Lighting Enhancement Aids Reconstruction of Colonoscopic Surfaces

High screening coverage during colonoscopy is crucial to effectively prevent colon cancer. Previous work has allowed alerting the doctor unsurveyed regions by reconstructing 3D colonoscopic surface from videos in real-time. However, lighting inconsistency of can cause a key component reconstruction system, SLAM optimization, fail. In this we focus on problem videos. To successfully improve consistency videos, have found necessary correction that adapts intensity distribution recent video frames. achieve real-time, designed and trained an RNN network. This network gamma value gamma-correction process. Applied our light-weight model significantly boosts success rate, making larger proportion segments reconstructable improving quality already reconstructed segments.