heterogeneity from Raman Spectrum via CP Decomposition

November 10, 2020

Presentation, INFORMS Annual Meeting, Virtual

Monitoring the Raman Spectrum Fluence in the Ultraviolet 3D Curing process is essential in the optical science. Traditional method based on SEM scanning suffers from complexity and time-consuming. Wavelet decomposition based method encounters difficulty in frequency selection and interpretation. This research proposes a novel unsupervised learning algorithm based on CP decomposition, which clusters cured samples via heterogeneous features extracted from Raman spectrum fluence data. The effectiveness of the proposed method is demonstrated by a real case study.

Functional Data Analytics for Detecting Bursts in Water Distribution Systems

October 01, 2019

Presentation, INFORMS Annual Meeting, Seattle WA, USA

It is crucial to develop effective and efficient algorithms to detect bursts in water distribution systems from spatially and temporally correlated hydraulic measurements. Traditional anomaly detection methods based on basis expansion may be applicable, falling short of accurate estimations of burst magnitude and starting time. This research proposes a spatio-temporal decomposition based burst detection method, which estimates the ST profile parameters according to the expected magnitude of spars burst. The effectiveness of the proposed method is demonstrated by a simulation case study.

Effective and Efficient Moving Object Detection by a Moving Camera

November 07, 2018

Presentation, INFORMS Annual Meeting, Phoenix, AZ, USA

Detecting and tracking moving targets with video surveillance systems is challenging, especially when using a moving camera. Conventional algorithms using projective transformation and frame differencing are inaccurate and slow. A new algorithm that combines a type of optical flow and color features is proposed to improve both detection accuracy in and tracking speed. Case studies of a variety of complex scenarios were conducted to demonstrate its effectiveness and efficiency.