Background: In many applications of image processing, the enhancement of images is often a step necessary for their preprocessing. In general, for an enhanced image, the visual contrast as a whole and its refined local details are both crucial for achieving accurate results for subsequent classification or analysis.
Objective: This paper proposes a new approach for image enhancement such that the global and local visual effects of an enhanced image can both be significantly improved.
Methods: The approach utilizes the normalized incomplete Beta transform to map pixel intensities from an original image to its enhanced one. An objective function that consists of two parts is optimized to determine the parameters in the transform. One part of the objective function reflects the global visual effects in the enhanced image and the other one evaluates the enhanced visual effects on the most important local details in the original image. The optimization of the objective function is performed with an optimization technique based on the particle swarm optimization method.
Results: Experimental results show that the approach is suitable for the automatic enhancement of images.
Conclusion: The proposed approach can significantly improve both the global and visual contrasts of the image.
Keywords: Image enhancement, global visual effects, local visual effects, normalized incomplete beta transform, optimization, particle swarm optimization.
[http://dx.doi.org/10.1109/42.363111] [PMID: 18218535]
[http://dx.doi.org/10.1109/83.908502] [PMID: 18249627]
[http://dx.doi.org/10.1109/TIP.2016.2639450] [PMID: 28113318]
[http://dx.doi.org/10.1109/TIP.2018.2810539] [PMID: 29570085]
[http://dx.doi.org/10.3390/s20020495] [PMID: 31952325]
[http://dx.doi.org/10.1109/TPAMI.2010.161] [PMID: 20733228]