Abstract
Image enhancement techniques are very important to image processing, which are used to improve image quality or extract the fine details in degraded images. In this paper, two novel objective functions based on the normalized incomplete Beta transform function are proposed to evaluate the effectiveness of grayscale image enhancement and color image enhancement, respectively. Using these objective functions, the parameters of transform functions are estimated by the quantum-behaved particle swarm optimization (QPSO). We also propose an improved QPSO with an adaptive parameter control strategy. The QPSO and the AQPSO algorithms, along with genetic algorithm (GA) and particle swarm optimization (PSO), are tested on several benchmark grayscale and color images. The results show that the QPSO and AQPSO perform better than GA and PSO for the enhancement of these images, and the AQPSO has some advantages over QPSO due to its adaptive parameter control strategy.
Highlights
Image enhancement is one of the important image techniques in low-level image processing, the purpose of which is to improve the quality of an image for machine analysis or visual perception of human being [1]
The results show that the quantum-behaved particle swarm optimization (QPSO) and adaptive QPSO (AQPSO) perform better than genetic algorithm (GA) and particle swarm optimization (PSO) for the enhancement of these images, and the AQPSO has some advantages over QPSO due to its adaptive parameter control strategy
“Lenna” and “Waterlilies,” are used as benchmarks to evaluate the performance of the QPSO and AQPSO algorithms in color image enhancement
Summary
Image enhancement is one of the important image techniques in low-level image processing, the purpose of which is to improve the quality of an image for machine analysis or visual perception of human being [1]. If grayscale image enhancement is applied directly to the three components (R, G, B) of a degraded color image, they may be prone to produce color artifacts which look very strange for human beings [1, 5,6,7,8] Another color space is HIS which stands for three main attributes: hue, intensity, and saturation, generally used to distinguish one color from another [6, 7]. We propose an adaptive control method for the CE coefficient and used the improved QPSO for the enhancement of grayscale and color images, using two proposed new objective functions for estimating the parameters of the Beta functions.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
