I’ve been doing HDR for 3 or 4 years and have spent all that time beating my head off the issue of noise in the Photomatix output. The tone mapping settings I use don’t help any since they accentuate texture and we all know that digital noise is a great example of high frequency texture. Even when I am holding back a bit on the gamma, though, the results are typically somewhat grainier than I would like. Part of the time I convince myself that this is fine and that the noise is adding to the result – grainy black and white pictures are really common and often considered artistic after all – but most of the time I end up in Photoshop mixing in chunks of one or other of the original exposures to try to cover over the splotchy mess that Photomatix turned my clear blue sky into.

During this time, I had read plenty about noise reduction software but hadn’t really bothered looking into it in any great detail since I had seen what Lightroom’s noise reduction did and it wasn’t impressive. If Adobe couldn’t make a huge difference, surely no-one else would be doing a great deal better so why spend $70 or so on yet another plug-in? As it turns out, however, I was hopelessly wrong.

I’ve spent some time this week playing with a couple of noise reduction plug-ins and am, frankly, gobsmacked at what an amazing job both of these products do at reducing image noise but, more importantly, preserving fine detail. Anyone can get rid of noise by blurring an image enough but these tools get rid of the noise AND keep all my nice sharp edges and tiny details crisp and clean. I have no idea how they manage it but it truly is a wonder to behold!

Topaz DeNoise as seen when launched from within Photoshop (click for larger image)

The first tool I looked at was Topaz DeNoise. I’ve read a lot about Topaz Adjust and seen a lot of great images which use it, but had not heard quite so much about their noise reduction tool. It turned out to be a good find – clean user interface, pretty easy to use and quite a few presets for common noise reduction scenarios. Using it on some of my worst Photomatix images, it did a respectable job of cleaning up the skies without smearing the detail but it did seem to leave some rather odd low frequency artifacts behind. This may have been due to the fact that I was using it without having read the whole manual, I suppose, but in the time I spent playing with the tool, I didn’t get as good a result as I managed to get with the second package I tried.

The user interface of the Noiseware Pro Photoshop plug-in. (click for larger image)

Noiseware from Imagenomic is a piece of software that made my jaw drop. It’s ability to remove noise and clean up an image is almost unbelievable. For HDR, where I am typically keen to remove noise from a sky without affecting other areas, it’s ability to remove noise based on particular colours is fantastic but, even without tweaking any of those sliders, I was stunned by how good a job it did of tidying up my images. I’ve included a couple of examples below showing 100% sections of a particularly noisy image (you can see the original here). I wish I had played with this software a lot earlier since I would have saved many hours masking skies in Photoshop had I known it was so impressive.

Detail before using Noiseware (click for larger version)

Details after using Noiseware (click for larger image)

If you’ve been using this kind of software for a while, you’re probably laughing at me right now but, if you’re not, take some advice from a guy how has wasted a great deal of time trying to solve this problem and treat yourself to a couple of hours with one of these pieces of software. Both are available as free evaluation downloads. Like me, though, I expect you’ll have your credit card out within 5 minutes.

Edit: I was so impressed with Noiseware Pro that I asked the nice people at Imagenomic if I could offer a discount code. They were nice enough to agree and are offering 15% off the product to readers of this blog. Click here to get to the order form then use code “DaveWilson” when you are checking out to get the discount. Apparently they like my work too since they are also featuring it on their gallery page.

Wells Fargo Tower, originally uploaded by DaveWilsonPhotography.

The nice folks over at HDRSoft, the makers of my HDR software of choice, Photomatix Pro, have signed me up for their affiliate program. This is a win-win-win arrangement as far as I can see since all parties benefit – you get a discount, HDRSoft get a sale and I get a commission. I’ve seen several others advertising HDRSoft discount codes but didn’t realise before this week that the arrangement involved a seller commission. In the spirit of full disclosure, though, I thought it would be right to mention this.

If you are buying a copy of Photomatix and use the coupon code “DaveWilson” when you are checking out, you will receive a 15% discount on the purchase, and I will thank you very much.

Despite the mutually beneficial arrangement, I should point out that I have been a huge fan of this software for several years now and, as regular readers will know, have pretty much been acting as an HDRSoft salesman for most of this time!

Barn, Johnson City, Texas, originally uploaded by DaveWilsonPhotography.

I’ve always considered myself an early adopter but, thinking about it, I’m actually rather behind the times these days. We are probably the only house in Austin without an HDTV (which is doubly weird given that I used to work in software development for HD satellite and cable settop boxes) and both my computer monitors are still CRTs. My computer, until this weekend, was a rather clunky and extremely unreliable HP system running Windows XP on a dual-core AMD Athlon 64. Much as I love Adobe Lightroom 2, running it felt very similar to wading through syrup.

Things changed this week, though. In anticipation of my upcoming expedition to Utah and in celebration of the start of a new tax year, I’ve bought my first laptop and fully intend this to become my main digital darkroom machine with the HP relegated to the role of file and print server. The new machine is a Dell Studio XPS 16 with a Core2Duo CPU, 6MB L2 cache, 7200rpm hard disk, upgraded graphics card and 64 bit Windows 7. It’s an absolute screamer (for a Windows machine)!

I’ve installed 64 bit versions of all my software where these versions are available. Lightroom 2, Lightroom 3 Beta, Photoshop CS4 and Photomatix Pro 3.2 all come in “double-wide” versions and the performance difference compared to my old machine is staggering. Lightroom exports happen in a very small number of seconds and, even more impressively, I can view the effects of slider changes in the Develop module in real time! No more “move the slider, wait for the effect”!

Couple this superb performance with the fact that the display is absolutely gorgeous (allegedly having a colour gamut encompassing the whole AdobeRGB space) and I’m definitely enjoying this as my new mobile office.

Oh, I almost forgot – it’s red too.

The game for this year’s FIRST Robotics Competition, “Breakaway”, has been announced and it looks like it should be a fun one! Teams now have a frantic 6 weeks to design, build and program their machines for the competition.

If you have read this blog for any length of time, you will know that I highly approve of Dean Kamen’s FIRST organisation. This weekend sees the kick-off for the 2010 FRC competition and the start of the 6 week design and build period leading up to regional competitions in March and April. If you can make it to any of the events, I would encourage you to go along and see just how cool they are. Talking of cool, they even managed to get Neo to do a PSA spot for them this year…

Edit: A more knowledgeable colleague just pointed out that the Keanu PSA was actually prepared for last year’s competition. That answers my question about why it doesn’t contain any footage from the 2009 competition. Oh well…

I got into a conversation with Guy Tal, Jim Goldstein and Pete Carr on Twitter this afternoon and it quickly became clear that it wasn’t going to work in 140 character chunks so I reckoned a blog post would be in order. This is something I’ve been meaning to write about for a while now so I’m happy I now have an excuse.

The issue under discussion related to the relationship between the number of bits used to store a High Dynamic Range image and the light levels that the image can store. An HDR image is typically thought of as being represented with 32 bits per color component (32 bits of red, 32 bits of green and 32 bits of blue for every pixel). Some of the most common HDR file formats actually use less than 32 bits per pixel but, regardless, the question comes up about what each of those bits represents and how the number of bits dictates whether an image is “High Dynamic Range” or not.

I had always assumed that the dynamic range of an image depended to some extent on the absolute maximum light level that could be recorded. On reading Christian Bloch’s rather good “HDRI Handbook” last year, I was surprised, however, to read that the dynamic range is defined instead in terms of the ratio of the number of discrete values (2**32, for example if we’re talking about a 32 bit number) a measurement can represent divided by the smallest measurable difference (the change in the signal represented by 1 least significant bit). Nowhere in this calculation does any absolute value appear – the dynamic range is a ratio of the largest and smallest values that can be represented but it says nothing about the actual quantities that those values represent.

This didn’t seem intuitive to me. Where was the reference? How would you know how bright a given pixel would be? Thinking about this for a while, however, it became clear that I was forgetting to take into account the camera as a whole rather than merely the sensor.

The camera’s sensor has a fixed, maximum signal that it can record and a certain number of bits of resolution. For example, a 12-bit sensor can differentiate between 2**12 (or 4096) different levels of red, green and blue light. At some absolute level of light, the sensor saturates and outputs its maximum value. If you increase the light level falling on the sensor above this amount either by keeping the sensor exposed to the light for longer or by increasing the brightness of the light falling on the sensor, you get no new information and the signal stays saturated.

Thinking about this from a photography point of view, however, this is exactly as you would expect and corresponds to overexposure. In this case, we close the aperture, reducing the brightness falling on the sensor, or speed up the shutter to reduce the total amount of time the sensor is exposed to the light. These changes reduce the total amount of light falling on the sensor and allow us to take another, hopefully correctly exposed image. Although the maximum light level the sensor sees has dropped and it no longer saturates, it still records 4096 different levels falling on it. The recorded dynamic range is the same but we’ve shifted the recorded values so that all of the actual image brightness levels fall within the recording capabilities of the sensor.

This is exactly as you would expect in a film camera too – if the film is overexposed, it “saturates” to opaque and you can’t store any more information (there is the complication of logarithmic vs. linear response here but let’s gloss over that for now since it’s not really relevant to this discussion). In these cases, you reduce your exposure to get the amount of light hitting the film such that you don’t saturate the medium.

Considering things this way, it is now clear (to me at least) that the absolute value of light represented by the sensor’s (or film’s) maximum output value is irrelevant. The photographer adjusts the exposure to ensure that the brightest highlight in the image is at or just below the sensor’s saturation point and, hey presto, you end up taking maximum advantage of the sensor’s dynamic range. The absolute maximum light level that will cause the sensor to saturate is, therefore, related to the sensitivity and not the dynamic range. If the sensor saturates with very little light hitting it, we end up with a high sensitivity (or high ISO) sensor that allows us to record images in lower light than one which saturates at higher light levels.

So how does this tie back in to the normal photographic system of defining exposure in terms of stops or EVs? For every bit you add to a sensor, the number of values it can represent doubles and, as a result, its theoretical dynamic range will also double (forgetting about noise which reduces this somewhat). Thinking about exposure calculations, you know that increasing your exposure by a stop also doubles the amount of light hitting the sensor. There is, therefore, a direct correlation between 1 stop and 1 bit. Reducing your exposure by a stop divides the sensor output in two or shifts the value one bit to the right. Adding a stop to your exposure does the opposite, doubling the sensor output or shifting the value one place to the left.

Using the bit shifting idea, we can, therefore, get some idea of how much dynamic range an HDR image may have based upon how we recorded it. I typically use 3 images bracketed 2 stops apart when shooting HDRs. If the original scene contains very bright highlights or lots of dark shadow areas, I will use more brackets but 3 is usually enough. My Nikon D90 has a 12 bit sensor and, at low ISO values, has close to .12 bits or 4096 levels of dynamic range. Adding +2 stops to the exposure has the effect of multiplying the sensor output values by 2**2 or 4 or shifting them 2 bits to the left. Taking away 2 stops has the opposite effect, dividing the output values by 4 or shifting them 2 bits to the right. When I combine all three exposures, -2 stops, 0 stops, +2 stops, into an HDR image, therefore, the maximum dynamic range I could possibly record would be 12 + 2 + 2 bits or 16 bits of information. Even though I may save this image in a 32 bit file format, I’m not actually storing pixel values that extend from the lowest to the highest possible value in the file format. Regardless of the fact that I’m not using the full dynamic range of the recording system (the 32 bit pixel component representation), I still have 16 bits of information per colour component which is twice the number of bits I would have had if I had saved in JPEG and 4 bits (or 16 times) more than I would have got from a single RAW file from the camera. The dynamic range of my image is something like 65536:1 versus 256:1 for JPEG or 4096:1 for my 12 bit RAW format.

Confused? If so, leave a comment and I’ll try to clarify this a bit!

I’m trying something new this morning. This post is being typed and posted from somewhere around 30,000 feet above south eastern Oklahoma thanks to the wonders of American Airlines new inflight WiFi. For $10, I get a broadband connection from my airliner seat – how amazing is that? I had expected a dial-up type experience but the bandwidth is actually rather impressive – streaming video, at least of the standard definition variety, plays smoothly and I’ve noticed no delays significantly longer than I would nornally see at home or in the office. Obviously, I have no idea how many people are using the service but I have to imagine that a full flight such as this containing mostly business people would be using the service pretty heavily.

The downside of this, of course, is that flights are no longer an opportunity to read, play games or generally do non-work things (like post to blogs?) during work time. I guess I had better get back to the email….

When I was working in digital settop box software, one of the biggest problems we used to have was ensuring that the audio and video stayed synchronised with one another. Arguably there’s nothing worse in TV playback than having the audio run slightly ahead of the video – even a 10mS audio lead is really noticeable and really messes up your viewing experience. Interestingly enough, if the audio is significantly behind the video (up to 100mS or so), you don’t worry so much since your brain is used to compensating for the delay caused by the speed of sound being so slow. I digress, however…

This morning, my 6 year old son introduced me to some videos on YouTube that give a whole new meaning to audio/video synchronisation. These animations do a superb job of synchronising the animated video content with the soundtrack. They are completely mesmerising. Take a look at this one then head over to YouTube for various others. You can buy DVDs at www.animusic.com.

I just received a rather odd message via Flickr mail. It was from an Italian photographer advertising one of his images. It was also in Italian, a language which is not my best (I barely know how to order beer in Italy let alone carry on a conversation about photography). As I usually do, I headed over to Yahoo Babelfish (or http://babelfish.altavista.com as it used to be – oddly, the “legacy URL” still works nicely) and pasted in the text. The final paragraph of the original message was:

“Flickr è la migliore applicazione per la gestione e la condivisione di foto online. Se sei curioso di sapere per che cosa lo utilizzo, guarda il mio profilo oppure naviga tra il mio album.”

…and the web site translated this as:

“Flickr is the best application for the management and the sharing of photo online. If you are curious of knowing for I use it what, watches my profile or is annoying between my egg whites.”

I do find unsolicited messages like this rather annoying (though I think this was the first example of Flickr spam I have seen) but, in this case, it ended up having great entertainment value and didn’t disturb my egg whites at all.

A couple of days ago, I heard about a new company in California offering an excellent service to photographers. Fotomoto offers the ability to provide print purchases directly from your web site without the need to tag, keyword and upload high resolution images to yet another site. I applied to join their beta program and was invited this evening.

Setting the service up on my photoblog was trivially simple. After logging on to the Fotomoto site and setting pricing for the various print sizes, I copied a small block of JavaScript and pasted it into the footer of the photoblog page. Immediately, all the images have a couple of new, unobtrusive links added beneath them, one to allow print purchases and the other offering to send an eCard containing the image.

Some of my images were taken in places which do not allow commercial sale of photographs taken there and one was taken with an old digital camera whose resolution was not really up to producing high quality prints so I had to return to the Fotomoto control panel to indicate which images were not for sale. Overall, the installation and customisation was incredibly easy.

So far, this looks as if it should be a superb service. I’ll purchase a couple of test prints and report back once I have checked out the print quality. Assuming it’s on a par with ImageKind, I will likely move all my online print purchasing there since FotoMoto do not charge an annual fee (they take a 15% commission on each sale instead which, given my current sales volume, works out better for me)