Basic sharpen filter

[maxgaard]

"Basic sharpen filter" creates a thin line of varies colors at the bottom border of the image and this shows especially clear when using "full screen" mode. :mad:

[Pantuflo]

Please, try to follow the forum rules ([Only registered and activated users can see links]) as possible.

This is not the way to report a possible bug and this kind of posts will surely be ignored.

What about... posting Ctrl+Tab info, uploading a screenshot or a video sample, giving us some information...?

[z]

I've written several things about this and don't understand why you guys still do not believe my words. Basic/rare filters of audio/video are the result of the author's study (electronic engineering) in his undergraduate days (almost 10 years ago). The KMPlayer started as an analyzing tool of video signal and the basic/rare filters will not be improved anymore. That's the reason why the player has the plugin section of video dsp unlike winamp. I strongly insisted on the removal of basic filters when KMP began to evolve into the player more than which it had been but, as you can guess, it might have been hard to get rid of the days and memories of his strenuous study.


Try advanced sharpen filter.

The default is Asharp ([Only registered and activated users can see links]) filter (more advanced) but you'll be pleased with unsharp masking sharpen filter if you prefer the basic sharpen filter. Once selected, it is possible to toggle the advanced sharpen filter by "alt+s". The terminologies used in KMP keep the standard of most applications not to confuse users. In almost all the cases, it may be posssible to find a valid information by googling. Before searching, pay attention to the explanation described in each section of the preferences. It is also self-explanatory, however, many users just overlook and ignore it complaning the lack of information.

Remember that the tree of the preference is very well organized to explain the category in which each feature belongs to and works.

description from ffdshow, there are still many explanations on internet.
xsharpen
[Only registered and activated users can see links]
This filter performs a subtle but useful sharpening effect. It operates by running a small window over the frame and each center pixel is either passed through untouched (depending upon a threshold setting), or mapped to either the brightest or dimmest pixel in the window, depending upon which is nearest to the center pixel. If the center pixel is mapped, it is also alpha-blended with the original pixel value using a configurable strength setting. The result is a sharpening effect that not only avoids amplifying noise, but also tends to reduce it.

Unsharp mask
[Only registered and activated users can see links]
This filter implements a well-known contrast enhancement/sharpening technique called unsharp masking. The basic idea is that a blurred version of the original image is created and then subtracted from the original image. The resulting image will have contrast boosts around features in the image. The result is a very pleasing sharpening that is superior to simple sharpening with a convolution kernel or high-pass boost.

msharpen
[Only registered and activated users can see links]
[Only registered and activated users can see links]
MSharpen is designed specifically for anime, it also works well with normal video. The filter is very effective at sharpening important edges without amplifying noise. The filter operates by detecting edges and then applying a sharpening kernel only in the edge areas.

Descriptions are copied from original AviSynth and VirtualDub filters documentation (C) Donald Graft


[Only registered and activated users can see links]
[Only registered and activated users can see links]

asharp is an adaptive sharpening filter and was written by MarcFD. The filter is based on the popular Unsharp Masking, a very powerful method of sharpening. But in contrast to simple Unsharp Masking, asharp allows adaptive sharpening. This means asharp can sharpen the video without amplifying noise! Note that this filter has been optimized for quality, not speed.