<title>Comparison of multiple compression cycle performance for JPEG and JPEG 2000</title>

Abstract

Ideally, when the same set of compression parameters are used, it is desirable for a compression algorithm to be idempotent to multiple cycles of compression and decompression. However, this condition is generally not satisfied for most images and compression settings of interest. Furthermore, if the image undergoes cropping before recompression, there is a severe degradation in image quality. In this paper we compare the multiple compression cycle performance of JPEG and JPEG2000. The performance is compared for different quantization tables (shaped or flat) and a variety of bit rates, with or without cropping. It is shown that in the absence of clipping errors, it is possible to derive conditions on the quantization tables under which the image is idempotent to repeated compression cycles. Simulation results show that when images have the same mean squared error (MSE) after the first compression cycle, there are situations in which the images compressed with JPEG2000 can degrade more rapidly compared to JPEG in subsequent compression cycles. Also, the multiple compression cycle performance of JPEG2000 depends on the specific choice of wavelet filters. Finally, we observe that in the presence of cropping, JPEG2000 is clearly superior to JPEG. Also, when it is anticipated that the images will be cropped between compression cycles when using JPEG2000, it is recommended that the canvas system be used.

Image coding design considerations for cascaded encoding-decoding cycles and image editing: analysis of JPEG 1, JPEG 2000, and JPEG XR / HD Photo

“…The use of the "canvas" system in the JPEG 2000 operation is important to enable such usage, as previously noted 5 . But the basis functions on the lowest frequency coefficients in the JPEG 2000 case can get very long as the decomposition level deepens, so such a scheme might require retaining a substantial number of extra coefficients and using correspondingly wide cropping boundaries in the output.…”

Improving Multi-generation Robustness of Learned Image Compression

“…Multi-generation coding is a process of repeatedly compressing decoded pictures (Horne et al 1996), and the resulting generation loss will have a non-negligible impact on multimedia applications. For traditional coding standards like JPEG and BPG (i.e., HEVC Intra Coding), prior works have shown that inconsistent mode decisions, rounding and clipping (RC) errors are sources of generation loss (Joshi, Rabbani, and Lepley 2000;Wegner et al 2015). Some work (Hurd and Rosiles 2000) devoted to eliminating the generation loss and achieving idempotent encoding in JPEG-LS, but additional bit allocation is needed.…”

“…Another factor that affects multi-generation robustness is the adaptive ability of the model. The researchers showed that error caused by RC operations has little effect on JPEG (Joshi, Rabbani, and Lepley 2000). For LIC that employ nonlinear transformations, the ability to accommodate this error is critical, especially since there is no similarly distorted image in the training set.…”

Better Compression With Deep Pre-Editing

“…Referring to the constraint, recall that the compression-decompression operation is idempotent, i.e. applying it more than once on a given image results with the same outcome as using it once [32]. Thus, the constraint aims to say that x is a feasible outcome of the compression algorithm with the given budget of R bits.…”