BetterPhoto Q&A
Category: New Answers

Image Size

To love this question, log in above
August 11, 2008

Your camera's sensor is apparently 12 MP; that is, there are 12 million [+/-} pixels crammed onto its surface. But, when you upload your images to a computer and check out the resolution in Image>Image Size, you see the image dimensions in pixels [which, when multiplied together, should equal 12 MP.]

You should also see the image dimensions in inches. This is an easily understood figure and reflects the Resolution in the next box.

I think you have a Canon camera; you are most probably looking at a Resolution Value of 72 ppi. That's the resolution you can see on your monitor. Multiply that resolution by the image dimensions and you should get the same pixels values as depicted in the upper section.

Make sure you uncheck the Resampling Box and change the resolution to 300 ppi [what you'll need for proper printing.] Your image dimesions should change to 8X12, if you camera works like my Canon 30D and you're using one of the Photoshop products.

With the Resampling Box unchecked, the size of the image [shown next to the words "Pixel Dimension"] should be some number with an "M" as its unit. This is for megabytes "of data," not pixels. And, it doesn't change so long as you keep the Resampling Box unchecked.

To love this comment, log in above
August 11, 2008

Have fun!

To love this comment, log in above
August 12, 2008

To love this comment, log in above
August 12, 2008

When I take a picture using hi-Res JPEG + RAW, the JPEG image [as uploaded] is +/- 3.6 megabytes. The RAW file is +/- 8.3 megabytes. The difference reflects the camera computer's use of the JPEG algorithm which tosses away data.

If I load the JPEG file and uncheck the Resampling Box, as described above, any editing I do might impact the file size [megabyte value.] For example, say I set the image size at 8X12. Then, if I crop the image to dimensions of 4X6, I will throw away 3/4 of the available data. This is one of the most significant issues you'll encounter when editing. Now, to resize the cropped image to 8X12, you must check the Resampling Box and allow the program to interpolate [and create data.]

To some degree, this may or may not impact image quality.

But, the problem you described originally still seems to suggest that when you upload picture file and open it for the first time, you open an image with a 72 ppi [screen] resolution. It appears sharp on screen but if you print it at 72 ppi, the print will be pixelated.

If you feel your image on screen is not sharp without any editing, several other things are posssible. Obviously there could be a problem with the camera's focusing "system." Have your camera checked by Canon.

Another possibility - you may be "the problem" by either failing to hold the camera steady [correct this using a tripod] or by using the wrong shutter speed or ISO setting. You can check this out by setting the ISO to 100, mounting the camera fitted with a "cable release" on a sturdy tripod, and shooting at a subject with distinct horizontal and vertical lines. Download your camera [don't rely on the LCD] and check for image sharpness.

Remember one other thing - essentially all digital images must be sharpened using an editing program. The Tool, if you aren't using it, is called the Unsharp Mask [in the Photoshop family of editors.]

To love this comment, log in above
August 12, 2008

How your 12 megapixel camera relates to file size:

As you know, your camera sports a digital sensor chip with a surface covered with light sensitive sights. When the shutter clicks, light from the outside world momentarily flashes onto the surface of the sensor. Light energy cause each sight on the sensor to gain an electric charge in proportion to the intensity of the light strike (photon). This charge will make-up one picture element, the smallest possible component of an image otherwise known as a pixel.

Based on the magnitude of the charge the camera’s software assigns a numerical value that best represents the charge. Most cameras are capable of detecting 256 different light energy levels (some quite a bit more). Thus the software will transmit an eight digit number using only the digits one (1) and zero (0). This is binary code we call digital. It takes eight digits to transmit a specific value between 0 and 255 (256 discrete values).

Since you will be taking pictures in color, each pixel will be assigned values that represent the intensity of the three primary colors of light i.e. red – green – blue. Since you camera sports a sensor chip with 12 million (12MP) sights, the digital data generated by one exposure is a whapping 12 million bytes times 3 = 36 million (36MB) to convey the image’s intelligence.

That’s a lot of file space requirements for just one picture. It’s possible to compress this data into a smaller space. Consider a big thick book filled with words. If you could agree on a shorthand method to replace some of the printed stuff i.e. repetitious words – spaces – punctuation marks – white spaces etc., you could maybe concentrate the size of the book. Enter JPEG (Joint Photo Experts Group). They wrote software that looks at the data contained inside the picture and they invented a code for repetitious data like expanses of blue sky and mundane vastness of sameness. The software consists of math algorithms that cast out data. With this algorithm about 90% of the data can be discarded before the image looses high quality. This method is called lossy as data is actually discarded. Using JPEG notation the files are likely shaved back to between 1.3 ~ 2.5 MB.

For those who can’t stomach lossy compression, you can save in RAW. This is a propriety algorithm that does little in the way of compression. The file size is likely 10 ~ 18 MB. The data is so complete you can use impressive editing software routines to correct for under – over exposure – white balance and a whole host of adjustment and modifications you never even dreamed of.

I hope this helps connect the dots i.e. chip pixel sights to file size.

Alan Marcus (marginal technical gobbledygook)
ammarcus@earthlink.net

To love this comment, log in above
August 15, 2008

To love this comment, log in above
August 16, 2008