Premium
Removing Spikes from Voltammetric Curves in the Presence of Random Noise
Author(s) -
Jakubowska Malgorzata,
Kubiak Wladyslaw W.
Publication year - 2005
Publication title -
electroanalysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.574
H-Index - 128
eISSN - 1521-4109
pISSN - 1040-0397
DOI - 10.1002/elan.200503276
Subject(s) - interpolation (computer graphics) , noise (video) , algorithm , computer science , set (abstract data type) , curve fitting , polynomial , process (computing) , interpretation (philosophy) , biological system , mathematics , artificial intelligence , mathematical analysis , image (mathematics) , machine learning , programming language , biology , operating system
The presence of noises and spikes in the experimental curve generates problems in the process of visualization and interpretation of the data. The procedure for spikes removal from voltammetric curves which are additionally distorted by random noise of different amplitude is proposed. The developed method bases on the assumption that spikes are measuring points corrupted by the gross errors. The statistical tests for detecting such errors are applied. Recognized spikes are replaced by points obtained by polynomial interpolation. The method eliminates close to 100% of spikes and reconstructs the ideal curve shape better than other algorithms. The considered new procedure gives the improvement of analytical parameters and may be applied for the set of curves in voltammetric experiment. Because the method is statistically valid it may be used for automatic spikes removal in intelligent computerized measuring systems in analytical chemistry.