Every month. W3Techs records the percentage of websites using popular image file formats. For a long time, PNG was slowly creeping up on JPEG. In the latest numbers, PNG has pulled ahead, being used on 74.1% of websites, against JPEG’s 73.8%. GIF stands at a distant third with 36%, followed by SVG at 3.8%.
Keep in mind, this isn’t the same as saying more PNG images are on the Web than JPEG images. If pages which hold JPEGs contain more of them, there may be more JPEG images even if they aren’t on more sites. Since galleries of JPEG photographs are common, this is a plausible situation.
You can see the trend by looking at the RSS feed. Unfortunately, all links point to the same URL, but if you can view the RSS preview, you can get the old numbers on a monthly basis. The web page is updated daily, which is why the numbers quoted above don’t match the June numbers in the RSS feed.
Posted in News
Tagged images, JPEG, PNG
Search for “imagegate,” and you’ll find lots of articles claiming there’s a malware risk in JPEG files. Look more closely, and you’ll notice they don’t provide any support for the claim. They all take an article from Check Point as their source, but there are two little problems with that: (1) The article doesn’t blame JPEG files, and (2) as I noted in my last post, it’s inept reporting.
Looking very closely at the video which accompanies the Check Point article, I see that it shows a file called “payload.jpg” being uploaded. This must be what all these sites are going by. You have to look really close to see the blurry file name coming up, and I give these sites credit for examining it more closely than I did.
If this was Check Point’s way of tipping people off that the weakness is in JPEG, it’s a strange way to do it. Did they think that ordinary users would catch it, but malware authors would be tricked by the article’s reference to non-image formats like JS and HTA as “images”?
None of the articles I’ve seen question why Check Point tipped them off in this way or note the inconsistencies in the article and the video. None of them ask why Check Point refuses to give any information until “the major affected websites” fix the problem, when a format vulnerability impacts any software that reads the files.
Doesn’t anyone know how to do journalism any more?
Update: Facebook says that “Imagegate” is bull.
My video for today briefly covers graphic file formats from the forties to the present. I made some interesting discoveries along the way, especially Laposky’s CRT “Oscillons.”
Some of us know that JPEG uses the Discrete Cosine Transform (DCT) to do lossy compression. A few people have more than the vaguest idea of what that means, and I don’t claim to be one of them. Today, though, I found a video on YouTube which gets me closer, even though I’m not very inclined toward advanced mathematics. It’s included in my file formats playlist on YouTube.
The FotoForensics site can be a valuable tool in checking the authenticity of an image. It’s easy to alter images with software and try to fool people with them. FotoForensics uses a technique called Error Level Analysis (ELA) to identify suspicious areas and highlight them visually. Playing with it a bit shows me that it takes practice to know what you’re seeing, but it’s worth knowing about if you ever have suspicions about an image.
Let’s start with an obvious fake, the iconic New Horizons image of Pluto with the equally iconic Disney dog superimposed on it. The ELA analysis shows a light-colored boundary around most of the cartoon, and the interior has very dark patches. The edge of the dwarf planet has a relatively dim boundary. According to the ELA tutorial, “Similar edges should have similar brightness in the ELA result. All high-contrast edges should look similar to each other, and all low-contrast edges should look similar. With an original photo, low-contrast edges should be almost as bright as high-contrast edges.” So that’s a confirmation that the New Horizons picture has been subtly altered.
Let’s compare that to an analysis of the unaltered image. The “heart” stands out as a dark spot on the ELA image, but its edges aren’t noticeably brighter than the edges of the planet’s (OK, “dwarf planet”) image. The tutorial says that “similar textures should have similar coloring under ELA. Areas with more surface detail, such as a close-up of a basketball, will likely have a higher ELA result than a smooth surface,” so it seems to make sense that the smooth heart (which is something like an ice plain) looks different.
A report from the 69th meeting of the JPEG Committee, held in Warsaw in June, mentions several recent initiatives. The descriptions have a rather high buzzword-to-content ratio, but here’s my best interpretation of what I think they mean. What’s usually called “JPEG” is one of several file formats supported by the Joint Photographic Experts Group, and JFIF would be a more precise name. Not every format name that starts with JPEG refers to “JPEG” files, but if I refer to JPEG without further qualification here, it means the familiar format.
A while back I posted on the painfully poor choices in creating thumbnails of JPEG2000 files. Since then I’ve come to realize that support for image file processing in Java is even worse than I’d realized. Now I’m trying to make thumbnails from TIFF files. At first I went with JAI, even though it hasn’t been supported for five years and relies on implementation-dependent classes. I’d done this before successfully, but now I’m trying to do it in an EJB under JBoss. This runs into a NoClassDefFoundError trying to get com.sun.image.codec.jpeg.JPEGCodec. A web search suggests there’s some obscure trick necessary to access com.sun.image, but I couldn’t figure it out. It occurred to me that for what I’m doing, javax.imageio should be sufficient to do the job. It can read an image file, standard Java classes can scale the BufferedImage it produces, and then it can write the scaled image to a file.
Only one trouble: javax.imageio knows nothing about TIFF. A search on imageio and TIFF leads to suggestions to use JAI.
Really, what kind of language is that poor in dealing with common image formats?
Here are a few of the news items I mentioned recently on the old blog, for your convenience:
- A workshop on JHOVE2 will be held after the conclusion of iPres 2009 in San Francisco, on October 7, 2009. This will include, for the first time, a presentation of the prototype code.
- JPEG XR, formerly known as Microsoft HD Photo, is now an international standard, as reported in a JPEG press release.
- JHOVE 1.4 is now available on SourceForge. The main change is that PDF/A compliance is more accurately identified than before, and is based on the final standard rather than a draft.
Posted in News
Tagged ISO, JHOVE, JPEG, JPEG XR