A New Method for Removing Internal Scattering Noise in iToF Camera

Amplitude Modulated Continuous Wave (AMCW) Indirect Time-of-Flight (iToF) cameras are widely used in 3D data acquisition due to their real-time operation, compactness, and cost-effectiveness [1]. However, environmental factors pose a significant challenge to accuracy, with internal scattering being a particularly prominent one. Internal scattering occurs when light reflects between the lens surfaces and the image sensor, causing pixel crosstalk and disrupting depth data, particularly in scenes with both foreground and background objects. To address this, the Point Spread Function (PSF) is commonly used for denoising by modeling light distribution [2]. However, existing PSF models often fail in complex scenarios due to their reliance on simple Gaussian or ideal distributions, which cannot handle variations in intensity or depth. This paper proposes an improved PSF model using wavelet transforms and multi-scale approaches to separately capture local and global scattering characteristics for more accurate correction. Our core algorithm formulas are as follows: