Error-Free Demodulation of Pixelated Interferograms: Updated review and circumstantial evidence of plagiarism by 4D Technology Corporation

TL;DR

This paper reviews the development of error-free pixelated interferogram demodulation, highlights potential plagiarism by 4D Technology Corporation, and discusses ethical issues in scientific research.

Contribution

It provides an updated review of error-free demodulation methods and presents circumstantial evidence of intellectual property appropriation by 4DTC.

Findings

Analysis of 4DTC algorithms shows they were limited to narrow bandwidth fringe demodulation.

A new error-free demodulation method was introduced by Servin et al. in 2010.

Circumstantial evidence suggests possible plagiarism of Servin's ideas by 4DTC.

Abstract

This paper examines the historical development that led to error-free phase demodulation of pixelated spatial-carrier interferograms between 2004 and 2010. It also evaluates evidence suggesting that 4D Technology Corporation (4DTC), in their SPIE 7790 publication [18], may have adopted key concepts from a manuscript submitted by Servin and collaborators, which at that time was still under peer review [17]. We analyze 4DTC algorithms based on 2x2 and 3x3 spatial convolution phase shifting algorithms (SC-PSA), developed from 2004 to 2009 [9-15], and show that they were fundamentally limited to narrow bandwidth fringe demodulation. Notably, only one 4DTC's paper [13] acknowledged major phase errors when these methods were applied to wideband pixelated fringes, but the problem remained unsolved. In 2010, the team at Centro de Investigaciones en Optica (CIO), led by Servin, entered the field [16,17]. At that point, CIO and 4DTC were the only two known groups pursuing error free demodulation of wideband pixelated interferograms. In June 2010, Servin et al submitted to Optics Express [17] a definitive error-free solution introducing a new method: the complex fringe-carrier product followed by step-edge low-pass Fourier filtering. This approach fully superseded the 2x2 and 3x3 kernels used by 4DTC [9-15], reaching the theoretical slope limit for pixelated sensors. Given the notable phase demodulation similarities between papers [17] and [18], and that the presentation of [18] was just seven weeks after Servin submission [17], it is plausible that 4DTC or an affiliate had early access to the submission. The circumstantial evidence suggests that core ideas from [17] were incorporated into 4DTC August presentation [18]. This paper retraces that scientific path and examines the ethical implications of uncredited appropriation in optical research.