Shadow Elimination to Detect Moving Vehicles on Highway

Kalpesh Dongre, Nilesh Kasat

Abstract


This paper presents a method to detect moving vehicles accurately without their shadow when shadow is also in motion along with the vehicles. The concept is based on segmentation of moving object from its background. This segmented object contains shadow also along with moving object. Once the moving object is separated accurately from its background next part is to detect object without considering shadow. Here we are deriving spectral property of shadow by considering photometric color invariants normalized RGB. This photometric color invariants of background and object is utilized to separate out shadow from object. Threshold is used here which is applied to the difference of photometric color invariants of background and object for detecting object accurately. The experimental result indicates that the algorithm of detecting shadow is fast and exact, improves the accuracy of the moving vehicle detection.

 

 

Keywords: Shadow; elimination; segmentation; moving object


Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v2i4.677

Refbacks

  • There are currently no refbacks.




Copyright (c) 2016 International Journal of Advanced Research in Computer Science