Detection based on Hough Transform is a good way frequently employed to detect straight-line targets, but it bears problems as difficulty in selecting an appropriate threshold coefficient and target misjudgment. To solve those technical problems mentioned above, this paper proposes a novel algorithm based on Hough Transform and Mean Shift Multi-Scale Clustering (MSMSC-HT). Firstly, the outline of targets is extracted and a primary selection is conducted by taking a low threshold. Then, the primary targets are treated with Multi-Scale Clustering, and the class centers can be obtained via Mean Shift algorithm. Finally, the target number and attitude parameters can be calculated adaptively by optimization of scales. For multi-target shaped in straight lines in sky background, this algorithm proposed can avoid the difficulty in the choice of an appropriate threshold by taking clustering method, thus can successfully complete detecting mission. Experimental results verify the efficacy of the proposed algorithm.
[1]
Yee Leung,et al.
Clustering by Scale-Space Filtering
,
2000,
IEEE Trans. Pattern Anal. Mach. Intell..
[2]
Rolf H. Möhring,et al.
Resource-constrained project scheduling: Notation, classification, models, and methods
,
1999,
Eur. J. Oper. Res..
[3]
Francisco Sandoval Hernández,et al.
Mean shift based clustering of Hough domain for fast line segment detection
,
2006,
Pattern Recognit. Lett..
[4]
Dorin Comaniciu,et al.
Mean Shift: A Robust Approach Toward Feature Space Analysis
,
2002,
IEEE Trans. Pattern Anal. Mach. Intell..
[5]
Babak Nadjar Araabi,et al.
Hough Array Processing via Fast Multi-Scale Clustering
,
2000,
Real Time Imaging.