[Retracted] An Improved Image Steganography Framework Based on Y Channel Information for Neural Style Transfer

Table 1

The mean values of PSNR and SSIM of each scheme.

Evaluation indexes		Scheme 1 (only gray-scale image)	Scheme 2 (gray-scale image and Y channel)	Scheme 3 (gray-sale image with Y channel and structural loss)
Between original secret images and extracted images	PSNR	11.4829	13.0054	24.1432
	SSIM	0.5438	0.7529	0.8027
Between cover images and stego images	PSNR	31.7026	31.3511	26.0245
	SSIM	0.8976	0.8150	0.7850

The bold fonts indicate that the corresponding schemes have the highest PSNR and SSIM values, that is, the secret images extracted by Scheme 3 are most similar to the original images. And the values between cover images and stego images of Scheme 1 is the highest, but the visual effects of three schemes are acceptable since the stylized stego images are almost the same to each other.