Improving color reliability of digital textile images via optimized acquisition and preprocessing

Area of Science:

• Textile Science
• Digital Imaging
• Colorimetry

Background:

• Digitalization and automation in the textile industry necessitate accurate image-based analysis.
• Color reliability is crucial for textile research, manufacturing, and digital commerce applications.
• Existing methods often suffer from color distortion in digital textile images.

Purpose of the Study:

• To develop an integrated imaging and preprocessing framework for improved color reliability in digital textile images.
• To establish optimal settings for image acquisition environments and preprocessing strategies.
• To enhance the precision of textile image analysis across various applications.

Main Methods:

• Compared color distortion in 11 textile samples under different background colors (white/black).
• Evaluated image fusion as a preprocessing step for color reliability and image quality.
• Derived optimal settings for image acquisition and preprocessing.
• Validated proposed guidelines by comparing color distortion with conventional correction methods.

Main Results:

• Appropriate physical environments and image preprocessing significantly increase color stability and reliability.
• The proposed framework yielded an average improvement of 21.2% in visual fidelity compared to original fabrics.
• Image fusion proved effective in enhancing color reliability and image quality.

Conclusions:

• The integrated imaging and preprocessing framework enhances the color stability and reliability of digital textile images.
• Optimal acquisition environments and preprocessing strategies are key to accurate textile surface analysis.
• This work provides guidelines for achieving higher visual fidelity in digital textile representations.