logiWIN - What is de-interlacing and what de-interlacing methods can be applied on interlaced video input
What is de-interlacing and what de-interlacing methods can be applied on interlaced video input?

ITU input video standards (ITU656 and ITU1120) support only interlaced scan video, unlike RGB or YUV, which support only progressive scan video. ITU input video standard can encode resolution and refresh rate of both PAL and NTSC video standards.

The process of producing half-resolution fields (odd and even) at double frame rate is known as interlacing and it's a feature of both PAL and NTSC signals. Interlaced scanning produces visual defects called interlace artefacts or combing, which occur when objects in the image are in motion. The process of converting interlaced video to the one which consists of full frames and does not exhibit interlace artefacts is called de-interlacing.

Default de-interlace mode for logiWIN is bob de-interlace. This algorithm does not make any differences between fields. logiWIN displays every field, one after the other, and there is no loss of information.  Fields should be stored in the same block of video memory (set the same address in both MEM_OFFSET_ODD and MEM_OFFSET_EVEN register). Each interlaced frame is split into 2 frames (= the 2 former fields) half the height. If full resolution is wanted then the input image should be vertically scaled with scaling factor 2.

logiWIN can be programmed to use weave de-interlacing algorithm. When in weave mode, if odd and even fields are stored in the same block of video memory, then the image is in full resolution but with artefacts if motion is displayed.

For both de-interlace modes, odd and even fields can be stored in different blocks of video memory. This option is used for advanced de-interlace algorithms. Since logiWIN separated odd and even fields in video memory, CPU can fetch each of them and process according to the chosen algorithm.


Additional Comments

Copyright © 2017 Xylon d.o.o.