bencardoen / colocalization.jl Goto Github PK

Add support for reference point cloud based metrics, such as CBC (coordinate based correlation)

• Split API in point cloud/voxel based on input type
• Add KDTree for fast point cloud distances
• Add CBC
• Add Ripley's K

Document metrics, especially the ones that are more complex in interpretation (Hausdorff distance)

• Add option in addition to otsu, to use adaptive segmentation in object detection
• Add API to expose choice in segmentation
• Add API to provide parameters

• Do a per-object segmentation
• For each object, record the descriptive stats of coloc metrics
• For each object, compute the distance to nearest object
• For each object, record the size of itself, and the size of the nearest object

Currently only xyz is exported, change the code so the original metadata is exported too