A multi-dimensional image information fusion algorithm based on NSCT transform

The depth picture is in keeping with human imaginative and prescient, however typically, the goal can’t be fully distinguished from the background. Polarization photographs can distinguish the goal extra successfully and spotlight the contour and texture particulars, though it doesn’t conform to human visible notion.

By using picture fusion methods, these two varieties of photographs will be mixed to disclose multi-dimensional options successfully. This fusion course of compensates for the constraints of data obtained from a single picture sensor, offering extra dependable and correct goal info.

Researchers led by Prof. Ming Zhao at Huazhong University of Science and Expertise (HUST), China, are curious about a picture fusion algorithm that fuses two photographs with totally different info dimensions into one picture.

Their thought is to preprocess the polarization picture and visual picture, decompose the picture into high-frequency sub-bands and low-frequency sub-bands by the nonsubsampled contourlet rework (NSCT), fuse the sub-bands in line with the fusion guidelines of the designed preserved edges, and at last acquire the fusion picture by NSCT inverse transformation.

The researchers predict potential purposes, comparable to utilizing fused images for electrical grid video surveillance, that would enable targets in some complicated environments to be highlighted.

The paper is published within the journal Frontiers of Optoelectronics.

Supplied by
Increased Education Press