Research on the WebP Image Format
WebP is a new image compression technology proposed by Google. It is derived from the VP8 video coding format; it also supports lossless and lossy compressions. Compared with the present popular image format, WebP has smaller volume than JPG and PNG in th
- PDF / 313,813 Bytes
- 7 Pages / 439.37 x 666.142 pts Page_size
- 48 Downloads / 294 Views
Research on the WebP Image Format Zhanjun Si and Ke Shen
Abstract WebP is a new image compression technology proposed by Google. It is derived from the VP8 video coding format; it also supports lossless and lossy compressions. Compared with the present popular image format, WebP has smaller volume than JPG and PNG in the condition of the same picture quality conditions. By coding with MATLAB, this experiment adopts the PSNR and SSIM on WebP and JPG, PNG for comparison test, then implement data compression and reduction test. Since MATLAB does not identify the WebP format, this experiment uses canvas to process WebP images, thus measuring the PSNR and WebP of SSIM. The results of experiment show that WebP has high value in the life, it will develop well in the future. Keywords WebP
35.1
Image compression PSNR SSIM
Introduction
With the rapid development of the Internet, people are placing a greater demand on the quality of the picture. It is necessary to reduce the volume of pictures in the condition of ensuring the fineness. But nowadays, the optimization for JPG, PNG, and GIF has reached the extreme. It needs great courage to make something new and change the current situation, and Google has given us a new choice: WebP. WebP, which is based on VP8 coding, has higher compression ratio than JPG, and it has brought a new power for the Internet [1].
Z. Si (&) K. Shen College of Packaging and Printing Engineering, Tianjin University of Science and Technology, Tianjin, China e-mail: [email protected] © Springer Science+Business Media Singapore 2016 Y. Ouyang et al. (eds.), Advanced Graphic Communications, Packaging Technology and Materials, Lecture Notes in Electrical Engineering 369, DOI 10.1007/978-981-10-0072-0_35
271
272
35.2
Z. Si and K. Shen
Picture Format Evaluation Criterion
35.2.1 Peak Signal-to-Noise Ratio (PSNR) and SSIM PSNR [2] is an objective standard for evaluating images, based on counting and averaging the pixel grey level. Formula is as follows: 2552 PSNR ¼ 10 lg ð35:1Þ MSE This measurement compares the original picture and the rebuilt picture in three aspects: brightness, contrast ratio, and structural similarity index. SSIM ¼ ½lðx; yÞa ½cðx; yÞb ½sðx; yÞc
35.3
ð35:2Þ
Design of Experiments
35.3.1 Contrast of Picture Size The experiment selects a number of PNG images, and transforms these images into WebP, JPG, and PNG whose qualities are 85 respectively. Then measure the size of these images and compare the result with the size of original image (Fig. 35.1). Experimental environment: Windows 7 professional Software: Photoshop CC, iSparta Fig. 35.1 Contrast of picture size
35
Research on the WebP Image Format
273
35.3.2 PSNR Test Experiment Because MATLAB does not recognize the WebP format and we can only use Chrome to open the WebP images, so this experiment uses HTML5 canvas to process these images and write Javascript code to calculate the PSNR value. First, paint the picture on the canvas. Then, we use ‘getImageData’ function to get the pixel information of im
Data Loading...
