Denoising and feature extraction for control chart pattern recognition in autocorrelated processes

Abstract

The main purpose of this paper is to develop a neural network-based recogniser for control chart pattern recognition in autocorrelated processes. First, we apply a multi-resolution analysis approach based on Haar Discrete Wavelet Transform (DWT) to denoise, decorrelate and extract distinguished features from autocorrelated data. Second, we introduce a supervised neural network for control chart pattern recognition. The performance of the neural network using features extracted from wavelet analysis as the components of the input vectors is explored and compared. In this study, we investigated three types of unnatural patterns, namely increasing and decreasing trends, cyclic patterns, upward and downward shifts. Extensive comparisons based on simulation study indicate that the proposed neural network performs better than that using raw data as inputs.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 11 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.