Archive for December, 2008

Happy New Year from Japan

Hey Guys,

Just wanted to wish you all a Happy New Year and a great 2009. This is a shot from my recent trip to the hot springs. It’s the first time I’ve seen Mt. Fuji in its entirety. The mountain is known to be rather camera shy (right Kiko?) and after living here for 4 years, it was great to finally set my own eyes on its snow capped slopes. I know “resolutions” suck, but I want to go on record to say that I plan on getting better at DSLR photography, blogging more here and at my design blog, and doing more personal projects in order to keep my skills sharp or acquire new ones.

Anyway, hope you guys have a great holiday. The Cadre Blog will definitely keep going in 2009 (unless Kiko decides to chuck his server…).




So, here’s a shot from the new trailer of Wolverine. This is where you begin to notice that someone isn’t paying attention or has only been following recent comic book versions of his admittedly now convoluted history. A young Logan with…that’s right…bone claws. Yeah, because that’s what makes him a mutant. The trailer sells the who’s who of comic book characters featured or making “wink-wink, nudge-nudge” cameos in the film. Can you spot them all? Anyway, plenty of places you can find the trailer on the internets. Carry on!


Dragonball Evolution

I wasn’t a huge Dragonball fan, but this looks beyond terrible.

It’s been a while…

…since I posted anything. So here are a few Star Wars cards I did for SOE. Only 2 have been approved, so until Lucasfilm lets that one go, keep these on the down low. These were way more trouble than they were worth to do…

tusken-flesh-hunter-swg-set02-fdsm.jpg black-sun-minion-swg-set02-fdsm.jpg gamorrean-battleaxe-swg-set02-fdsm.jpg

影像處理 photography process

I initially was going to write this post to specify how I sharpen my photos in response to Ben’s request. I’ve decided to bore you even further by documenting my entire photography process, from the physical action of taking the photo to the last post processing (sharpening).

Now, I like to take all kinds of photos, but it seems my landscapes gets the most attention, so I’ll go over that process. Maybe you can gleam some helpful info, or if you guys have any comments, I’m all ears.

I often take my landscape photos during sunset or sunrise. Sunrises are harder for me because public transportation are usually not in service by then, so I sometimes have to stakeout the night before. I tend to carry a compass with me to indicate where east and west is to get any idea where the sun will be located. While I’m waiting for the right lighting condition, I scope out the location, seeing which areas have interest and where the sun will be in relation.

Now for the actual shot. I set my aperture low, usually F/11 to make sure both my foreground and background are sharp. I also use a tripod to lock down my shots. Now, this is a situation I’m sure most of you have experienced. When taking a photo with a bright background against a dark foreground, either the foreground will have the right exposure and the background will be blown out, or the background will have the right exposure and the foreground will be completely black (like taking a picture inside a room, with the window in the background being completely white). Traditionally, photographers taking landscape pictures will use a graduated filter to balance the bright sky against the darker landscape. I use a different approach called HDR (High Dynamic Range). To have the necessary material, I bracket my shots, which for me, means taking different exposure of the same shot. This insures that I have detail in the bright areas and detail in the dark areas.


I then use Photomatix to combine these bracketed shots into a single file. Photoshop also has this feature, but I like Photomatix results better. Some people may say that I’m cheating. I’d like to say that how the shot looked in real life is between the raw images and the final HDR image. Keep in mind that the human eye has a far greater range of levels than a camera. That’s why when we look at a scene in real life, we can see all those details, but the camera cannot capture that.


After creating the HDR image, I bring into Photoshop for any further editing. I usually leave sharpening as my last process.


Now, when you sharpen an image, the idea is to sharpen the areas that have sharp contrast, like edges, and not sharpen broad surfaces, like skies and skin, because this will sharpen any noise and make your images look grainy. To accomplish this, you will need to create a mask that will only sharpen the edges. Look at the RGB channels and see which channel has the best contrast. Copy that and paste it into a new layer.


Run a surface blur filter to remove any residual noise or subtle contrast.


Tweak the levels if necessary to produce a better contrast.


Next, run a find edge filter (I never thought this filter would be useful). I think you can now get the idea of the areas that will receive or not receive the sharpening treatment.


Clean up any areas that you don’t want to any sharpening effect. Once satisfied, run a very low Gaussian blur (this will make your sharpening effect not look so obvious) and again tweak the levels if necessary (I try to have large areas of pure white and pure black).


Inverse this image so that now, whatever is black will not be sharpened and whatever is white will be sharpened.


Going back to your actual image, create a duplicate layer and run a sharpening filter. I usually use the Unsharp mask filter. While playing with the settings, you’ll see how your image is being sharpened and any noise being exaggerated. Once satisfied, apply the mask you created to the layer that received the sharpening treatment.


I’m not sure if you can see the difference. The left has no sharpening, the middle has everything sharpened, and the right has the sharpening with the mask applied. Hopefully this quick rundown was of some interest.


“edit” In response to Kiko’s question, I’m elaborating a little further on HDR. For best results, you need at least three different exposures. Now, there are times when objects are moving in the scene, like people, cars, waters, etc. Sometimes the HDR product will look fine, but often, you’ll get a ghosting effect where you can see the object moving through time. Sometimes, you’ll get some weird artifacting. Take this HDR shot for example. The left side is the HDR treatment with 3 bracketed shots. At first glance, the water looks passable, but I wanted more definition in the water, like on the right side.


This is where the power of RAW images comes in handy. From my bracketed shots, I take the medium one, and through Photoshop, I “artificially” bracket my shots. RAW images still contain enough dynamic range to produce pseudo bracketed shots. Note, this step is also a way to produce HDR shots if you didn’t or couldn’t take bracketed shots at the time of your shoot. Because my actual bracketed shots are spaced by 2 stops, I’m reduplicating that exposure here (note the exposure slide bar).


Once I’ve created my pseudo bracketed shots, I save them out as TIFFs and apply my HDR process, using the same settings from my HDR process on the real bracketed shots. Once satisfied with the way the water turned out, I mask that and apply it over my original HDR shot.


I generally try to avoid this particular process of creating HDR shots from pseudo bracketed shots because whatever noise that’s in the original shot will be significantly more apparent. Look at the shot above, and you’ll see some pretty bad noise in the darker areas. Remember, you’re combining three exposures from the same shot, which means you’re amplifying the same noise by three times, whereas when you combine exposures from an actual bracketed shot, the noise you’re combining will be different from each shot, essentially smoothing out the noise.

Glad you’ve found this to be of some interest. If you have any further questions, I’ll keep updating this post.

Photography Advise

For the Cadre members with DSLRs, have a question.

What is your preferred metering mode?
I am getting used to my D90 and I love the results, but I find that I am having some so-so results with the Matrix Metering mode. This is probably mostly my fault as my old film camera and even the Cybershot I got from Kiko were center-weighted or spot metering cameras; now that I have Matrix Metering, I sometimes get shots that look great, especially outdoor shots, but in some cases where I want dramatic mood or something, I usually get a lot of underexposure and have to futz around locking off focus & exposure and reframing. Unfortunately, this means losing a moment sometimes… Anyway, let me know about your experience with Matrix Metering and when exactly you dial that in (if ever). Thanks!