Post-processing of gated images by using seed growing fusion approach

Abstract

This paper introduces an image fusion technique on a gated image. Gated imaging is a powerful technique to capture images in turbid condition. By controlling the time delay of the shutter to capture the returning light, we can obtain the image at different depth. Modular transfer function (MTF) is used to evaluate the quality of the gated images. Images with useful information (clear and sharp objects) are gathered. Then, the images are segmented into squares. Every segment of every image is analyzed using MTF. The segment with the best contrast is selected to be used in the final image. The best of each segment is gathered and are used to reconstruct a new image with most of the useful information preserved in them. Then, post-processing is done on the constructed gated image for a seamless image fusion. Seeded region growing is one of the image segmentation techniques. This technique is utilized to look for the region with the same feature and segment the images into regions. The constructed gated image provides an image with monotonous coloration. Therefore, seeded region growing technique easily identifies the borders where there is abrupt change in color feature. The segment edge is one of these borders. An algorithm is used to check for the edges of the segment. Then, the segment edges are chosen for further processing. Gaussian filter is applied to the rough edge of the segment with obvious contrast. Gaussian filter is useful to blur images so that the color histogram will show a smooth color distribution. After the filter is applied, the rough edges will appear smooth and the objective of seamless image fusion is achieved. The final image has the useful information preserved and the borders of the segment are not obvious.