4.7 Article

A Projection Transformation Method for Eliminating Crosstalk in Networked Thermopile Arrays

Journal

IEEE SENSORS JOURNAL
Volume 22, Issue 4, Pages 3263-3269

Publisher

IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/JSEN.2021.3139081

Keywords

Sensors; Voltage; Crosstalk; Sensor arrays; Voltage measurement; Resistance; Newton method; Crosstalk elimination; projection transformation; networked thermopile arrays; Newton-Raphson iterative algorithm; non-linear least squares

Funding

  1. National Natural Science Foundation of China [61601455, 61874137]
  2. Scientific Instrument Developing Project of the Chinese Academy of Sciences [ZDKYYQ20200007]

Ask authors/readers for more resources

In this work, a projection transformation method suitable for small-scale networked thermopile arrays is proposed to eliminate crosstalk and reconstruct thermal images accurately.
The networked thermopile array in row-column fashion suffered from the crosstalk caused by parasitic parallel paths. To measure accurate two-dimensional thermal information, it is of great importance to eliminate the crosstalk. In this work, based on the response voltage non-linearity of thermopile pixels as well as the response voltage non-uniformity of array, a projection transformation method that suits for small-scale networked thermopile arrays is proposed. It transforms the crosstalk elimination problem into an overdetermined non-linear least squares problem, and is solved by the Newton-Raphson iterative algorithm. Its reconstruction error is analyzed carefully by Monte Carlo simulation, and its ability to implement non-uniformity and defective pixel correction is also discussed. Its validity is verified experimentally using a3x3 networked thermopile array. The experimental result shows that the original IR image can be reconstructed from measured IR images by this method, and the reconstruction error is significantly decreased from 17.03% to 2.10%.

Authors

I am an author on this paper
Click your name to claim this paper and add it to your profile.

Reviews

Primary Rating

4.7
Not enough ratings

Secondary Ratings

Novelty
-
Significance
-
Scientific rigor
-
Rate this paper

Recommended

No Data Available
No Data Available