If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#46
|
|||
|
|||
Does the .png image format have a text metadata field?
In message , Java Jive
writes: On 14/02/2020 13:52, J. P. Gilliver (John) wrote: In message , R.Wieser writes: Java Jive, Paint will load and save PNG files, but know even less about them than Oh, only PNG? I'd thought maybe the W10 version did JPEGs by now. Oh yes, I was taking that for granted! But it still doesn't have any means of displaying, let alone altering, JPEG metadata. If it loads one that already _has_ the metadata though, and subsequently saves it, does it mess it up, as apparently PSP does? -- J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf If a cluttered desk is characteristic of a cluttered mind, what does an empty desk mean ? |
Ads |
#47
|
|||
|
|||
Does the .png image format have a text metadata field?
In message , R.Wieser
writes: [] And by the way: GDI+ (which contains the support for EXIF properties) does not know how to write an EXIF property into a PNG image. "EXIF Pilot" can; it _could_ be proprietary or a special, but it presents it under an EXIF tab, _implying_ it's part of some standard. :-) When you use third-party programs all bets are off. Nothing stops a Of course. Though see below. program from putting or extracting an EXIF (or any other meta-data containing) blob into/from one of the earlier mentioned PNG "chunks". Same goes for other image formats ofcourse As for "some standard" ? Yeah, I guess so. As long as that means "the way that I (the program) did it" :-p Regards, Rudy Wieser So far, I've found (for JPEG files): Windows Explorer (I'm on Windows 7) (not a "third-party" software, I'd venture) can add text (header in Explorer is "Comment:") that IrfanView can see, though not edit (displays it inside IV under EXIF with the header "XPComment"). That's the only heading there (in IV - EXIF). [Try it to see what I mean.] EXIF Pilot can add text that Windows Explorer displays (still under "Comment:" in Explorer; replaces that _in Explorer_ put there earlier using Explorer). If that file is immediately examined in IrfanView, it shows (but still can't edit) under EXIF, _both_ lots of text: that put there earlier using Windows Explorer still shows by "XPComment", whereas that added by EXIF Pilot shows by "UserComment". If I now _edit_ the text in Explorer, and then yet again examine the file in IrfanView, the changed text appears, under "XPComment", with "UserComment" having disappeared - not just the text, but the heading as well. Comment text entered (just under Comment) using IrfanView is displayed by Brother's Keeper (optionally, as several image softwares/sources just put their own name there, and BK users mostly don't want to see those). -- J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf If a cluttered desk is characteristic of a cluttered mind, what does an empty desk mean ? |
#48
|
|||
|
|||
Does the .png image format have a text metadata field?
"J. P. Gilliver (John)" wrote
| If it loads one that already _has_ the metadata though, and subsequently | saves it, does it mess it up, as apparently PSP does? My PSP16 doesn't do that. It gives you an option. Maybe you're not noticing the save options? PSP5 loses the data because it creates a new header, using only the necessary parts, and doesn't see the EXIF data. Any program that doesn't see it will drop it out. It's not necessary for reading the file. There are specific file markers that ID things like the beginning of image data. The optional parts of the header don't relate to that. They're only there if you look for them. |
#49
|
|||
|
|||
Does the .png image format have a text metadata field?
On 14.02.20 10:30, J. P. Gilliver (John) wrote:
In message , R.Wieser writes: John, Can you explain in baby steps how you get to the "image properties - summary dialog", All I have to do, on XP, is to rightclick a file and select "properties" from the popup menu. I than get a dialog with two tabs, "General" and "Summary". Clicking the Ah. I get five tabs - General, File Hashes, Security, Details, and Previous Versions. (The File Hashes one is because of some utility I've added that I think Paul put me onto.) "Summary" tab gives me a "page" with a button at the bottom showing either " Simple" or "Advanced ". When it shows "Advanced " the page above it shows textboxes named, from the top, "Title", "Subject", "Author", "Category", "Keywords" and "Comments". But ... As you have W7 what you see might be rather different. I was looking for an image to show you what I see, but could not find anything better than the below. Just ignore the "Security" tab: http://www.cs.mun.ca/jdk1.6/technote.../security2.gif I'm not disbelieving you, just not sure where you're starting from. From a different version of Windows than you are looking at. :-) You might be looking at something like this: https://www.sevenforums.com/attachme...80418003-file- details-properties-add-change-remove-details-1.jpg Yes, that's what I get (plus the File Hashes as mentioned above). (notice the "comments" entry directly above the big, red, right-pointing arrow) Regards, Rudy Wieser I hadn't realised these were editable! The edit boxes don't appear until you hover over them! You can also edit them in the "details pane": in W7, an extra pane appears across the Explorer window, beneath the folder tree pane on the left and file list on the right. This pane can be enlarged (to show more "details") by dragging its divider up. Again, I hadn't realised you could edit in there! However: although .jpg does indeed have a Comment field which can be edited (either in Details under Properties, or in the details pane), .png doesn't; it has a lot fewer details altogether, and the only editable one is Date taken. (Do you see a comment field for a .png file under XP?) Not in the few I checked. Just all info about the res,color etc. |
#50
|
|||
|
|||
Does the .png image format have a text metadata field?
John,
That's the only heading there (in IV - EXIF). [Try it to see what I mean.] Can't. XP, remember. Apart from not having IrfanView nor knowing which version ofcourse. Though see below. I'm sorry, but I do not quite get which point you're trying to make there. The only thing I see is that you recognise that each program does (and shows) things its own way. Which is something I already mentioned. Regards, Rudy Wieser |
#51
|
|||
|
|||
[OT]Does the .png image format have a text metadata field?
On Fri, 14 Feb 2020 10:10:51 +0100, "R.Wieser"
wrote: Shadow, Since almost all image formats are compressed, do you know a tool that can un-compress the data to look for "hidden" text or files? There's a lot of apps that do steganography, What makes you think that steganography is limited to the uncompressed data ? :-p One form of it might make use of a slightly altered compression engine, which returns the same uncompresed (short) byte sequences for two input values, and regard those two input sequences as being a Zero and One. As for your question, a standard Windows installation comes with a few DLLs (GDI, GDI+) that will uncompress images (into memory) and allow you to inspect their contents (using "GetPixel" style calls). But don't assume that that is all you need to do. The data itself might have been encrypted and strewn around the image in a non-lineair pattern (possibly even using a seeded random generator). It will make it rather hard to find the bits themselves, and the order in which they should be read. In other words, steganography normally doesn't stop at layer #1. :-) I'm beginning to get that. But *bad guys* would tend to hide terrifying messages like "don't kill the president" using freeware/open-source programs which don't leave a money trail. I doubt many of them are expert programmers or into crypto. Whatever, it was just a passing curiosity. []'s -- Don't be evil - Google 2004 We have a new policy - Google 2012 |
#52
|
|||
|
|||
[OT]Does the .png image format have a text metadata field?
Shadow wrote:
On Fri, 14 Feb 2020 10:10:51 +0100, "R.Wieser" wrote: Shadow, Since almost all image formats are compressed, do you know a tool that can un-compress the data to look for "hidden" text or files? There's a lot of apps that do steganography, What makes you think that steganography is limited to the uncompressed data ? :-p One form of it might make use of a slightly altered compression engine, which returns the same uncompresed (short) byte sequences for two input values, and regard those two input sequences as being a Zero and One. As for your question, a standard Windows installation comes with a few DLLs (GDI, GDI+) that will uncompress images (into memory) and allow you to inspect their contents (using "GetPixel" style calls). But don't assume that that is all you need to do. The data itself might have been encrypted and strewn around the image in a non-lineair pattern (possibly even using a seeded random generator). It will make it rather hard to find the bits themselves, and the order in which they should be read. In other words, steganography normally doesn't stop at layer #1. :-) I'm beginning to get that. But *bad guys* would tend to hide terrifying messages like "don't kill the president" using freeware/open-source programs which don't leave a money trail. I doubt many of them are expert programmers or into crypto. Whatever, it was just a passing curiosity. []'s There are two ways to pass a message in a JPG. Formats that use 4CC codes, the parser ignores things that are not part of the format. For example: ABCD length series-of-bytes-encrypted-compressed If ABCD is a 4CC that doesn't exist in JPG, it is ignored while you're looking at the picture. Then, you need a simple tool to look for ABCD to find the "packet" with the information, whatever it is. This would be considered an out-of-band method. The next method is in-band, and uses the dots you would hope would have been used for your purposes. ******* An example of steganography, is how the Secret Service has a series of yellow dots embedded in color prints on color printing devices. The Secret Service is tasked with currency crimes, such as counterfeit currency. To aid them, a color printing device adds yellow dots to the image, and via some self-synchronizing mechanism, they can get a serial number from the dot pattern. Presumably the pattern is fine enough and repeats enough (redundancy) so that items as small as a dollar bill will have a serial number in it. This does not guarantee the printer can be tracked down. If the customer pays cash for the printer, then the evidence trail ends. Except for circumstances where a retailer "scrapes your face" for a transaction, and of course we'll never know they're doing that. You can see the yellow dots, in the picture on this page. You have to use the zoomed-in version of the picture to see them. https://en.wikipedia.org/wiki/Machin...ification_Code The chicanery with printers, is amazing. https://en.wikipedia.org/wiki/Counte...tates_currency "Albert Talton, was sent to prison for counterfeiting the United States one hundred-dollar bill and the United States twenty-dollar bill. Produced over 7 million dollars in counterfeit US currency using a standard inkjet printer, and was convicted and sent to prison in May 2009." They figured it out, when he was caught smiling at the cash when buying new cartridges for the printer. Nobody ever smiles when buying ink :-) Paul |
#53
|
|||
|
|||
[OT]Does the .png image format have a text metadata field?
Shadow,
I doubt many of them are expert programmers or into crypto. They don't need to be. They can just grab a standard encryption program off the web. Or just use Cesars ring method. Or XORing the text with a repeated word, or ... The idea is to make the recognition of used bits harder by removing any obvious patterns. Like five zeroes, a one, and than two zeroes (a space character) Heck, they could even go without any encryption, as you would already need to know where to look and what to look for to be able to get the bits in the first place. Regards, Rudy Wieser |
#54
|
|||
|
|||
[OT]Does the .png image format have a text metadata field?
"Mayayana" on Thu, 13 Feb 2020 19:05:04
-0500 typed in alt.windows7.general the following: "Shadow" wrote | Since almost all image formats are compressed, do you know a | tool that can un-compress the data to look for "hidden" text or files? | There's a lot of apps that do steganography, so there must be | something to identify the method used (a generic un-packer). | The NSA would be naked without one. LOL. | Mysterious. So it hides extra bytes? Hides it as "noise". Every Nth byte / pixel encodes the message. "Obviously" it can't be compressed, that will mess up the encryption. (Sort of like the story of the "spy" who sent "grandfather is deceased" and the telegrapher replaces "deceased" with "died". The contact send a telegram back "Is grandfather deceased or demised?".) Since each format is different I don't know of any easy way to track it, or to create the hidden text. Hides it as "noise" -- pyotr filipivich Next month's Panel: Graft - Boon or blessing? |
#55
|
|||
|
|||
Image formats
On 14/02/2020 00:02, Mayayana wrote:
I don't know of anything to recommend PNG, but it could make sense online where high image quality in 24-bit is needed. There's no other widely supported formate for that. The transparency feature is also nice, but I'm not aware of any program for creating finely detailed transparency. For instance, if you want a logo for "John's English Diner" and you want a knife and fork semi-transparent, how do you do that in one image? I don't know. What? You don't know anything to recommend PNG? But you then immediately give a reason to recommend it! PNG is about the best widely recognised format for totally lossless stored photos. You might say one never needs 100% lossless but sometimes you do, like if you are doing complex photoshop type editing operations on a photo and want to be able to save your work and carry on later without the saving and re-loading introducing compression artefacts. -- Brian Gregory (in England). |
#56
|
|||
|
|||
[OT]Does the .png image format have a text metadata field?
Shadow on Fri, 14 Feb 2020 16:42:09 -0300 typed in
alt.windows7.general the following: On Fri, 14 Feb 2020 10:10:51 +0100, "R.Wieser" wrote: Shadow, Since almost all image formats are compressed, do you know a tool that can un-compress the data to look for "hidden" text or files? There's a lot of apps that do steganography, What makes you think that steganography is limited to the uncompressed data ? :-p One form of it might make use of a slightly altered compression engine, which returns the same uncompresed (short) byte sequences for two input values, and regard those two input sequences as being a Zero and One. As for your question, a standard Windows installation comes with a few DLLs (GDI, GDI+) that will uncompress images (into memory) and allow you to inspect their contents (using "GetPixel" style calls). But don't assume that that is all you need to do. The data itself might have been encrypted and strewn around the image in a non-lineair pattern (possibly even using a seeded random generator). It will make it rather hard to find the bits themselves, and the order in which they should be read. In other words, steganography normally doesn't stop at layer #1. :-) I'm beginning to get that. But *bad guys* would tend to hide terrifying messages like "don't kill the president" using freeware/open-source programs which don't leave a money trail. I doubt many of them are expert programmers or into crypto. They don't have to be. That's what the Schmart Guys in the Back office are for. They are the ones who have come up with the app/program which takes the plan text and hides it in the picture. (and decrypts the picture for the message.) Sometimes the picture is the message. Whatever, it was just a passing curiosity. []'s steganography, like so much, works best if nobody suspects anything. "Oh look, a picture of a kitten." Yes, but the code for every tenth pixel is actually the message "embedded" in the data. (Goes back to some of the earliest examples: drawings of the patterns on butterfly wings which "just happen" to match the layout of the defenses.) Crypto, like camouflage, is all about giving Them something to see which is not out of place, in the time they have to look. I recall reading of encrypted messages which when decrypted, were little more than revolutionary rants. But those rants had the encoded "secret" message concealed within. [technical aside: encrypted is what you do to the message which makes it unreadable. Rot-13 for example. "Jr unir gjragl svir pvtnef.", Wbua'f Hapyr unf n Zbhfgnpur. "Codes" are where words or phrases are substituted, and while the result may be grammatical and make sense, the meaning is not on the face of it. "We have twenty five cigars.", John's Uncle has a Moustache. The Chair is against the wall. Les sanglots longs des violons de l’automne ["the long sobs of the violins of autumn"], "The word Gullible is not in the dictionary." Etc, etc, etc.] As I said, I email a picture of a cat, and a flower. The position of the cat and the flower carry the message. I am remembering a story of a bunch of men recruited for a project. They decided to go, promising that they would send a photo when they got there. If they were standing, all was good, come and join us. If they were sitting, don't. Comes the letter, there is the photograph, all of them are laying on the grass. -- pyotr filipivich Next month's Panel: Graft - Boon or blessing? |
#57
|
|||
|
|||
[OT]Does the .png image format have a text metadata field?
Paul on Fri, 14 Feb 2020 15:18:04 -0500 typed
in alt.windows7.general the following: The chicanery with printers, is amazing. https://en.wikipedia.org/wiki/Counte...tates_currency "Albert Talton, was sent to prison for counterfeiting the United States one hundred-dollar bill and the United States twenty-dollar bill. Produced over 7 million dollars in counterfeit US currency using a standard inkjet printer, and was convicted and sent to prison in May 2009." They figured it out, when he was caught smiling at the cash when buying new cartridges for the printer. Nobody ever smiles when buying ink :-) EURion Constellation. An anti-counterfeiting trick which causes modern color copiers to fail when you try to photocopy banks notes. -- pyotr filipivich Next month's Panel: Graft - Boon or blessing? |
#58
|
|||
|
|||
Image formats
"Brian Gregory" wrote
| PNG is about the best widely recognised format for totally lossless | stored photos. | The most widely recognized in browsers. So it's good for high quality images online when quality is important. But it's still not widely used. It's just too big compared to JPG. | You might say one never needs 100% lossless but sometimes you do, like | if you are doing complex photoshop type editing operations on a photo | and want to be able to save your work and carry on later without the | saving and re-loading introducing compression artefacts. | I use BMP or TIF for that. I could use PNG but it's relatively bloated and it's also complex. Try saving a big image as TIF and then PNG. PNG is a lot of work to pack and unpack. It's slow. And as I wrote to Java Jive, I did a test yesterday and found that PNGs are much bigger than TIFs. So, yes, I see it as potentially useful for online, in a limited way, but otherwise of no value. I only have a handful of PNGs on my system and they all seem to be charts from webpages. The only value unique to PNG is transparency, but I don't usually have a reason to want that, and I don't know of any tools to take advantage of it. I actually wrote code at one time, or rather finished some faulty code, for rendering PNGs. It's a nightmare. I think there are something like 36 variations. But at the time Windows Explorer couldn't render a PNG and I was making an Explorer Bar. Yet I never really had occasion to use the code because I never had occasion to look at a PNG. |
#59
|
|||
|
|||
Image formats
Mayayana wrote:
"Brian Gregory" wrote | PNG is about the best widely recognised format for totally lossless | stored photos. | The most widely recognized in browsers. So it's good for high quality images online when quality is important. But it's still not widely used. It's just too big compared to JPG. | You might say one never needs 100% lossless but sometimes you do, like | if you are doing complex photoshop type editing operations on a photo | and want to be able to save your work and carry on later without the | saving and re-loading introducing compression artefacts. | I use BMP or TIF for that. I could use PNG but it's relatively bloated and it's also complex. Try saving a big image as TIF and then PNG. PNG is a lot of work to pack and unpack. It's slow. And as I wrote to Java Jive, I did a test yesterday and found that PNGs are much bigger than TIFs. So, yes, I see it as potentially useful for online, in a limited way, but otherwise of no value. I only have a handful of PNGs on my system and they all seem to be charts from webpages. The only value unique to PNG is transparency, but I don't usually have a reason to want that, and I don't know of any tools to take advantage of it. I actually wrote code at one time, or rather finished some faulty code, for rendering PNGs. It's a nightmare. I think there are something like 36 variations. But at the time Windows Explorer couldn't render a PNG and I was making an Explorer Bar. Yet I never really had occasion to use the code because I never had occasion to look at a PNG. But surely you look for a libpng, before writing your own ? I'm not much of a programmer, but if a library is available, I'll use it. Each of the image formats should have a library. libjpg, libtiff, libpng, there should be one for each. (Watch configure on FFMPEG sometime, and the sheer number of libraries it pulls up! There are lots of libraries, so you don't have to write your own.) If you need to cobble together conversion pipes, this is an available method. I still use my collection occasionally for stuff. The naming convention has suffered over the years, but that's the price of someone messing around with it. (I think the original author stopped at some point, and someone else worked on it.) One of its early sins, would be insufficient WxH capabilities in the first version. I don't think they need to be meek about line buffers any more. http://netpbm.sourceforge.net/doc/pngtopam.html http://netpbm.sourceforge.net/doc/ppmtobmp.html The problem now, is a shortage of tools for images with over 4 billion pixels :-) (Yeah, I only have one file like that. It's 10GB or so. I'd show it to you, but... the image hosting sites would not like that.) Paul |
#60
|
|||
|
|||
Image formats
In message , Mayayana
writes: [] So, yes, I see it as potentially useful for online, in a limited way, but otherwise of no value. I only have a handful of PNGs on my system and they all seem to be charts from webpages. The only value unique to PNG is transparency, but I don't usually have a reason to want that, and I don't know of any tools to take advantage of it. [] Even that's not totally unique (is this a rare case where that phrase can be validlly used!): one of the other formats, GIF I think, has the option of declaring one of its colours as the transparent one. (Come to think of it, .ico must as well). I gather PNG has _degree_ of transparency, which is different - the others have only yes or no. -- J. P. Gilliver. UMRA: 1960/1985 MB++G()AL-IS-Ch++(p)Ar@T+H+Sh0!:`)DNAf The trouble with the death penalty has always been that nobody wanted it for everybody, but everybody differed about who should get off. - Albert Pierrepoint, in his 1974 autobiography. |
Thread Tools | |
Display Modes | |
|
|