Chapter 1: Invited Addresses and Tutorials on Signals, Coding,
  Systems and Intelligent Techniques

Title: Combining Features from Wavelet Decomposition for Vehicle Detection

Author(s): Tomaž Romih, Žarko Cucej, Peter Planinšic, Panos Liatsis

Address: Faculty of Electrical Engineering and Computer Science, University of Maribor, Slovenia | Faculty of Electrical Engineering and Computer Science, University of Maribor, Slovenia | Faculty of Electrical Engineering and Computer Science, University of Maribor, Slovenia | Information & Biomedical Engineering Centre, School of Engineering & Mathematical Sciences City University, Northampton Square, London EC1V, U.K.

Reference: 12th International Workshop on Systems, Signals and Image Processing pp. 477 - 481

Abstract/Summary: In this paper, we present the use of the wavelet transform for image object detection. We decompose image into first level subbands and remove the approximation subband. Inverse wavelet transform gives us an image with enhanced details, which we use for segmentation of image areas to objects. We take into consideration the size scaling of object near and away from camera. A comparison of the performance of the suggested algorithm is performed.

Order a copy of this article