Digital Photography Articles

Everything you need to know about digital photography. Articles on digital photography workflow, technical articles, JPEG compression, file naming strategies, photo cataloging software (digital asset management), photo software reviews, equipment reviews and archiving the digital photos on CD or DVD.

Learn how to use programs for organizing digital photographs. Also includes a beginners guide to digital photography or those who are making the change from film to digital.

 

Is there an question or article that you would like to see?
If so, please feel free to leave a comment below!

Latest Photography Articles

 

Latest Comments in Photography Sections

2010-03-18JPEG Huffman Coding Tutorial
Thanks a lot for this tutorial which truly helped ...
2010-03-17Designing a JPEG Decoder
Hi, I am new in image programming. I just land ...
2010-03-16JPEG Huffman Coding Tutorial
Hello Calvin, I have a RAW data, it is compressed ...
2010-03-15Folder hierarchy strategy
Hi, great site. I'm scanning shoe-boxes full of ...
2010-03-14Comparison of Backup Software
I did not phrase the question properly-I see that ...
2010-03-14Comparison of Backup Software
What do I use for syncing when I use Backup4all ...
2010-03-13Digital Photography Articles
Hi Sir, First, thanks for creating such an ...
2010-03-13JPEG Chroma Subsampling
The JPEG standard allows for Hx and Vx to be from ...
2010-03-11Comparison of Photo Catalog Software
There's a version of Picasa for Mac.
2010-03-10Monitor sRGB mode preset
Hi Calvin, I just wanted to say thanks for this ...
2010-03-10JPEGsnoop - Options
Great website and program, thanks I have several ...
2010-03-09Designing a JPEG Decoder
I really thank you for your help. Although I'm ...
2010-03-09Designing a JPEG Decoder
Please help me... I really need an answer... I'm ...
2010-03-05JPEGsnoop - Options
Hi Calvin, Thanks for great website. I enjoyed a ...
2010-03-04Digital Photography Articles
Hey Cal: I wanted to follow up on my question ...
2010-03-03What is an Optimized JPEG?
Hi, found your articles useful in learning about ...
2010-03-02Undelete your Photos!
Transferred pictures from memory stick to desktop ...
2010-02-27JPEGsnoop - Identify Edited Photos
Thank you so much for creating such an invaluable ...
2010-02-27Undelete your Photos!
Hi, I have lots of corrupted images which I ...
2010-02-26JPEGsnoop - Identify Edited Photos
Your software is great. It has found 3 images to ...
Gallery
Girls at Party

 

 

 

Organizing and Naming Photos

How to be organized when you have thousands of photos on your computer. File name schemes that allow a mix of digital photos, scanned photos and non-photos on the same drive.

Read More

 

Technical Articles

For those interested in knowing the details of how digital photos are stored, articles on how JPEG compression works, and other in-depth tutorials:

 

 

Archiving & Storing Photos

Your entire photo collection can vanish in an instant, so a proper methodology in archiving your photos is crucial. Here you will find strategies to help automate the digital photo backup process.

Read More

 

Importing Digital Photos

Software used to transfer your images from memory cards to your hard drive.

Read More

 

Beginner's Guide to Digital Photography

Articles for those who are either new to photography and want to start with digital, or those who are experienced with film and want to upgrade to digital.

 

 

Digital Photography Equipment

Articles on digital cameras and related equipment.

 

 


 


Reader's Comments:

Please leave your comments or suggestions below!
2010-03-13william wallace
 Hi Sir,

First, thanks for creating such an informative website as well as your amazing JpegSnoop!

Next, my question. I guess this is more in the area of Digital Image Analysis, but I thought I'd ask it here...is it possible to determine the quality of a jpeg image (low quality and blurry versus high quality and detailed) by examining the jpeg file itself (instead of viewing the image)?

The reason I ask is that I have a folder of approx. 10,000 jpeg files and I'd like to sort them from highest quality to lowest quality, but I'd like a way to sort them automatically without having to view each individual picture and adding metadata to each file (which would be quite time consuming). Thanks for any info!

-Bill
 Thanks! Interesting question. I am sure that there are specialized tools out there that can assess the degree of "bluriness" for a given photo, but I'm not aware of any. Without being too fancy, it may be possible to determine the number of MCUs that have have an increased proportion of high-frequency image components (suggesting that the image has areas with more detail). This type of analysis is not foolproof, but could help differentiate images that are mostly blurred versus those that contain a lot of detail. However, this would not catch issues where the focus was wrong (ie. autofocus locked on the background instead of your subject). JPEGsnoop does count the frequency of huffman entries, but it would need to be presented in a different manner for it to be useful as described above. So, in short, no the current tool won't directly help you with this search :)
2010-03-04Jay
 Hey Cal:

I wanted to follow up on my question last year about detecting layers in a flattened image (Yes, I know that is part of the process in saving a PSD to a JPG. Photoshop has that warning every time you go to the SAVE screen).

I have tried using error level analysis to find the answer, but from my understanding of it, there has to be at least one resave of a JPG along with a physical change to the original image, for it to spot anything.

Here is the scenario:

I received a letter of recommendation, an email with attachment, from a chap I wish to hire. He claims that he had it scanned for him.

It looks like it was created in MS Word, using Arial, and then printed off on white stationary with a black and white business logo at the top and an address line along the bottom after the closing.

The letter was not signed (somewhat of a red flag).

The text part of the letter looked darker than the logo -- which may mean that the logo was in color or greyscale -- but I could not find anything unusual - like repeating patterns, different blocking, ringing artifacts, etc. - that would lead me to conclude the text was added to an existing JPG.

After I "Snooped" it, it identified the software as CS2 from the Exif data, but the compression signatures came out as something else, like IrFanView.

Given the Photoshop IRB that JPEGSnoop found with a matching pair of signatures that are not from PS, I take this ro mean that the scan may have been made within Photoshop, saved as a JPG (let's say "Save at 60"), and then dragged through IrFanView for a resave.

Do you think that the resave was to hide any evidence of layers?

I also tried using pixel equalization to spot any unnaturally light or dark pixels, as well as Edge Detection.

I am thinking that this dude scanned a sheet of blank letterhead (creating one layer) and then created a text layer to merge with the first when saving it as a flattened JPG.

But, how do I prove that?

Thanks for your help.

------

One thing that may not be apparent is that when you save an image with Photoshop, all layers are usually "flattened" in generating the single JPEG image. This single image uses only one set of quantization tables. In fact, if you resave an image with various tools, it is generally only the last tool that will define the quantization tables used to encode the final JPEG image.

So, to answer your question: while JPEGsnoop can often identify that an image was generated by Photoshop, it cannot infer further details about the layers that were used to generate the image (at least on the basis of the quantization tables). Therefore, cropping will make no difference to the quantization tables used (in fact it may change them, depending on the tool that you are using for the cropping).

The best way to accomplish what you are after is to use a tool that supports one of the many imaging algorithms that can decompose the error level analysis and other characteristics that point to the fact that an image was created from a composite.
 I can't help you with the "proof", as that image analysis can best be answered by others. However, to answer your question regarding the mismatch of quantization tables (signature) from the Photoshop IRB metadata: it is quite possible that Photoshop was indeed used previously and then the file was again resaved using another graphics editor (ie. other than Photoshop). The Photoshop signatures are reasonably unique.
2009-10-01Ivey
 I have a DVD and i am trying to see if it has been altered the person in the DVD said that some one changed the DVD but i can not see any sign of it being altered by eye. Can anyone tell me if this program can tell me this information. Or is there anyway to tell. Thankyou!!
2009-08-02Newtodigital
 Your site is a great resource for someone like myself who is transitioning from film to digital. The articles about naming files and folders and cataloging sw are particularly helpful. They get me to think before clicking (and losing) thousands of shots.

Further down the road, I will be interested in web galleries, either custom or pre-fabed, and POD photobooks. There are plenty of these around, but comprehensive comparative reviews like the articles mentioned above are hard to come by. If they are already at your site, what are the links? If not, perhaps you can provide some. TIA.
 Thanks! I have recently began experimenting with photobook services, and may be able to provide some input on these at a later date (especially if I find a decent competing service to try -- right now I've been using blurb). I do have a number of web design-related articles I've considered writing, and may consider expanding coverage in that area soon. Thanks for the great suggestions!
2009-02-23Jay
 Here's a task for which I'm interested in using JPEGsnoop, and i'd like your advice on how to use it.

I have an image that was saved in Photoshop that I suspect was created from two layers, and I think I know which parts of the image belonged to each layer.

Id there a way to compare two separate parts of an image to see if they originated from separate layers?

For example, the 2nd image/layer looks like it came from an image with a lower quality setting (higher compression ratio) than the first. If I cropped off parts of each image from the merged product of the two, would they produce different tbles in JPEGsnoop?
 Hi there Jay --

One thing that may not be apparent is that when you save an image with Photoshop, all layers are usually "flattened" in generating the single JPEG image. This single image uses only one set of quantization tables. In fact, if you resave an image with various tools, it is generally only the last tool that will define the quantization tables used to encode the final JPEG image.

So, to answer your question: while JPEGsnoop can often identify that an image was generated by Photoshop, it cannot infer further details about the layers that were used to generate the image (at least on the basis of the quantization tables). Therefore, cropping will make no difference to the quantization tables used (in fact it may change them, depending on the tool that you are using for the cropping).

The best way to accomplish what you are after is to use a tool that supports one of the many imaging algorithms that can decompose the error level analysis and other characteristics that point to the fact that an image was created from a composite.

I have current plans to incorporate some basic imaging algorithms into upcoming versions of JPEGsnoop, which may address some of what you're looking for. Until then, a search for error level analysis tool may reveal some alternatives.
2009-02-16Jay
 You've created a great program that can capture most signs of editing. Where it can be fooled is by in-camera editing features that modify an original image but do not change the camera's signature.

In other words, you are correct by your guarded use of the terms, "high-probability of being original" and "you cannot prove conclusively that a photo is an original," which I take to mean that it is easier to identify an altered photo.

BTW, in terms of a percentage range, what exactly do you mean by "high probability?"

Thanks, and keep up the good work
 Thanks! You're right in that JPEGsnoop could possibly mark an image as being original even though in-camera editing facilities may have been used. There is no way that these edits can be detected with any certainty, and this wasn't the intention of the image assessment algorithm. Since in-camera editing is (and will likely remain) extremely limited in capability, this doesn't present much of an issue for those aiming to detect more hidden edits.

Yes, it is far easier to "prove" that an image has been edited than that it hasn't. One of the main reasons why I do indeed refrain from indicating that an image is guaranteed to be original is that I know for a fact that it is possible to fake out such analysis.

As a proof of concept, I have succeeded in doing this myself. However, the chance that someone has gone through this challenging process is quite remote. I have never seen any examples of this "in the wild", so I would peg the probability or originality extremely high. Of course, if you are dealing with a situation where someone wanted to cover their tracks at all costs, anything is possible.
2009-01-19Ben
 I like the look of the JPEG fixing information. Have you attempted to do the same with RAW files?

I have a set of Canon Raw .CR2 files that became corrupted whilst travelling. The embedded previews come up fine, but then when it goes to process the larger preview it comes up with the corruption. I suspect it would something along the lines you were saying with JPEGs and interuuption of the bitstream.

I suspect the corruption occured when transferring images either from CF card to storage drive, or from the storage drive to laptop.

Regards,
 Hi there Ben -- I have not had a look at recovery of Canon RAW files, as I believe they have been encoded in JPEG Lossless mode, which I have yet to review. I'll add it to my list of features to take a look into. You are correct in that the integrity of the embedded preview is virtually unrelated to the integrity of the primary RAW image.
2008-09-08susie
 I am looking for a camera that would take professional size pictures I can sell but I need them to be large enough. I wanted to know what size megapixel camera would take a 12MB size picture?
2008-01-09Lensmeister
 I hope someone can help. What I really need is advice on how to/or what to use as a database/catalogue for my photos. I have over 10,000 digitial images and over 800 rolls of 35 mm film that I am scanning. What I need is something that I can have a thumbnail of the images, all the exif, and other data I store (the 35 mm I want to make items similar to EXIF).

I am totally lost ... can someone help please ?
Thanks :)
2007-11-24Thomas Kaiser
 JPEG Guru needed for improving Linux webcam driver :-)

I developed the Linux driver for the Pixart PAC7311 webcam chip and have a problem with picture quality. Pixart claims to use "standard" JPEG but they do not. They included special markers after (or before) each MCU which I must remove to decode the picture.
You can find some information here http://www.kaiser-linux.li/index.php?title=PAC7311

If you are intrested helping me I can give you more detailed information. Just drop me a mail :-)

Thomas
 Hi Thomas -- I'm always interested in challenges. Send me some example frame captures from the device and I'll see what I can figure out. Inserting markers after every MCU seems excessive, so I'm curious as to why this is being done.
2007-10-01Ken
 Hi Calvin,

Thanks for updated JPEGsnoop software. It seems like a handy tool when you need to "go deep" into a file.

While I am still researching software for my workflow, my journey has led me to again reconsider shooting RAW. Unfortunately, I have read far too many articles both for and against it for my own good. What are your feelings towards RAW? You always seem to be talking about JPEG files, but I was not clear if you shoot RAW or JPEG. And, if you do shoot RAW, what are your thoughts on RAW converters? I would be very curious to hear your thoughts and opinions.

Thanks,

--Ken
 Great question -- Yes, I have certainly focused a lot of my technical interest on JPEG, as it is more prevalent and often misunderstood in internet forums. Since it is a reasonably "open" standard, unlike the proprietary RAW formats, it's actually possible to understand how it works :)

I do shoot RAW, but I generally limit it to times when I a) intend to deliver the shots to others (e.g. a modeling shoot / sporting event), b) place significant importance on the images or c) find myself in difficult lighting conditions. The rest of the time I often choose to shoot JPEG. While the RAW images from my camera (Canon 40d) are certainly nicer and provide far more processing latitude than the JPEGs, one can't forget to factor in the additional processing time that the RAW format usually entails, along with the increase in storage requirements (especially when you keep both RAW and the converted JPEGs). Some people have managed to dial in a RAW processing workflow (importing, converting, XMP tagging, buddy-file synchronization, etc.) that is reasonably efficient, but I am not convinced that my current toolset is what I will ultimately end up using.

When I travel for extended periods of time (e.g. 1 month+) in remote regions, I am also more apt to shoot JPEG because it is more important for me to have enough redundant portable storage to keep copies of the images safe -- file space is already a difficult issue on the road.

As for RAW converters, I have not yet completed a fair evaluation of the different products on the market (Photoshop ACR, Lightroom, Bibble, Phase One, etc.). When I do, I may be in a better position to share my opinions and refine my own RAW workflow.
2007-09-26 
 Great site. Any chance you could add an RSS or ATOM feed?
 Good idea... It has been suggested to me in the past, and since I wrote my own CMS system it's going to take a little bit of work. I think that I would prefer to use that instead of the notification system that I have built in the meantime. Thanks!
 
2007-08-20Kal
 I'm trying to decide between a DSLR (pentax K100D) with say 50-200 or 70-300 Tamron or Sigma lens and Sony H7 or similar strong Optical Zoom. I'll be visiting Kruger Nnational Park later and wuld like to be able to take some wildlife Zoom photos as we are to remain in the car while in the park. You gave us a formula to convert things lens specs to zoom in X units. So..is the 300 in 28-300 better than the 300 70-300 for my purpose? What about comparing a 50-200 versus 70-300. I guess my question is whether it is the ratio that is important or the larger number in the lens i.e. 300 (versus 200).

Also to get the same zoom as say a 70-300 on the DSLR what zoom should I look for in a Point and shoot (recognize of course the quality will be lower in P&S)?
 For wildlife, you're going to want a decent long telephoto lens. While in Etosha and Moremi (in southern Africa), I shot from the vehicle with my 100-400/L and tended to stay in the 250-350 range (this is on a DSLR with a 1.6x multiplier, which is common). You might find that 200mm is a little short, and that the 300 provides you with more opportunities. When comparing lenses, unless there is a significant difference in optical quality, you should start by making sure that you get the range that you need (i.e. is 200 enough, or do you need 300mm?) In your case, I would probably opt for the longer zoom (70-300).

Ignoring the maximum focal length for a moment, it's important to keep in mind that it is much harder to produce a zoom lens with high image quality across a large range versus one that only covers a narrower range. In other words, the 28-300 lens has a 10.7x (300/28 = 10.7), while the 70-300 lens has a 4.3x. For a similar price point, it is a pretty safe assumption that the 70-300 will provide better image quality across the range than the 28-300 superzoom. Nowadays, the superzooms are becoming better, but I would tend to stay away from them unless I was truly requiring an all-in-one for convenience.

As an aside, if you are taking photos from the vehicle, consider buying a bean bag that you can place over the window sill, as you won't be able to use a tripod.

To get an equivalent of 70-300 (on a DSLR) in a point & shoot camera, one would look for an "35mm equivalent focal length" of 110-480, which is very unlikely (this assumes that your DSLR has a focal length multiplier of 1.6x). Note that the rating will say "equivalent" as the sensor size is much smaller in a point & shoot camera, making the real focal length comparison not as simple.
2007-07-17Darren
 Hi again. It would be great if you could expand the photo section to include articles about digital video too. You mention that you also take video footage and this is an area I'm just getting into, but haven't much of an idea about the best way to do things.

Thanks!
 Thanks for the suggestion! I will definitely give it some thought -- While I have recently decided to start writing a series of technical web design articles, digital video is certainly another area of interest too. I'll put it on the list!
2007-05-20JL
 Hi,
Can you tell me if there is any program that can scan my imported pictures to get the accurate date? We just had our computer cleaned up and when I put them back on the computer a ton of them (which had been taken a long time ago with a different, cheaper digital camera) got imported incorrectly. I'm currently trying a trial version of adobe 5.0.Thanks so much :)
 In what way did you find that the photos were imported incorrectly? I assume that you are seeing that the File Created Date is recent, when you expected it to be the date that the shot was taken. If so, when you right-click on one of these photos and then click on the Summary tab then the Advanced button, do you see the correct date listed in one of the properties (e.g. Date Picture Taken)?

If you see the correct date here, then you're in luck. That means that the image's EXIF information still retained the original photo date. There are many utilities that will use the EXIF date information to update the other file timestamps as well. The most popular utilities to do this are: Exifer, jhead, and AttributeMagic. I'll add this to the list of articles to write, as this is a very common need for digital photographers. For example, you might have forgotten to set the date on your digital camera (or travelled to a different time zone), and later need to update all your photos.
2007-05-07S
 Noticed your post on dpreview forum. If you're still interested in example files then send me a mail. I can provide any number of Sony H7 files, including Gregrag McBeth color chart and ISO res chart.

Also, thanks for an interesting site. Hadn't seen it before, but you've done a great job providing a lot of useful info. Downloaded JPEGsnoop and checked several H7-files. I was wondering how they managed to compress all the files so small and with so little variation in file size and it seems like you've got the explanation. Found a lot of variation in the "Approx quality factor" from around 97% to 87% on the files I checked.
 Thanks! Email sent.
2007-01-26Rachel
 First of all, fantastic site! Should me mandatory for anyone using digital photography. I've learned so much.

Have you looked into any of the online photo storage/sharing sites? I'm interested in using one for an additional back up of my collection and would like to be able to easily share with friends and family. There are so many, it's hard to know which one to choose. With your expertise, I wondered what you thought?
 Thanks Rachel! I have used pbase in the past (smugmug was another close alternative) and found it to be a great source for getting critical input on photos from other photographers. That said, the interface is not totally intuitive, so I wouldn't recommend it for those who are looking for something very easy to use.

As I made the transition to host my own photo galleries, I never invested much time in comparing all of the various photo sharing sites out there, but I would suggest that you have a look at the following two review pages:
You mentioned an interest in using these photo sharing sites as a backup for your photo collection. Generally, I don't think they do a particularly good job at that because of the fact that they often store much-reduced versions of your photos, not the originals. Some do keep fullsize originals, but you will have to be very mindful of your file storage limit restrictions.

A much better plan would be for you to create a backup of your photo collection yourself (on hard drives, DVD-R, remote backup, etc.) instead of relying on a photo sharing site to achieve this goal.
2006-12-30Justin
 Are there any software applications that allow you to sync the categories etc with 2 computers?
2006-11-19KimF
 Your website is incredibly helpful in 2 of my latest endeavors - organizing and backing up digital photos. Thanks for all your efforts!
2006-09-22Chris
 My Fuji Finepix S5600 has problems getting a clear image of bright red birds using single auto focus unless I focus first on another object at that distance, eg fence bird is sitting on. Images are out of focus even though focus point right in bird image, generally 8 to 10 x optical zoom. Other colours appear to be OK giving a sharp image even at maximum zoom.
Can you explain why.
Thanks in advance
2006-03-05Kathy
 Thank you so very much! You are the first person to finally give me some solutions and answers to this problem. A person can't move on and learn more unless the issues at hand are taken care of!
 Glad I could help!
2006-03-04Kathy
 Would love to hear your comment on digital camera's "purple fringing" and what causes it and how to avoid it.
 Hi Kathy -- "Purple fringing" has been a contentious issue for those who have tried to explain it. Some discussions on it (#1 #2 #3) have proposed some potential causes (blooming, microlenses chromatic aberration, InfraRed sensitivity, etc.), but I have yet to see any definitive answer.

As for how to avoid purple fringing, it seems that there is some consensus:
  • Use apertures f4 or smaller
  • Avoid high-contrast subjects in poorly-lit environments
  • Use a lens hood
  • Avoid using focal lengths wider than 28mm
2006-03-02Sarah
 Hi. I am researching digital photography vs. film, and I actualy prefer film to digital for a few different reasons. I won't go in to details now. I am wondering if you may share your opinion of the matter. Just email me at the address above. Thank you!
 HI Sarah -- I have posted a number of my personal experience and thoughts regarding digital versus film. I'd be interested to hear your opinions on how you prefer film.
2005-09-12 
 

The photographer has an amazing eye for capturing astoundingly beautiful images. I am overwhelmed by such beauty, especially the images of B.C.

 

Thanks!

2005-09-11Darren
 

Thank you for taking the time to create such a fantastic website! I've been reading it quite a lot lately and have picked up loads of hints and tips.

I'd be interested to read your thoughts and views on RAW formats - workflow, pros/cons etc.

 

Thanks Darren! Glad you found it useful... I will add discussions of RAW workflow to my to-do list. Thx!

2005-04-13Gerry Pasternack
 

This is abstracted from your response to me in dpreview forum on how to track performance of Monaco Optix-XR calibrator:

"So, I managed to use a combination of pulling apart the generated profiles and then compared this with the IT-8 target LAB color data in Photoshop to determine my own delta-E values and drift. A fair bit of hassle, but I could determine a lot more about the quality of my profiles than would ever be obvious from the application itself."

Can you be more specific? What program do you use to read the ICC Profile contents, and how do you analyse the data?

Thanks, Gerry

 

Hi Gerry --

Very soon I had intended to open up my section on color management, but at this stage it is not yet ready. One of the articles that I was going to publish was going to detail this exact process. However, in the interim: you can derive objective data from several of the files that are generated by Monaco EZcolor (MonacoEZcolorDevice.log, the monitor *.icm profile, and the IT-8 reflective target data (eg. MONR2004.06.01.mrf). The two key points are that some ICC profiles have text-based sections that include the actual lookup table, and Monaco's target files can be opened as a text file!

2005-03-16Keith Alexander
 

As I'm sure you know happens so often on the Internet, I just *stumbled* across your TERRIFIC web site today. I teach digital photography courses at local colleges and at photography "clubs," and maintain a web-site of resources for my students. I had just printed up a list of "Internet Resources on Digital Photography" to give as a handout to my class tonight, but now I have to go edit that file and re-print it with your site added (with three stars next to it, I might add)!

Terrific site!!!! Bravo!

OK, here's my own lame little site (it's oriented towards both my computer and digital photography students, so forgive me). You can go visit at: http://groups.msn.com/Mr-AlexandersNewsletter

 

Thanks Keith for the nice comments... glad you found it useful!

 


Leave a comment or suggestion for this page:

(Never Shown - Optional)