Color 3D display and security via monochrome scenes

Area of Science:

• Computer Vision
• 3D Display Technology
• Image Encryption

Background:

• Color 3D displays offer superior realism but rendering monochrome scenes in 3D color remains a significant challenge.
• Existing methods lack effective solutions for achieving vivid color 3D visual effects from monochrome content.

Purpose of the Study:

• To propose a novel color stereo reconstruction algorithm (CSRA) for generating color 3D information from monochrome scenes.
• To develop a deep learning-based approach for accurate color stereo estimation (CSE).
• To implement an efficient and secure 3D image encryption scheme utilizing the developed algorithm.

Main Methods:

• A deep learning-based color stereo estimation (CSE) network was designed to extract color 3D information.
• A self-made display system was used to verify the visual effect of the color 3D display.
• A 3D image encryption scheme was achieved by integrating the CSRA with two-dimensional double cellular automata (2D-DCA).

Main Results:

• The proposed CSE network successfully obtained color 3D information from monochrome scenes.
• The self-made display system demonstrated a vivid color 3D visual effect.
• The CSRA-based encryption scheme achieved real-time, high-security 3D image encryption with a large key space and parallel processing capabilities.

Conclusions:

• The developed CSRA effectively addresses the challenge of color 3D display for monochrome scenes.
• The proposed method enables high-quality color 3D visualization and secure 3D image encryption.
• The integration of deep learning and 2D-DCA offers a promising direction for advanced 3D imaging applications.