Ihsan Auditia Akhinov, B.S. Rahayu Purwanti, Hasvienda Muhammad Ridwlan, Muhammad Nurtanto, Anggi Mardiono, Rangga Aulia Rusli
This study focuses on reducing noise in 2D-to-3D image conversion using a combination of Masking Algorithms, Triangulation, and Kalman Filter Formulas (KFF). Although these methods are widely applied, masking techniques for minimizing height errors due to noise are rarely discussed. The objective is to reduce noise through a hybrid masking method, followed by KFF for accuracy enhancement. Experiments were conducted on planar objects with white and black surfaces, captured along a laser line to generate 2D images with dominant red hues. The 3D dimension was reconstructed using HSV-HPF, RGB, and a combined RGB-HSV (HPF) masking approach in OpenCV. Results show that the combined masking (COMB) method achieves the lowest noise levels. Integration with KFF further improves accuracy, achieving a Root Mean Square Error (RMSE) of 0.087375, compared to 0.117 without KFF. Accuracy testing used RMSE with Q = 0.3 and R = 0.8. The findings indicate that KFF reduces noise by approximately 27% and significantly improves precision in the 2D-to-3D image conversion process. © Published under licence by IOP Publishing Ltd.
Politeknik Negeri Jakarta, Depok, Indonesia; Department of Technology and Vocational Education, Universitas Negeri Jakarta, Jakarta, Indonesia
Research at a Glance
Register to unlock
Topics & SDG Alignment
Register to unlock
Collaboration
Register to unlock
Author Profile (Selected)
Register to unlock
References Overview
Register to unlock
Journal & Source
Register to unlock
Metadata & Integrity
Register to unlock