Lossless compression method applicable to Bayer image format

Abstract

The invention discloses a lossless compression method applicable to the Bayer image format, which comprises the following steps: carrying out correlation evaluation and calculation by separating a current pixel with the above part by a line of pixels and separating the current pixel with the left side by a column of pixels, adopting two storage lines of line 0 and line 1 to respectively store odd-numbered lines and even-numbered lines of pixel values, and using the line 0 to carry out the correlation evaluation if the current pixel is in the even-numbered line; using the line 1 to carry out the correlation evaluation if the current pixel is in the odd-numbered line; that is, giving corresponding values to variables of N1 and N2, then calculating L which is equal to min (N1, N2) and calculating H which is equal to ma multiplied by (N1, N2), and determining the encoding method according to the distribution position fallen by the current pixel value P. The lossless compression method does not need to carry out pre-treatment, can overcome the calculations of three-color channel decomposition or image structure conversion and the like, inherits the characteristics of fastness, high efficiency and capability of being applicable to physical realization with low power consumption and small area of the FELICS method, and has good compression effect.

Description

technical field [0001] The invention relates to an image compression method, in particular to a lossless compression method suitable for Bayer image format. Background technique [0002] At present, the vast majority of CCD and CMOS image sensors use a filter arrangement format called Bayer to record the three primary colors (red, green, blue, RGB) information of the image. This pattern is a red-green filter interval Alternate arrangement of rows and green and blue filters at intervals (such as figure 1 shown), so the Bayer format is the most common data output format for CMOS and CCD image sensors. [0003] Although the Bayer format can save image transmission bandwidth and storage space compared with the full-color format, it is still necessary to compress the image in the Bayer format in some cases to further reduce the image size (such as low-power and small-volume embedded image sensors, or image sensors with resolutions up to millions or even tens of millions of pixe...

AI Technical Summary

Problems solved by technology

Obviously, from theoretical analysis, it can be seen that Figure 4 This method of assessing pixel-to-pixel correlation cannot be applied to figure 1 Bayer format images in , because figure 1 Any pixel and its adjacent pixels at the Top and Left positions are all non-same-color pixels, and the correlation is naturally very poor; a large number of actual experiments have also shown that FELICS has a poor compression effect on Bayer images, such as Figure 5 The image shown is the test object, the compression ratio is reduced from 3.3708 in the normal RGB format to 1.1321 in the Bayer format

It can be seen that the FELICS compression method is not suitable for Bayer format images

Images