An Implementation of the Exposure Fusion Algorithm

Hessel, Charles

doi:10.5201/ipol.2018.230

Cited by 11 publications

(20 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To do so, the dynamic range of each input image is distributed in several new images created from it. To this aim, the authors use the remapping function g defined in Equation (2). Its only parameter is the restrained range width β ∈ (0, 1].…”

Section: Sequence Extensionmentioning

confidence: 99%

“…Algorithm 1 describes the whole process, from the quality measurements to the multiscale fusion. We refer to [2] for pseudo-codes of the downsampling and upsampling procedures.…”

Section: Exposure Fusionmentioning

confidence: 99%

“…This final step over-stretches the colors to [0, 1] by allowing a percentage of clipping (typically 1%) in both sides of the histogram. This operation is called Robust Normalization [2,3]. We refer to [2] for a pseudo-code of this algorithm.…”

Section: Robust Normalizationmentioning

confidence: 99%

“…The EF algorithm, contrarily to most HDR imaging techniques, does not create an intermediate HDR image but directly constructs the final LDR image by seamlessly fusing the best regions of the input sequence, using the Laplacian pyramid. A precise description of the method along with its two main limitations, namely, an out-of-range artifact and a low-frequency halo, can be found in [2].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Extended Exposure Fusion

Hessel¹

2019

Image Processing on Line

Self Cite

View full text Add to dashboard Cite

Extended Exposure Fusion (EEF) is a high dynamic range imaging technique. It was recently proposed as an improvement of an earlier technique called Exposure Fusion (EF), itself widely used thanks to its very good results. But EF has two well-known artifacts: an out-of-range artifact and a low-frequency halo. The extended version solves them, and delivers fused results with enhanced contrast everywhere in the image. We give in this paper a precise description, analysis and implementation of the extended exposure fusion. We notably verify that the artifacts are removed and that the local contrast is improved compared to the original exposure fusion. Source Code The Matlab/Octave source code, the code documentation, and the online demo are accessible at the web page of this article 1. Usage instructions are included in the README.txt file of the archive.

show abstract

Section: Sequence Extensionmentioning

confidence: 99%

“…Algorithm 1 describes the whole process, from the quality measurements to the multiscale fusion. We refer to [2] for pseudo-codes of the downsampling and upsampling procedures.…”

Section: Exposure Fusionmentioning

confidence: 99%

Section: Robust Normalizationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Extended Exposure Fusion

Hessel¹

2019

Image Processing on Line

Self Cite

View full text Add to dashboard Cite

show abstract

“…Since its publication, this method has received considerable attention, being both effective and efficient. A precise description of the method along with its main limitation, namely, an out-of-range artifact can be found in [7].…”

Section: Introductionmentioning

confidence: 99%

Simulated Exposure Fusion

Hessel¹

2019

Image Processing on Line

Self Cite

View full text Add to dashboard Cite

Simulated Exposure Fusion (SEF) is a single-image contrast enhancement method. It is built upon a high dynamic range imaging technique called Exposure Fusion (EF), introduced in 2007 and widely used since then, which aims at fusing a bracketed exposure sequence into a high quality image. Simulated Exposure Fusion extends the initial method to the case where only one image is available, and delivers an image with enhanced contrast. We propose in this paper an implementation of this method, along with its precise description and analysis. Its results are compared to state-of-the-art enhancement algorithms and appear to be artifact-free, even in extreme enhancement conditions. Furthermore, they inherit from EF's celebrated natural aspect. Source Code The Matlab/Octave source code, the code documentation, and the online demo are accessible at the web page of this article 1. Usage instructions are included in the README.txt file of the archive.

show abstract

Enhancement of Region of Interest from a Single Backlit Image with Multiple Features

Yadav,

Mouli

2021

Communications in Computer and Information Science

View full text Add to dashboard Cite

Backlit images are a combination of dark and bright regions and the objects in the image generally appear to be dark for human perception. The region of interest (ROI) in general confines to the object(s) present in the image or some regions of the image. Such ROI in backlit images have low contrast and it is difficult for visualization. Enhancement of ROI in backlit images is necessary in order to view the contents properly. In this paper, a novel and simple approach for the enhancement of ROI of backlit images is proposed. This approach considers several features including tone mappings, exposedness, gradient, median filtering, etc. and finally, the fusion of the results has been done. The novel contribution in the proposed method, though seems to be trivial, attained best results without applying pyramid based operations namely Laplacian pyramid and Gaussian pyramid. Efficacy of the proposed method is evident from the experimental results which confirm that the proposed approach gives better results both qualitatively (visualization) and quantitatively (objective evaluation) compared to the existing methods.

show abstract

An Implementation of the Exposure Fusion Algorithm

Cited by 11 publications

References 9 publications

Extended Exposure Fusion

Extended Exposure Fusion

Simulated Exposure Fusion

Enhancement of Region of Interest from a Single Backlit Image with Multiple Features

Contact Info

Product

Resources

About