Ultrafast pattern based line detector for real time image processing

Line detector algorithms are used for feature extraction, segmentation, pattern recognition, matching and object detection in computer vision applications such as detection of road lanes and power lines. The fastest line detector so far published in the literature is presented in this thesis. This line detector algorithm is 14.88 times faster than the next fastest line detector, the EDLines, and 94.7 times faster than the line segment detector and more reliable than the Standard Hough Transform. An ingeniously simple method for detecting and combining patterns with advantage of several look-up tables is introduced to recognize and fit straight-line patterns. As the first step, it recognizes any possible 4 × 4 pixel line patterns among the binary edge pixels and then uses the slope and the connection points look-up tables to decide whether the connected patterns form a line or not. This algorithm is implemented and tested in MATLAB to compare the speed performance. MATLAB executable file is created by compiling C++ code. The experimental results are presented as graphics and tables in terms of the speed. According to the results, the proposed algorithm is an ultrafast line detector for real time image processing. This algorithm is especially designed for high definition images. Key Words: Line Detector, Feature Extraction, Segmentation, Pattern Recognition, Image Processing, Matlab