iMonAd [Afbeelding bedrijfslogo]


[In aanbouw]


General information

iMonAd turns your iPhone into a inexpensive RGB analyzer.

In the following explanations not all possibilities are given, there are more combinations of settings possible.

Please be aware of the color of your light source. When pointing direct into a light source your will measure the source direct (be aware of high intensities). When you measure a surface with reflected light, the results will be a combination of the source and the color shift of the surface. For our eyes (they will adapt) everything we know is white will be white, but they aren't. Only specified grey cards will reflect all colors correct. When you compare monitors, they will act as light source and are not affected by reflective color shifting. When you compare colors under a light source, use the same light source for every object. This means the same light source, not a comparable (for our eyes the same) source.

This app can not and will not replace a professional color analyzer like the DataColor Spyder4.

This manual is valid for V1.20

Run in normal screen

The upper dark gray window contains the exposure value and average gray on the entire camera image.
The gray value is the standard value calculated over the 3 color channels.
The next 3 values represent the exposure value in different units. The values are referenced to ISO100.
This EV value can be used in a table to get the exposure time for a certain aperture and vise versa (when you have to do this often, please look for iZoner).

The second window, showing RGB, shows the three separate color channels (red, green, blue) as figures.
It is even possible to compare these color to standard tables to get the name of the color (although they can be gray shifted, i.e. all channels have an equal offset).
The second row in this window, gives you the color in a different system.
These values are calculated over the entire image.

The third window shows the selected WP threshold. This is the minimum level a pixel (all three color channels combined) must have to be used.
Why is this done? Statistically is it shown that the brightest pixels of an image are the closest to white.
So when you want to perform a test on color balance, it is the best to use only these pixels.
The threshold level ranges from 0.5 to 1.0 (RGB 127/127/127 to 255/255/255).
When you set this value to 1.0, only exact white pixels are analyzed (and will be 1.00 in neutral offset). This is a not so useful setting, a reasonable value is around 0.7
The WP# figure shows the number of pixels that are above the threshold level.

The next and largest window needs some more explanation.
Every row consists of three columns, one for each color channel RGB.
Only those pixels above the white point threshold are evaluated.

The first row indicates the 70% bandwidth of the histogram. What does this mean?
In the histogram the highest level for a color channel is searched. Then from this point in both directions (darker and brighter) the points are searched where the level is 70% of the peak. The separation of these two points is the bandwidth. The narrower the bandwidth, the more uniform in brightness a color is reflected.

The second row is like the RGB separation in the second window, except for the selection of the pixels.
In the first case all pixels are used, in the latter only them above the white point threshold. Under normal circumstances these values reflect more accurate figures.

The next row shows the relative gain. The highest channel is referenced as 1.0000 and the others must be multiplied by the displayed figure to get the same RGB level.
The closer these values are to each other, the whiter the pixels are.

The last row shows the point in the histogram where the integral area is 50% of the total area.

The big figure under "RGB neutral offset" shows the average (square root from the sum of all three channels powered to two) of the three RGB@WP values. 
The more this figure is to 1.00 the better it is when you meter a pure white (every white sheet of paper is different and certainly not neutral white) or better a professional gray card.

The last big number is the estimated white balance color temperature. You can check e.g. your monitor settings from a white window, normally your monitor should be something around 6500K
Be aware that this only works for neutral gray or white surfaces without any color shift. You can also check the change in color when you use a light dimmer. In this case use a piece of white paper and observe the shift while dimming. 

The small black square with the colored bars shows the values of the three color channels from 0-255 and the gray bar bar runs from 0 to 16 EV for the image.

When you tap on the small camera view the current values for "RGB neutral offset" and "color temperature" are saved. From now on these values are subtracted from the live values. In front of the displayed label "rel" is shown and the values show the difference. The live values can be restored by tapping the view again.

Run in histogram screen

Some of the following pictures are taken with iZoner, these show the same graph part.

This view is the most difficult to explain, but contains for the eager photographer lots of information.
Let's give it a try.

In the upper left corner the live view of the camera is visible.

When you use a single source light the reflected light on a gray card should show three different channels close to each other. When you turn on other sources with different physics (LED and a normal lamp), you will see that the color channels are drifting apart. When you take a picture with your camera, using this light setup, there will be a color cast in your photo.
It is advisable to change the lights or put a color filter over one of them (this can be checked with iMonAd).

To do this test, even a white piece of paper will work (as long as there is no color in it), but the best is a standard gray card.

If you think the image is constantly changing in brightness, you can freeze it by tapping on the picture. Tapping again starts the live view.

What do we see here?
This is a wide tonal picture, all values from black (=0) to white (=256) are present.
The total dynamic range of the iPhone sensor is used. We also see that are lots of dark parts in the scene.

Just one more last example.
In this picture we see three distinctive parts. Left a black peak (border of PC screen), on the right side a lower/wider peak representing the screen itself.
In the middle a wall reflecting three separated color peaks, indicating the is a color in that paint in combination with several light sources.

To get more feeling with this view and learn to see colors/tones, you can try lots of different scenes while waiting at the doctors or dentists.

On this view there are some figures to help you.

Peak: highest value for a color channel


iMonAd doesn't require lots of settings, most of the time you will not touch it anymore.
As not all iPhones are exactly the same (or the sensors are from a different batch), there can be a difference in results.
When you think this is the case you can adjust the color offset per channel. When you measure a gray card under daylight (6500K) conditions all three channels should be very close to each other.

The white threshold level sets the gray level (a combination of RGB) a pixel must have at least to be used for the calculations in the lower part of the run screen.

On the run screen you find an EV level. With this figure you can find in tables the exposure time/aperture combination at 100ISO. When you compare this to your camera there might be a difference. If you want to, you can adjust is to get the same figures. This is not necessary for the color analysis.


Tapping on this screen will bring you to other worlds.

Globe : to this help page.
Twitter : start following us on Twitter. You account must be set up properly.
View our... : to our other App in the store.

If you enable "show hints", tapping on the run screen will show a short information on certain parts.

Be aware that for the other worlds, an internet connection is used and can be charged by your provider.



Start ] Omhoog ]

U kunt een e-mailbericht met vragen of opmerkingen over deze website verzenden aan
Copyright 2007-2012 Alfa Kilo Beheer B.V.
Laatst bijgewerkt: 25 juli 2012