この続きは、会員限定コンテンツです。
会員登録をすると、全文をご覧いただけます。
Infrared (IR) spectroscopy is widely used for molecular analysis, but noise significantly impairs spectral interpretation. This article reviews key noise types and reduction methods—such as SG, wavelet, HHT, PCA, and CNN—and introduces MI-6’s automated workflow that dynamically selects optimal denoising techniques per spectrum, improving peak detection accuracy and signal clarity.
Sivakorn Kanharattanachai
MI-6 Ltd.Data Scientist
He is an experienced nanomaterial engineer who then transitioned to computer engineering with expertise in text mining, deep learning,
and imbalanced data learning. He previously served as a data scientist at Charoen Pokphand Group (CP) Thailand, where he specialized in time series analysis,
satellite image processing, optical character recognition (OCR) development, and automated data systems implementation.
He is currently working as a data scientist at MI-6, focusing on advanced feature extraction and spectral data analysis through deep learning methodologies.
人気の記事
お役立ち資料
Loading...
