Holographic Image Segmentation for Computational Load Reduction

The patent divides the original image into multiple segments based on depth information, creating multiple depth-resolved images. Each segment corresponds to a specific depth range, allowing the diffraction fringe pattern to be calculated separately for each segment. This segmentation reduces the overall computational complexity by breaking down the large-scale calculation into smaller, more manageable calculations for each depth layer.

The patent introduces a depth dimension by segmenting the image along the z-axis (depth direction). Instead of processing the entire 2D image simultaneously, the system creates multiple 2D images at different depth planes and processes them separately. This dimensional transformation from a single 2D plane to multiple depth-resolved 2D planes enables more efficient computation of the diffraction fringe pattern.

The patent applies partial action by calculating diffraction fringe patterns only for specific depth segments rather than processing the entire image at full resolution uniformly. By focusing computational resources on relevant depth ranges and using adaptive segmentation, the system achieves acceptable image resolution while reducing unnecessary calculations in regions where high precision is not required.

The patent implements local quality by using adaptive segmentation where the number and size of segments vary according to the depth and content of different regions. Regions requiring higher resolution receive finer segmentation, while other regions use coarser segmentation. This localized approach to quality control maintains image resolution where needed while reducing overall computational burden.