Filter Selection for Optimizing the Spectral Sensitivity of Broadband Multispectral Cameras Based on Maximum Linear...

Area of Science:

• Optics
• Imaging Science
• Sensor Technology

Background:

• Vector analysis methods, including maximum linear independence (MLI), are used to select broadband filter sets.
• Traditional MLI is suboptimal due to a fixed initial filter selection criterion.

Purpose of the Study:

• To investigate and identify characteristics of optimal filter sets for broadband multispectral imaging.
• To develop an improved framework for selecting filter sets that enhance system performance.

Main Methods:

• Conducted exhaustive imaging simulations, evaluating each filter as the initial filter.
• Analyzed filter set performance based on condition number and geometric transmittance features.
• Proposed a two-step framework for optimal filter set selection.

Main Results:

• Optimal filter sets exhibit minimized condition numbers and specific geometric transmittance features.
• Key features include a distinct transmittance peak for the first filter and uniform peak distribution.
• Substantial overlaps between adjacent filter transmittance curves were observed.

Conclusions:

• The best-performed filter sets can be intuitively identified through vector analysis and experimental verification.
• The recommended two-step framework significantly enhances the performance of multispectral imaging systems.
• This research aids in optimizing the spectral sensitivity of broadband multispectral imaging sensors.