Figure 1 | Scientific Reports

Figure 1

From: Impact of image compression on deep learning-based mammogram classification

Figure 1

Images with different compression ratios (CRs). Subfigures (b)∼(k) are the same patched image corresponding to the area demarcated with a white rectangle of in the original image (i.e. subfigure (a)) compressed using different CRs. The caption of each subfigure includes both the CR and peak signal-to-noise ratio (PNSR). For example, the patched image tagged with the caption CR1K (64) has a CR of 1 K and PNSR of 64 dB.

Back to article page