A COMPARISON OF SĂLĂGEAN AND RUSCHEWEYH COEFFICIENT-BASED IMAGE ENHANCEMENT MASKS
ASERC 5th INTERNATIONAL CONGRESS ON HEALT, ENGINEERING, ARCHITECTURE and MATHEMATICS, Malatya, Türkiye, 23 - 25 Ekim 2026, ss.145-158, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Basıldığı Şehir: Malatya
- Basıldığı Ülke: Türkiye
- Sayfa Sayıları: ss.145-158
- İstanbul Gelişim Üniversitesi Adresli: Evet
Özet
Derivative operators of geometric function theory change the coefficients of an analytic function through a single parameter, and this makes them useful for designing tunable image masks. Image enhancement masks of size 5×5 were obtained from the Ruscheweyh operator in earlier work. In this paper the same mask structure is built with the Sălăgean operator and the two operators are compared. The Koebe function f1, the extremal convex function f2 and the extremal function f3 of bounded turning are used. To make the comparison fair, both operators are taken at the same inner ring strength t=c2/a2. Two theoretical results are obtained. For 0<t<1 and equal inner ring strength, the outer ring coefficient of the Sălăgean mask is always smaller than that of the Ruscheweyh mask. Under the Sălăgean operator the Koebe function and the convex function produce the same family of masks. Six standard images and three blur levels were used, and 1512 enhancement experiments were run on a 14-point grid of t. On this grid the best results of the two operators were very close. At every blur level the highest mean SSIM was obtained with the Sălăgean f3 mask (0.9494, 0.9048 and 0.8670 for σ=1.0, 1.5 and 2.0). To remove the effect of the grid, t was also optimized continuously for each image and blur level. The Sălăgean masks then gave a higher SSIM and PSNR in all 18 image–blur pairs for each of the three functions (Wilcoxon test, p<0.001), and the mixed results seen on the grid disappeared. The choice of operator matters less than the enhancement strength, but the weaker outer ring gives a small and systematic advantage.
Keywords: Geometric function theory, image enhancement, Sălăgean operator, Ruscheweyh operator, convolution mask, image processing.