Design of an autofocus capsule endoscope system and the corresponding 3D reconstruction algorithm

Area of Science:

• Medical Imaging
• Optical Engineering
• Computer Vision

Background:

• Traditional 2D capsule endoscopes provide insufficient image clarity for accurate medical diagnosis.
• Blurred images in capsule endoscopy hinder the accuracy of 3D reconstruction.
• Existing systems lack advanced focusing capabilities, limiting diagnostic potential.

Purpose of the Study:

• To design and develop a compact, autofocus capsule endoscope system.
• To enhance image clarity and improve 3D reconstruction accuracy for diagnostic purposes.
• To overcome the limitations of traditional 2D capsule endoscopy.

Main Methods:

• Development of a compact capsule endoscope system utilizing a liquid lens for electronic autofocus control without moving parts.
• Implementation of a novel 3D reconstruction algorithm tailored to the system's zooming capabilities.
• System evaluation through simulations and experimental validation.

Main Results:

• The developed system achieves a depth of field from 3-100 mm and a field of view of approximately 110°.
• Captured images demonstrate significantly improved clarity compared to traditional capsule endoscopes.
• The 3D reconstruction algorithm successfully matched more feature points, leading to higher reconstruction accuracy.

Conclusions:

• The proposed autofocus capsule endoscope system offers superior image quality for enhanced medical diagnosis.
• The integrated 3D reconstruction strategy effectively addresses challenges posed by image blurring and zooming.
• This technology holds promise for improving the speed and accuracy of endoscopic diagnoses.