When a cyclist is cycling on a suburban road, it’s a problem to notice fast rear approaching vehicles in some cases. Looking back frequently is not a good idea. Finding some technical way to help cyclist perceiving rear approaching vehicles is quite necessary. This project aims to find some proper sensor to detect rear approaching vehicles. It’s separated into three steps. First, choose the suitable sensor and capture data. Then, find proper analyzing tool to analyze the capture data. Last but not least, draw a conclusion after analyzing contrast. Microphone is chosen as the sensor to recording the sounds form rear approaching vehicles. ”iRig Recorder FREE” is the program to transfer audio format. And the analyzing tool is to be Matlab. Matlab audio analysis makes good frequency spectrum for each piece of audio data. According to the frequency spectrum, the unique amplitude change around 1000 Hz is found when there is a rear approaching vehicle. This change is always distinct with or without noise. After getting the spectrum of different audio sources, the cross-correlation coefficient between 800 Hz and 1200 Hz is computed to see the correlation level. Then according to cross-correlation coefficient between new captured data and knowledge data, we can determine if there is a rear approaching vehicle in the new data or not. So, this project proves that the cross-correlation coefficient of frequency spectrum can determine if there is rear approaching vehicles or not. The future work would be automatic computer detect depending on this method.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:hh-23565 |
| Date | January 2013 |
| Creators | Chen, Chengshang |
| Publisher | Högskolan i Halmstad, Sektionen för Informationsvetenskap, Data– och Elektroteknik (IDE) |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
Page generated in 0.001 seconds