商用無料の写真検索さん
           


Decoding the raw files : 無料・フリー素材/写真

Decoding the raw files / selkovjr
このタグをブログ記事に貼り付けてください。
トリミング(切り除き):
使用画像:     注:元画像によっては、全ての大きさが同じ場合があります。
サイズ:横      位置:上から 左から 写真をドラッグしても調整できます。
あなたのブログで、ぜひこのサービスを紹介してください!(^^
Decoding the raw files

QRコード

ライセンスクリエイティブ・コモンズ 表示 2.1
説明Previous: Raw file storageReading the files and assessing the amount of correctionWithout going into the details of how to correct exposure and colour while decoding a raw image (it is really a long story, if all possible effects and conditions are to be accounted for), here's an example of an imperfect exposure that I had to correct before throwing this frame at Hugin. It looks pretty awful, if you have never looked at a raw file before. It is certainly far from perfect: it has been underexposed by at least 1 EV. You will see that this is almost OK; we still have 11 bits to play with.I prefer to have my frames underexposed than overexposed, if possible. If this frame was overxposed by 1 EV, or even less than that, I would have to throw it away.Now, remember Nautilus — the file browser I mentioned on the previous page?It is configured to call ufraw on any photographic raw file, when I click on one. At least, so it is in Ubuntu, and it is easy to achieve the same effect in other systems. This combination allows me to very quickly run through the images I want to work with, and to figure out the corrections they require. My goal at this stage is to come up with the numbers describing the necessary corrections, rather than do those corrections and save the resulting files. In other words, it is a read-only exercise. I tweak the numbers until the image starts looking right. Or, in the case of panorama frames, approximately right. Because Hugin will command an exposure adjustment to make the frames blend well, any extra time spent making them look perfect at this stage will be wasted. All I need is to try and spread the pixel levels as widely as possible across the available dynamic range. Doing so will minimise the loss of information during the subsequent adjustments and blending.Now, the above picture shows how the image looks with the standard settings for my camera. What are the standard settings? No exposure compensation and the gamma curve that corresponds to my camera's profile. In this case it is a built-in profile, which I learned to trust over the years. The gamma curve for this camera has the gamma of 0.45 and linearity of 0.1. The linearity parameter is a rough and ready hack that allows to selectively vary the lightness of the darkest values in the image without having much effect on the mid-range and brighter values. And that is a frequent need. The linearity parameter specifies which fraction of the curve to make linear.My other camera works best with the standand gamma of 0.36 and linearity of 0.02. The standard values are obtained by testing the camera in the best lighting conditions (diffuse sunlight at mid-day), so the actual value you might need to correct an image will vary with the camera and with the time of the day, as well as possibly other condtions. But starting with the standard settings is a good way to estimate the amount of correction.A quick look at the histogram (in the bottom left corner), reveals an unused space at the bright end of the range. That space needs to be filled. At the other extreme, pressing the Indicate button will highlight the areas that will be hopelessly underexposed in the output image, unless we do something. The reds in the underexposed map indicate that the red channel fell through the floor, while the green and the blue channels are still in range. The yellow spots contain the pixels where both the red and the green channels are underexposed. It appears that the blue is fine across the entire image; if it was not, and all three channels were underexposed, the fact would be indicated by black spots.Now that we know what's wrong with the lightness distribution in all three channels, we can nudge it towards perfection (and there is plenty of room in that direction).Next: Adjusting the exposure level
撮影日2010-08-03 21:16:32
撮影者selkovjr , Chicago, Illinois
タグ
撮影地


(C)名入れギフト.com