So will you be illuminating the tape sequentially with various color LEDs and measuring transmission each time? Or will the setup be an RGB photodiode with 3 elements and a broad spectrum light source? Have you measured the light intensity range, put that next to the photodiode datasheet and figured out the typical current range you can expect from the sensor?
In short: have you done the theoretical work to get a ballpark estimate of how much gain you need from the analog frontend?