If I understand what you are trying to achieve, I don’t think what you are asking is possible.
Typically, the integration of radio frequency identification and video systems involves linking a specific video segment with the serial number of a particular tag or tags within a read field. So a camera would be trained on, say, the back room of an apparel store. As items were received into inventory and stored, the RFID system would record the serial numbers, along with the time that they were captured. This would be linked with the time that video was shot, so you could search the video file for serial number 12345, for instance, and see what was going on at the moment that the tag was in the read field.
You could also have a camera move to cover different read fields. So let’s say items were unpacked on two tables. The camera could swing to table 1 or table 2 when tags were being read, as long as both tables were not in use simultaneously.
Having a camera “follow” a tagged item is a bit more tricky, however. With passive ultrahigh-frequency (UHF) readers, the read field from a single circular-polarized antenna might be 15 feet wide by 20 feet long, and might be shaped like an American football. You could set up antennas to cover the entire 100-foot-by-200-foot area, but the RFID data would show only that a tag was within read field 1 or read field 2. It would not allow you to determine that the reader was on the left side of read field 1 and moving to the right side. So it depends, I guess, on how closely you want to follow something. If the camera covered the entire read field and then swiveled to the adjacent field, then it might work.
Another option would be a phased-array antenna system, which could more precisely locate tags within a three-dimensional space. But even these antennas are only accurate to approximately three or four feet. If I have misunderstood your question, please let me know.
—Mark Roberti, Founder and Editor, RFID Journal
Can You Recommend an Inexpensive UHF Vehicle Scanner? »