Applies the abnormal pixel correction as described in the article. (Since 3.8)
Images must be IM_GRAY. Source and Target must have the same datatype, and complex is not supported.
image_abnormal is optional, can be NULL. If not NULL, must be IM_BINARY and it will store the abnormal pixels distribution.
Can be done in-place.
threshold_percent is the percentage of the height that must have abnormal pixels candidates.
threshold_consecutive is the minimum number of consecutive abnormal pixels candidates to be considered an abnormal range. (usually the longest vertical ground feature in pixels)
- Based on "Detection and Correction of Abnormal Pixels in Hyperion Images" from T. Han, D. G. Goodenough, A. Dyk, and J. Love
im.AbnormalHyperionCorrection(src_image: imImage, dst_image: imImage, threshold_consecutive, threshold_percent: number[, image_abnormal: imImage]) [in Lua 5]
im.AbnormalHyperionCorrectionNew(src_image: imImage, threshold_consecutive, threshold_percent: number[, image_abnormal: imImage]) -> new_image: imImage [in Lua 5]