An adaptive strategy for adjusting contribution of frames in super resolution imaging

Jamalabadi, Marzieh; Ahmadyfard, Alireza

doi:10.1109/iraniancee.2014.6999523

Cited by 1 publication

(3 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since is the closest pixel to in the LR image , we also approximate by . This results into the following equation: (10) and, since , we get to approximate by using a weighted sum over all LR pixels as follows (11) where is the average noise used during the image degradation procedure (cf. (1)).…”

Section: Energy Function Minimizationmentioning

confidence: 99%

“…In the last decade, several multi-frame SR methods have been published including Bayesian methods [6], maximum a posteriori (MAP) [7], regularization-based methods [8]- [10], blind methods [11], wavelets [12], iterated back projection (IBP) [13], and structure-adaptive normalized convolution [14] to name a few. For these methods, SR is an inverse problem for which HR pixel values are recovered by combining LR pixels taken from all LR images.…”

Section: Introductionmentioning

confidence: 99%

“…They showed that HR images recovered after minimizing an energy function which includes this notion of zone of influence are of good quality, especially when the HR image is larger than the LR images. Jamalabadi and Ahmadyfard [10] recently proposed a method which gives different weights to the LR images while Karch and Hardie [16] pushed the idea further by using 3 types of weights: an interpolation weight, a local motion weight, and channel cross-correlation weight.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Novel Graph Cuts Method for Multi-Frame Super-Resolution

Zhang

Jodoin

et al. 2015

IEEE Signal Process. Lett.

View full text Add to dashboard Cite

In this letter, we propose a new graph cuts multi-frame super resolution method. The method is carried out in 3 steps. First, we project each high-resolution pixel onto the low-resolution images and select low-resolution pixels which fall within the zone of influence of . Second, we weigh the contribution of the low-resolution pixels via a soft switching function and add them to construct a virtual low resolution pixel. The high resolution image is then recovered after minimizing a Maximum a posteriori Markov Random Field (MAP-MRF) energy function. This is done by approximating our energy function to make it graph representable and minimize it with a graph cuts -expansion algorithm. Experimental results show that our approach outperforms state-of-the-art methods.

show abstract

Section: Energy Function Minimizationmentioning

confidence: 99%