The main problem with Bingo balls is that they are polished and very glossy, so it is hard to avoid light reflection. That is why the special process of double expousure is used here.

16 white LEDs are arranged in 2x2 groups. There are inner leds (on the green back plane on the photo), which illuminates the center of the ball, and the outer ones (the black front plane) for the ball edges. Each group can be addressed for "odd" and "even" LEDs, so MCU can illuminate individual ball slices. Two exposures are made, first with the goup of LEDs which are marked as red on the drawing, an the second with LEDs marked as blue (of course, all of them are actually white LEDs). During the second exposure, each pixel which is lighter than in the previous exposure, will be ignored, and only the darker ones shall be written in frame buffer. If there is no extra light which comes from the outside, the result is perfect exposure, with no shiny areas.

There are two disadvantages of this process. The ball must be motionless during both exposures, and it takes a lot of time. Most of low-cost surveillance cameras have some persistence, so you must wait for one more dummy frame between LEDs setting and exposure. For PAL cameras, one frame takes 20 ms and in the worst case the whole described process takes 5x20 ms, plus about 15 ms for frame conditioning (digital adjusting of black and white levels and contrast enhancement). For NTSC cameras, this time is about 20% shorter.

The next three photos show the first, second and combined exposure. They are taken from the monitor (connected to the VGA connector from the schematics) by photo camera, not photoshopped. Same is with all other photos on next pages.


First exposure Second exposure Combined image
    Back to projects 
0. Home
H. Hardware
L. Lighting
1. Frame fetch
2. Ball locate
3. Stretching
4. Unsharp mask
5. Component selection
6. Holes and scratches
7. Component shrink
8. Invert selection
9. Component list
10. Select best circle
11. Angle detection
12. Rotate selected
13. Select broad lines
14. Selected components list
15. Components scaling
16. Histogram
17. MSE and voting
C. Conclusion