Hi,
Jpg image format is a lossy format I would suggest png or tiff they are lossless images, In video editing that will give the best display (my opinion any way). video editing the bitrate determines the quality (another of my opinions).
Jim
p.s. I am very opinionated... Intel i7-2600@3.4Gz Geforce 560ti-1GB Graphic accelerator, windows 7 Premium 12GB memory

Visit GranPapa64's channel for your YouTube experience of the day!

Great Jim

yes, bitrate does = quality, but a soft image will look worst after 'production'

PNG looks to be my fav too at this point

Hi max_e_mum -

Firstly, JPEG format. It's absolutely true of JPEG format that its compression causes loss of quality. In some photo editors, opening a JPEG file (say 0001.jpg), without any editing, then saving it as 0002.jpg will cause further compression (& loss of quality). Of course, JPEGs lose quality every time they're opened edited & saved.

Resizing images will ALWAYS have a negative effect on quality, whether it's done in photo or video editing software. As a matter of course, information from the photo gets lost.

If you're referring to photos from your camera, it will typically store files in either/both JPEG &/or RAW format. You didn't mention what camera you were using. Can your camera shoot RAW?

If you have a series of JPEG images, converting them to TIFF, BMP or PNG will achieve very little. It will not improve the quality. The best thing to do is avoid processing them at all.

Max - you're asking a lot of yourself to want to (a) scale down the images by so much & (b) maximise quality. That's a tough ask.

Cheers - Tony
Visit PDtoots. PowerDirector Tutorials, tips, free resources & more. Subscribe!
Full linked Tutorial Catalog PDtoots happily supports fellow PowerDirector users!

Morning ynotfish

The camera is from security industry IP-Camera [I like security style as they can be a good distance from computer and allow lens changes]
So, like most older IP based cams I'm getting jpgs - which look great straight off the camera.
just seems when I open them again they already have jaggies - I've thought that was because of having the wrong codec on the machine - but I'm not sure what codec was used in camera.

Anyways, Its all about preserving as much quality in each image, a format that can be moved around - open and closed and in the final compression inside the video editor compresses down without the jaggies that jpegs can have when resized and compressed.

first rule of video capture and editing is to work at the biggest size, and only reduce the image size at the final encoding.

Part of my thought on the resize is, is to add a better contrast ratio and smooth out a somewhat grainy image [the cam is 5MP and have a C-mount lens running about as wide as 35mm might look, but the 'scene' is complex], a good higher definition vid.

By rights I should be able to put out a super deep looking smooth video.

Hmm - my reply vanished into thin air!

What I suggested was to attach one of the JPEG files straight off your camera so that members can advise about the best way to maximise quality in production.

Cheers - Tony
Visit PDtoots. PowerDirector Tutorials, tips, free resources & more. Subscribe!
Full linked Tutorial Catalog PDtoots happily supports fellow PowerDirector users!

captures are 51 minutes apart
for my little video I chose around 8am as it gave a good contast,
the two I've posted are more of the extremes (I think)

I think the problem is sensor size in the camera. Have not looked at your files but as a photographer who upload a ton of videos to Youtube I cannot see a loss of quality in .jpeg files.

Suspect the images have been captured on a small snsor camera - that would explain the loss in quality? Thanks, George

Without wishing to dampen enthusiasm for high quality, and not being sure what the final output is to be used for/viewed on, perhaps an alternative approach is to research which process/format loses least quality for your specific needs.

There is no doubt, using consumer software, that reducing resolution will lose quality. But is it best to do that with a dedicated image editor then import 640x360 images to PD or import full images and let PD do it's best (or worst)?

Purely as an experiment, I took one of your images and reduced it to the smaller size in PSE8 output as a .png.

Obviously blowing them both up to the same original physical screen size shows the dramatic difference in visual appearance.

I then imported both the full size and reduced size images into PD and ran each one at 1 sec intervals for 1 sec so that they alternated. In the preview the difference was observable as a blinking effect as the hi-res/lo-res alternated.

However, I then produced the 2 tracks using the ipod mp4 profile for 640x360 and to my surprise the blinking effect was no longer observable. I was so surprised that I did it 3 times.

My logic may be flawed but I think that it means that the hi-res image was resized by PD to 640x360 and became worse but the lo-res image did not seem to worsen any further.

Additionally, when viewing the original hi-res image and a paused lo-res video - both at the 640x360 physical screen size - it seemed to me to be difficult to discern noticeable differences?

The lo-res .mp4 is attached for reference.

Cheers
Adrian



Life is really simple, but we insist on making it complicated. (see below)
Confucius
AMD Phenom IIX6 1055T, win10, 5 internal drives, 7 usb drives, struggling power supply.

Hi max_e_mum -

Adrian - it seems you & I were chasing similar tails! If your logic is flawed, then mine is too. It's hard to spot any difference in the produced file... probably more to do with the low-res output.

It also appears that the EXIF data from the attached photos has been stripped. Are they straight off the camera? or have they been processed in some way? Neither LightRoom nor PhotoDirector displayed the data. Attached is all the information EXIFtool could extract, plus an example of data from one of my cameras' photos. That's just by-the-by. I was trying to follow up George's suggestion about the camera.

Histograms of the photos show some quite badly overexposed areas. If it were me, I think I'd be correcting that prior to importing into PD.

So - to give the 640x360 video its best chance of looking good, I think it would be well worth resizing & correcting the images before production.

Cheers - Tony




Visit PDtoots. PowerDirector Tutorials, tips, free resources & more. Subscribe!
Full linked Tutorial Catalog PDtoots happily supports fellow PowerDirector users!

Morning Chaps

The camera is a box-type CCTV with changable lenses.
It's primarily a Video Camera, well sort of

The sensor is CMOS - it was better I think - it may have been damaged on another shoot ...?
it seems to have lost some fine clarity and makes much more noise at night.

There are better lenses available
I own a Cosmicar TV Zoom Lens ES 12.5 ~ 75mm 1:1.8
Its 1/2" like the camera, but the Camera maker decided to do DC Iris whereas the Cosmicar is Video Iris
I've found converters for DC to Video, but not the other way around

The picture that was being shot is soft,
I've re-focused,
cleaned the lens
dropped some contrast which has stopped the crushing of dark colours around the shadows
upped the saturation
and on my 20" Dell IPS it looks correct


Of course all in camera settings have to work for the whole day, I set the capture at every 51mins. My first production used shots from around 8am

changing the images to png does create bigger files, but it does stop the jaggies from forming and thus keeps the contrast cleaner. and might just compress cleaner too.

PS - I did start this at AM

Basically you have a 4mp sensor with what looks like not a great lens. The image is not sharp and suffers from colour shifts along the image edges. You see green edges running along white objects within the image etc.

You can correct all this within photoshop (to a certain extent) and enlarge the image with other softtware (eg OnOne). Not sure if all that is worth the effort as buying a new camera would probably solve your problem.

I have it as a 16mb file now and the flaws are fairly obvious - as listed above. No (little) dynamic range within the image either so the shadows will block up or the highlights will be blown - as in your image.

Saving it as another format will help, but it will not correct the problems.

You appreciate this is only my opinion - someone may have more information to assist you. Thanks, George

Hey georgemon, yes none of the stuff is great, but it is working now. I think you are talking about the earlier images, not todays?

It was today's I downloaded. See no reason why it should need much compression in PD at that size. Thanks, George