Adding text to JPEG images

[rnmerchant]

I have been wondering about adding text to an image (useful, perhaps, with old historical family photos to identify people).

2 questions, perhaps not of great import. However.... Insofar as I understand every time one saves a JPG image the image is 'degraded' a bit further. Is this important enough an effect that one shouldn't use the 'add text' feature rather than using exif fields....? Mind you these images are just family interest and not particularly historically vitally important, so a small amount of degredation may not be that big a deal...

Bringing up the Text dialogue, it seems to default to MS Shell Dlg 2. This is a font I've never heard of. Is there anything important about this default?

Thx

[cday]

I have been wondering about adding text to an image (useful, perhaps, with old historical family photos to identify people).

I understand every time one saves a JPG image the image is 'degraded' a bit further. Is this important enough an effect that one shouldn't use the 'add text' feature rather than using exif fields....? Mind you these images are just family interest and not particularly historically vitally important, so a small amount of degredation may not be that big a deal...

No reason for concern in any normal situation, but understand that the text added can't be edited or removed to restore the original image, so best to keep a copy of the original images if there is any possibility of needing to use them again.

Bringing up the Text dialogue, it seems to default to MS Shell Dlg 2. This is a font I've never heard of. Is there anything important about this default?

Use any font you think suitable...

Edit:

An option which might be worth considering would be to place the added text on new image canvas, for example below the image. When captioning a series of images the new canvas could often be added as a batch operation.

Advantages would include the text possibly being easier to read if placed on a white background rather than directly on the images, and the fact that if required the added text could be removed later by cropping the image, or for that matter replaced by new text if the existing text is first removed by pasting background colour over it.

Note, however, that whichever method is used good quality text will require that images have reasonable pixel dimensions, however when that is not the case there may be some possibility of resizing images to larger pixel dimensions by a factor such as two, if the quality of the resulting images is acceptable

[michel038]

Another method :
- Install "Face tag" add-on ( https://www.xnview.com/wiki/index.php?t ... _Face_tags )
- ( you can enable settings>metadata>"Import xmp subject..." and "import xmp face tag" options to get "People" in categories filter up-to-date )
- Select images in Browser
- Metadata>Face tag
- Add names on people

To view names, open a photo in View mode
Enable "View>Show People's regions" option

Then you can get a screen capture with "Capture" tool, and save it as a copy of original.

LaurelHardy.jpg

.
You can use categories filter, People to find anyone in your tagged photos

[XnTriq]

• BetterJPEG
  “BetterJPEG is a JPEG editor specifically designed to avoid recompression loss when editing and resaving JPEG images. BetterJPEG takes advantage of the fact that JPEG images consist of a number of small independent blocks and does not recompress blocks that don't have to be changed.”
  • Putting EXIF info / custom text on the image
• RealWorld Photos
  “RealWorld Photos avoids the incremental quality loss caused by repeated .jpg re-saving.
  Unchanged areas of a photograph are not re-compressed during saving. They are taken directly from the original image. The whole process is completely transparent to the end user. It just works! You can even use layers and save the intermediate result in .rli format and still get the benefits of lossless re-saving when producing the final .jpg.”
  • Lossless JPEG Saving
    “A JPEG image is divided into tiles (called MCUs - minimum compression units) and each tile is processed separately. The size of the tiles is 8x8 pixels, but JPEG can save chrominance (hue) information at smaller resolution and therefore a 8x8 pixels of chrominance information can effectively affect for example 16x16 pixels.
    The JPEG codec in RW apps detects which tiles were not changed since the original JPEG image was opened and if such tiles are found, they are copied from the original image without any loss.”
• XnView Forum
  • JPEG Lossless Join
  • Labels for a beginner
  • Basic understanding of “JPEG Quality”?

[Brother Gabriel-Marie]

Adding text to images makes XnView an easy option for creating memes!
However, is it possible to have some extra features in that regard? Or do we have an addon for MP that will work?

Text Outlines - so we can use yellow letters over a white background
Drop Shadow - well, because everyone likes them and they make things easier to see

Those would be fabulous.

[Brother Gabriel-Marie]

Oh, sorry, I found the Image > Draw function... It actually has both outline AND shadow.
I prefer the toolbar method via Edit > Add Text however. The Draw panel is a little clunky.

[cday]

Oh, sorry, I found the Image > Draw function... It actually has both outline AND shadow.
I prefer the toolbar method via Edit > Add Text however. The Draw panel is a little clunky.

@user0:

Yes, if you only need one 'Draw' action.

[XnTriq]

Bringing up the Text dialogue, it seems to default to MS Shell Dlg 2. This is a font I've never heard of. Is there anything important about this default?

It's a logical font used by default for dialogs and controls (user interface) in Windows.

• Microsoft Learn: Using MS Shell Dlg and MS Shell Dlg 2
• Qt Project: Windows: Default to MS Shell Dlg 2

[rnmerchant]

Bringing up the Text dialogue, it seems to default to MS Shell Dlg 2. This is a font I've never heard of. Is there anything important about this default?

It's a logical font used by default for dialogs and controls (user interface) in Windows.

• Microsoft Learn: Using MS Shell Dlg and MS Shell Dlg 2
• Qt Project: Windows: Default to MS Shell Dlg 2

Thanks for this reference. It makes sense. I really should have just searched google rather than bothering busy other people!

Regards Richard

[rnmerchant]

• BetterJPEG
  “BetterJPEG is a JPEG editor specifically designed to avoid recompression loss when editing and resaving JPEG images. BetterJPEG takes advantage of the fact that JPEG images consist of a number of small independent blocks and does not recompress blocks that don't have to be changed.”
  • Putting EXIF info / custom text on the image
• RealWorld Photos
  “RealWorld Photos avoids the incremental quality loss caused by repeated .jpg re-saving.
  Unchanged areas of a photograph are not re-compressed during saving. They are taken directly from the original image. The whole process is completely transparent to the end user. It just works! You can even use layers and save the intermediate result in .rli format and still get the benefits of lossless re-saving when producing the final .jpg.”
  • Lossless JPEG Saving
    “A JPEG image is divided into tiles (called MCUs - minimum compression units) and each tile is processed separately. The size of the tiles is 8x8 pixels, but JPEG can save chrominance (hue) information at smaller resolution and therefore a 8x8 pixels of chrominance information can effectively affect for example 16x16 pixels.
    The JPEG codec in RW apps detects which tiles were not changed since the original JPEG image was opened and if such tiles are found, they are copied from the original image without any loss.”
• XnView Forum
  • JPEG Lossless Join
  • Labels for a beginner
  • Basic understanding of “JPEG Quality”?

XnTriq - thanks for your detailed and helpful response. This is what makes message boards so great!

I'm not going to worry about it too much but inasmuch as you refer to other programs (which don't) may I draw the inference that XnView actually does 'save as new jpeg' and thus lose a bit of the image quality when adding text? (These are 50-80 year old family photos - a small amount of image loss is not a disaster - but others might have greater concerns that me hence the clarification request...)

Regards Richard

[cday]

I'm not going to worry about it too much but inasmuch as you refer to other programs (which don't) may I draw the inference that XnView actually does 'save as new jpeg' and thus lose a bit of the image quality when adding text? (These are 50-80 year old family photos - a small amount of image loss is not a disaster - but others might have greater concerns that me hence the clarification request...)

I would refer you to my original response, that there is no reason for concern in any normal situation...

Regarding saving, when an image is edited in any way and then resaved, a new image is necessarily created.

However, if the images you wish to caption were or will be produced by scanning, there may potentially be quality considerations there, related for example to scan resolution and the pixel dimensions of the image files produced, other scanner settings and any subsequent image enhancement.

[XnTriq]

I really should have just searched google rather than bothering busy other people!

No worries, Richard! It's a justified question. I didn't know what a "logical font" is, so I learned something (-:

[XnTriq]

may I draw the inference that XnView actually does 'save as new jpeg' and thus lose a bit of the image quality when adding text?

Yes, with the exception of Metadata → JPEG lossless transformations, XnView re-compresses/re-encodes newly-created JPEGs. The extent of the so-called generation loss depends on the settings in Tools → Settings → Formats → Write → JPEG.

These are 50-80 year old family photos - a small amount of image loss is not a disaster - but others might have greater concerns that me hence the clarification request...

Are you going to add text in batch or one by one? Can you share a sample?

• XnView Forum
  • a picture problem for help
  • Image Compare - add 'find difference' feature
  • JPEG processing question
• Kornel Lesiński
  • How to compare images fairly
  • RGBA Structural Similarity

[rnmerchant]

Thanks XnTriq:

Just coming back to this question

Thanks for the clue about the Jpeg write settings! I'd not seen that. I've put the quality to 100% now (it was defaulted to 80%) so I'm happy to know about that option!

Regarding my 'adding text to images' question I had imagined just adding simple descriptors to images one by one. Names, dates, and so on. The idea of adding a slightly larger 'image canvas' has appeal for that little task.

I think that solves my questions! Thanks to everyone.

Richard