A short quiz on OpenCV and simple image processing operations

Adrian F. Clark

1. An image which is under-exposed will have a histogram that

   • has no significant peaks

   • has a peak at low grey-level values

   • has a peak towards the middle
   • has a peak at high grey-level values

   An under-exposed image appears dark and hence has many dark pixels. This will result in a histogram where most of the bins filled correspond to low grey levels; hence, the histogram's peak will appear at low grey-level values (normally towards the left of the histogram).

2. What is contrast stretching?

   • the range of grey level values
   • the standard deviation

   • expanding the available grey levels to fill the range used for displaying images

   • re-scaling the data to a given mean and standard deviation

   Contrast stretching is expanding the occupancy of the grey level bins in the histogram, usually for display.

3. OpenCV was initially developed in

   • Russia

   • USA
   • UK
   • Australia

   Perhaps surprisingly, the early versions of OpenCV were developed in Russia, at Intel's research center in Nizhny Novgorod.

4. In OpenCV, if Y indexes lines, X columns and C channels, the right nesting >of loops is

   • X inside Y inside C
   • Y inside X inside C

   • C inside X inside Y

   • C inside Y inside X

   The C subscript should cycle fastest in all languages you will use OpenCV with, followed by the X subscript. The Y subscript should cycle slowest. This order mimics the way in which the pixels are stored in memory and thus minimises cache misses etc.

5. Histogram equalization involves:

   • using the inverse of the histogram as a look-up table
   • using the reciprocal of the histogram as a look-up table

   • using the cumulative histogram as a look-up table

   • using the histogram as a look-up table

   To perform histogram equalization, one forms the cumulative histogram, which has near-vertical shape in regions where there are many pixels, and uses that as a look-up table, stretching out the contrast there.

6. After contrast-stretching, an over-exposed image which has only a few grey levels occupied will have a histogram that

   • has a nice peak of grey levels in its middle region
   • has all grey-level bins roughly equally occupied

   • tends to show small spikes across the entire range of grey levels

   • has a concentration of large values near high grey-level values

   A poorly-exposed image (either under- or over-exposed) which is contrast-stretched will normally exhibit only a few occupied bins, and they will be equally-spaced across the entire range of grey levels, giving it a `spiky' appearance.

7. With additive colours, yellow is a mixture of

   • red and green

   • red and cyan
   • cyan and red
   • blue and green

   The additive colour primaries are red, green and blue, ruling out some of the possibilities given. Blue and green combine to give cyan, while red and green (perhaps surprisingly) combine to give yellow.

8. For an image of unsigned bytes, how can it be contrast-reversed?

   • subtract 255 from each pixel
   • negate each pixel
   • subtract each pixel from 127

   • subtract each pixel from 255

   255 - 255 = 0 and 255 - 0 = 255, so subtracting from 255 does this.

9. What is plotted along the abscissa (x-axis) of a histogram?

   • grey level

   • mean value
   • standard deviation
   • number of occurrences

   The x-axis of a histogram plots the grey level; the y-axis plots the number of occurrences of that grey level.

10. In the HSV colour space, the complementary colour of green is:

    • blue

    • magenta

    • cyan
    • yellow

    The way to find this is to look at the colour wheel and draw a line across it from green -- you will find that the colour "opposite" it across the wheel is magenta.