@nicolesharp100

Hi.

It is possible to strip the colour information of any undesired colour channels but whether that is the same thing or as only getting information from the photo-sites recording a given primary colour or not I couldn't say as there is no such thing as a colour sensor. What you have is a black and white sensor with a Bayer filter in front to filter out the unwanted colour from the visible spectrum from a particular set of photo-sites. When you take out a colour channel from a video it may apply to all the dot matrix and not just the photo-sites that originally gave the image its colour.

However, if you made three copies of the extracted information, one for each colour channel (RGB) them combined them back again on three separate tracks, then you can reconstruct the original image but possibly it won't look exactly the same.

Within MEP colour channels can be muted as shown.

Then the resulting file exported if desired.

Ray.

A more complicated example would be to correct for chromatic aberration from atmospheric dispersion. This would require extracting the red, green, and blue channels (from an MP4 video or from a series of photos taken with an intervalometer) as three separate monochromatic videos, and then recombining the channels into a single color video with the red and blue channels vertically displaced from the green channel (since they become misaligned/dispersed by the atmosphere).

https://skyinspector.co.uk/atm-dispersion-corrector-adc/

Can Movie Studio 2023 do this?

@nicolesharp100

Hi.

I think you are talking about things that should be corrected in camera at the time of recording and not post processing as once information is baked into a file it would be difficult / cumbersome, to correct.

There was a plugin by FXHome called Ignite Enhance Ultra that included such a Chromatic aberration plugin but it has now been withdrawn. Even if you had it you would have to tune each channel by eye.

I don't know if any other company has developed such a plugin.

@nicolesharp100

Hi.

I think you are talking about things that should be corrected in camera at the time of recording and not post processing as once information is baked into a file it would be difficult / cumbersome, to correct.

There was a plugin by FXHome called Ignite Enhance Ultra that included such a Chromatic aberration plugin but it has now been withdrawn. Even if you had it you would have to tune each channel by eye.

I don't know if anyone else has developed such a plugin.

 

 

Chromatic aberration from a lens is different from chromatic aberration in the atmosphere. Correcting atmospheric dispersion should be fairly simple and could be done manually. The atmosphere acts as a prism to separate the different wavelengths of light, so that the red image is positioned above (vertically displaced from) the blue image. If you can extract each wavelength of light from the video, then you can realign the red and blue channels to be aligned with the green channel.

This isn't very different from aligning videos for 3D stereoscopy, except that the alignment is vertical instead of horizontal, and you're combining three color channels instead of left-right channels. If you also rotate the video by 90 degrees, then the atmospheric dispersion becomes a horizontal displacement needing a left-right alignment between the red and blue channels, similar to a stereoscopic video.

I am assuming this is a task that Movie Studio 2023 can handle if the software can extract and then recombine the red, green, and blue channels?

I suppose I should ask directly, if Movie Studio 2023 can extract the red, green, and blue channels from an MP4 color video, can it then recombine the three monochromatic videos back into a color video?

@nicolesharp100

Hi.

I don't think in the way you think you could do it with a DSLR still raw file. Raw files in a photo editing package are working with the raw data of the sensor reading each photo-site as it was recorded and not a compressed conversion like a jpeg where such data has already been mixed and given approximate values, white balance is automatically applied as would any other setting used in camera. All video codecs do this as far as I'm aware. You would need the raw data from the sensor and specialist software to work with it after. At present any encoded file from your camera would be the equivalent of working with a jpeg. It has already been mixed and processed by the camera including sharpening and any white balance information set by your camera when the video is taken. At best you can blanket remove blue, green, or red from all the pixels in the image but not from those that originally only captured the red, green or blue spectrum at each of the photo-sites involved. Some cameras do now allow the capture of a video raw file but the only program I'm aware of capable of handling camera video raw files at the editing stage is the full version of DaVinci Resolve (full version at around £295 /$)

Ray.

@nicolesharp100

Hi.

I don't think in the way you think you could do it with a DSLR still raw file. Raw files in a photo editing package are working with the raw data of the sensor reading each photo-site as it was recorded and not a compressed conversion like a jpeg where such data has already been mixed and given approximate values, white balance is automatically applied as would any other setting used in camera. All video codecs do this as far as I'm aware. You would need the raw data from the sensor and specialist software to work with it after. At present any encoded file from your camera would be the equivalent of working with a jpeg. It has already been mixed and processed by the camera including sharpening and any white balance information set by your camera when the video is taken. At best you can blanket remove blue, green, or red from all the pixels in the image but not from those that originally only captured the red, green or blue spectrum at each of the photo-sites involved. Some cameras do now allow the capture of a video raw file but the only program I'm aware of capable of handling camera video raw files at the editing stage is the full version of DaVinci Resolve (full version at around £295 /$)

Ray.

I know this cannot be done losslessly, but does Movie Studio 2023 have the ability to combine a red video, a green video, and a blue video into a single color video? Same question if you use a monochrome camera (instead of a color camera) to record three separate videos, one with a red filter, one with a green filter, and one with a blue filter.

@nicolesharp100

I wish it were just a case of worrying about how much quality loss loss there was.

It is easy to take away two primary colours from a video as I have shown above and export the file.

It is not as easy to keep the frames aligned once those rendered files are placed back into the video editor nor will combining and adjusting the level of each file as it overlays to give the same degree of brightness and contrast let alone colour balance. This frame drifting is due to the nature of the re-encoding and the way B and P frames work within the GOP structure of a file while rendering, causing the results to vary from frame to frame. It may be possible to stop that 'drift' of frames back and forth by exporting only using all I frames. Something I have not yet tried.

Ray.

@CubeAce You'd need to use uncompressed files or those with intra frame compression?

@ericlnz

Hi Eric.

When I decide to use an all I frame export I normally exceed the bit rate of the original file so the lowest bit rate corresponds to the highest bit rate of the file being processed. That normally give me a constant rate exported file.

So I end up with a file that exceeds the amount of streaming bits of the original files used. So for source files averaging 100Mb/s I end up with an exported file that looks like this.

General
Complete name                            : J:\Full Medway Gaming and Creative Festival 2022.mp4
Format                                   : MPEG-4
Format profile                           : Base Media
Codec ID                                 : isom (isom/avc1)
File size                                : 13.3 GiB
Duration                                 : 13 min 36 s
Overall bit rate mode                    : Variable
Overall bit rate                         : 139 Mb/s
Encoded date                             : UTC 2022-07-24 01:47:17
Tagged date                              : UTC 2022-07-24 01:47:17

Video
ID                                       : 1
Format                                   : AVC
Format/Info                              : Advanced Video Codec
Format profile                           : High@L5.2
Format settings                          : CABAC / 3 Ref Frames
Format settings, CABAC                   : Yes
Format settings, Reference frames        : 3 frames
Format settings, GOP                     : N=1
Codec ID                                 : avc1
Codec ID/Info                            : Advanced Video Coding
Duration                                 : 13 min 36 s
Bit rate                                 : 139 Mb/s
Width                                    : 3 840 pixels
Height                                   : 2 160 pixels
Display aspect ratio                     : 16:9
Frame rate mode                          : Constant
Frame rate                               : 50.000 FPS
Color space                              : YUV
Chroma subsampling                       : 4:2:0
Bit depth                                : 8 bits
Scan type                                : Progressive
Bits/(Pixel*Frame)                       : 0.336
Stream size                              : 13.2 GiB (100%)
Language                                 : English
Encoded date                             : UTC 2022-07-24 01:47:17
Tagged date                              : UTC 2022-07-24 01:47:17
Codec configuration box                  : avcC

Audio
ID                                       : 2
Format                                   : AAC LC
Format/Info                              : Advanced Audio Codec Low Complexity
Codec ID                                 : mp4a-40-2
Duration                                 : 13 min 35 s
Bit rate mode                            : Variable
Bit rate                                 : 192 kb/s
Channel(s)                               : 2 channels
Channel layout                           : L R
Sampling rate                            : 48.0 kHz
Frame rate                               : 46.875 FPS (1024 SPF)
Compression mode                         : Lossy
Stream size                              : 19.0 MiB (0%)
Language                                 : English
Encoded date                             : UTC 2022-07-24 01:47:17
Tagged date                              : UTC 2022-07-24 01:47:17

I find there is no upper of lower bit rate readings just the one 139Mb/s. Normally the file is identical position wise frame for frame to the original. Only the changes I make to the look of file is changed. I think that could work but haven't checked. I have a busy day today and tomorrow but may try later in the week. How this will work with the diffraction problem I'm not at all sure as a filter will not give the same result as being able to access a raw file format. For one thing you have gone from a possible sensor data rate of 16 bits (direct from some sensors but admittedly some are less) down to 8 bits per colour channel.

Then I'm also not sure that this process will give an accurate readout for each channel and has to be layered and blended.

There may be other ways of doing this, I'm not into astro photography or videography so I don't know, but there must be dedicated websites for this or programs that can handle such things. I know programs like DaVinci Resolve and Grass Valley's Edius programs can handle camera video raw files but then it all gets further down the the very dark rabbit hole of digital video processing at expert levels and frankly above my experiences.

Ray.