Image-based spectral transmission estimation using "sensitivity comparison"

Area of Science:

• Optical Physics
• Image Processing
• Color Science

Background:

• Digital cameras are widely used for spectral reflectance estimation.
• Spectral transmission measurement using digital cameras is underexplored.
• Existing methods often require prior sample knowledge or statistical datasets.

Purpose of the Study:

• To present a novel method for spectral transmission estimation using digital cameras.
• To develop a technique that does not require a priori knowledge of samples or statistical data.
• To evaluate the performance of the proposed method against existing techniques.

Main Methods:

• The sensitivity comparison (SC) method was developed for spectral transmission estimation.
• The method utilizes camera sensitivity in the absence and presence of transparent colored objects.
• Calibration and measurement require one shot each, similar to spectrophotometers.

Main Results:

• The SC method was validated using 138 Kodak Wratten Gelatin filters.
• Performance was compared with a training-based Matrix R method under varying noise levels.
• Testing on 33 manmade colored transparent films demonstrated promising results with various error metrics.

Conclusions:

• The sensitivity comparison method provides a viable approach for spectral transmission estimation with digital cameras.
• The technique is independent of sample-specific information, enhancing its applicability.
• The SC method shows potential for accurate spectral transmission measurements in practical scenarios.