When “megapixels” does not mean what you might think it means

If your camera has 48 megapixels, how many picture elements do you think your highest resolution photo has?

This specification table for a Sigma camera helps to understand a bit about how “megapixels” may or may not mean what you think it means.

From the table, this is a 48 megapixel camera (which is an amazing camera).

But down below, you can see that the highest resolution images are 4704 x 3136 picture elements wide x tall. Multiply those together and you get 14.76 megapixels, not 48 megapixels. Why?

Because megapixels is counting the sensors of which there are 3 at each pixel. Multiply 14.76 x 3 and you’ll come up with 44.2 million image sensor locations (which is close enough since the entire sensor surface is not used in most cameras). The Foveon sensor uniquely has all 3 sensor sites, per pixel stacked on top of one another.

There’s nothing wrong with these numbers – only a potential misunderstanding of what the term “megapixels” actually means.

SIGMA DP2 Merrill : Major Specifications
Image Sensor
Foveon X3® direct image sensor (CMOS)
Image Size
23.5×15.7mm
Number of Pixels
Total Pixels: 48MP
Effective Pixels: 46MP(4,800×3,200×3)
Aspect Ratio
3:2
Focal Length
30mm
35mm Equivalent Focal Format
Approx. 45mm
Maximum Aperture
F2.8
Number of Diaphragm Blades 9
9 Blades
Lens Construction
8 Elements in 6 Groups
Minimum Focusing Distance
28cm
Maximum Magnification Shooting
1:7.6
Storage Media
SD Card / Compatible with SDHC, Multi Media Card
Recording Mode
Lossless compression RAW data (12-bit, High, Medium, Low), JPEG (High, Medium, Low), RAW+JPEG, Movie (motion jpeg), Voice memo to still images (10sec.)
File Size / Still
RAW
High
4,704×3,136×3 (Approx. 45MB)
Medium
3,264×2,176×3 (Approx. 24MB)
Low
2,336×1,568×3 (Approx. 12MB
JPEG
High
Fine 4,704×3,136 (Approx. 10MB)
High Normal 4,704×3,136 (Approx. 5.6MB)
Basic 4,704×3,136 (Approx. 4.2MB)
Medium
Fine 3,264×2,176 (Approx. 5MB)
Medium Normal 3,264×2,176 (Approx. 2.7MB)
Basic 3,264×2,176 (Approx. 2MB)
Low
Fine 2,336×1,568 (Approx. 2.5MB)
Normal 2,336×1,568 (Approx. 1.4MB)
Basic 2,336×1,568 (Approx. 1MB)

Let’s take a look at another camera, the Canon 5D Mark II.

Type
High-sensitivity, high-resolution, large single-plate CMOS sensor

Pixels
Effective pixels: Approx. 21.1 megapixels

Total Pixels
Total pixels: Approx. 22.0 megapixels

And then look at the image sizes

File Size
(1) Large/Fine: Approx. 6.1MB (5616 x 3744 pixels)

(2) Large/Nomal: Approx. 3.0MB (5616 x 3744 pixels)

(3) Medium/Fine: Approx. 3.6MB (4080 x 2720 pixels)

(4) Medium/Normal: Approx. 1.9MB (4080 x 2720 pixels)

(5) Small/Fine: Approx. 2.1MB (2784 x 1856 pixels)

(6) Small/Normal: Approx. 1.0MB (2784 x 1856 pixels)

(7) RAW: Approx. 25.8MB (5616 x 3744 pixels)

(8) sRAW 1: Approx. 14.8MB (3861 x 2574 pixels)

(9) sRAW 2: Approx. 10.8MB (2784 x 1856 pixels)

With the Canon 5D Mark II, the image resolution of 5616 x 3744 is 21 megapixels, yet the camera likely has more “photosites” within the imager.

This leads to the conclusion that megapixel camera ratings do not necessarily  mean what you think they mean. And that the megapixel count may not be nearly as important as some think it is. Ultimately, lens quality, use of a tripod, sensor noise characteristics and noise removal algorithms, sharpening algorithms and general photographic skill may  be much more important in image resolution than the megapixel count – however it may be defined.

 

Enhanced by Zemanta

3d photos taken with a single camera

While hiking I took several 3D photos using a single Canon SX1 IS camera. This works by taking a photo and then sliding the camera a couple of inches to the right and taking the same photo again. The left and right images are then combined with Stereo Photo Maker to create a 3D image.

Obviously, when handheld, camera pointing is not perfectly aligned in the left and right images. Much of the alignment can be cleaned up in Stereo Photo Maker but there will be little artifacts, like lens barrel distortion, still present. But still, this is an interesting way of easily photographing 3D photos of stationary subjects.

(Click on any image for full size – all are in red/cyan stereo anaglyph format)

[singlepic id=172 w=640 h=480 float=]

 

[singlepic id=173 w=640 h=480 float=]

 

 

 

Enhanced by Zemanta