Extending depth of field by intrinsic mode image fusion

Abstract

Here, a versatile data-driven application independent method to extend the depth of field is presented. The principal contribution in this effort is the use of features extracted by Empirical Mode Decomposition, namely Intrinsic Mode Images, for fusion. The input images are decomposed into intrinsic mode images and fusion is performed on the extracted oscillatory modes, by means of weighing schemes that allow emphasis of focused regions in each input image. The fused image unifies information from all focal planes, while maintaining the verisimilitude of the scene. In order to validate the fusion performance of our method, we have compared our results with those of region-based and multiscale decomposition based fusion techniques. Several illustrative examples and objective comparisons are provided.

Multi-focus Image Fusion Based on Sparse Features

“…After initial subdivision, ( 1, , 4 ) k Ik     are corresponding to regions at level 1. At level 1, the first and the third blocks, namely 1 I and 3 I , are subdivided into are further subdivided if they meet the threshold conditions. Other quadrants will be subdivided similarly.…”

“…Then, the focused blocks or regions are integrated to construct final fused image. The spatial domain methods are easy to implement and have low time consumption [1]. However, if the size of the blocks is too small, the blocks selection is too sensitive to noise and is subject to incorrect selection from the corresponding source images.…”

Multi-Focus Image Fusion Based on Sparse Decomposition

“…Image fusion can be defined a process in which a single sharper image is produced by integrating a set of source images captured from the same scene with different focus points [1]. The traditional image fusion methods can be categorized into two groups: spatial domain fusion and transform domain fusion [2].…”

“…Then, the focused blocks or regions are integrated to construct final fused image. The spatial domain methods are easy to implement and have low time consumption [1]. However, the blocking artifacts may compromise the quality of the final fused image.…”

A Novel Pulse Coupled Neural Network Based Method for Multi-focus Image Fusion

“…The spatial domain fusion methods are easy to implement and have low computational complexity, while the spatial domain methods may produce blocking artifacts and compromise the quality of the final fused image. Different from the spatial domain fusion, the transform domain fusion methods can get improved contrast, as well as signal-to-noise ratio and better fusion quality [2], while the transform domain fusion methods are more time/space-consuming to implement.…”