Arbitrary ratio image resizing in the dct domain

Abstract

A method and system for arbitrary resizing of a compressed image in the discrete cosine transform (DCT) domain. First and second scaling parameters P and Q a re determined in accordance with the block numbers L and M and the scaling rati o L / M or M / L. A non-uniform, or uneven, sampling in the DCT domain is then applied to coefficients of successive blocks in the compressed image in accordance with the scaling parameters. In an embodiment, P blocks are sampled and transformed from a given block length to |P| according to a |P|-point IDCT, while Q blocks are sampled and transformed from a given block length to |Q| according to a |Q|-point IDCT. The non- uniformly sampled and transformed pixel domain samples are then regrouped into a predetermined block size and transformed back to generate the DCT coefficient of the compressed image. The proposed method significantly reduces the computational complexity compared with other DCT domain arbitrary ratio image resizing approach. It also facilitates the practical implementations since only the fast implementations of IDCT with length N (1 < N < 9) and DCT with length 8 are required.

Description

technical field [0001] The present invention generally relates to image or video frame compression. More specifically, the invention relates to resizing images at arbitrary ratios in the frequency domain. Background technique [0002] The spatial resolution of the source image or video frame is often larger than the screen size of the handheld device. Therefore, image downscaling or video transcoding with spatial resolution downsampling is usually performed on the server side of the service provider to reduce the spatial resolution of the source image or video frame to fit the display screen of the terminal device. Resizing an image or video frame is usually performed in the pixel or spatial domain by a low-pass filtering operation followed by downsampling. However, for JPEG images, and video frames that have been formatted according to common standards such as the H.26x series of standards and the MPEG-x series of standards, the images and video frames have been in a comp...

AI Technical Summary

Problems solved by technology

[0005] However, there are still two problems associated with these previously proposed algorithms for resizing at arbitrary ratios in the DCT domain

First, they are still computationally expensive

Second, the two methods are difficult to implement for practical applications

The second method requires fast execution of DCT and inverse discrete cosine transform (IDCT) operations at every possible length, which is non-trivial even if some fast implementations exist for synthetic lengths

Images