The trick now is to implement an appropriate switch to count the items or events in question. All it needs to do is make or break contact, or produce a high-low-high voltage transition. Let’s assume we’re counting items passing by on a conveyor belt. The implementation will depend on the items themselves. Are they all the same size and orientation? How heavy are they? Are they randomly placed and timed? In some cases a simple mechanical spring-mounted “flip” switch can be used: The item brushes up against the switch and pushes it to the side, activating it. The spring then restores the switch to the off position to await the next item. If the items are small, irregularly positioned or shaped, this technique may have difficulty. Another approach is to use a photo-electric scheme. Here, a beam of visible or infrared light is shown across the conveyor belt. Opposite it is a detection device. When an item passes by, the beam is broken and a signal generated. This scheme will work with oddly shaped or placed items as well as very light items as no physical contact is made. Not everything will work with this, though. Possible examples of difficult items include aquariums and thinly sliced chunks of Swiss cheese (really, any oddly shaped device with gaps might cause multiple triggers).
Mechanical switching will require some fabrication but the idea is straightforward enough. For photo-electric, there are a few options. One possibility is the use of a light dependent resistor or CdS cell and a strong light source. The CdS cell will exhibit high resistance under low light and the resistance will decrease as light levels increase. The CdS cell could be connected directly to an input port pin with the internal pull-up resistor enabled. A major challenge here is the ambient light. Another possibility is an infrared emitter-detector pair (IR LED plus IR phototransistor). This would require a little more circuitry but would tend to be less sensitive to ambient light conditions.
Alignment of photo-electric devices and their separation distance can be critical to proper and consistent triggering. For ease of testing in lab, it is best to focus on “proof of concept”, keeping the devices only an inch or two apart and simulating the items to be counted by passing a black card between the sensor/detector pair.