Abstract—Sea target detection is an important goal for military purpose and navigation. A new nonlinear quadratic transform is presented to generate target-based color space (TCS) for sea target detection. RGB color space is converted to TCS through a 3*3 matrix, using quadratic transformation. PSO algorithm is used for maximum differentiation between target and non-target clusters which are classified with FCM clustering method. The task of PSO is to minimize the Error value produced by FCM classifier. Experimental results show the ability of the new method in finding sea targets in color images. The superiority of the TCS space to the relative previous works and other color spaces is conspicuous.

Index Terms—Quadratic transformation, Color space, Sea target detection, Particle Swarm Optimization, Fuzzy C-Mean.

Saeed Mirghasemi is with the Computer & IT Engineering Department, Islamic Azad Univercity of Parand, IRAN (e-mail: s.mirghasemi@gmail.com)
Hadi Sadoghi Yazdi is with the Computer Engineering Department, Ferdowsi University of Mashhad, IRAN (e-mail: h-sadoghi@um.ac.ir)
Mojtaba Lotfizad is with the Electical & Computer Engineering Department, Tarbiat Modares University of Tehran, IRAN (e-mail: lotfizad@modares.ac.ir)

Cite: Saeed mirghasemi, Hadi Sadoghi Yazdi and Mojtaba Lotfizad, "Linear and Quadratic PSO Based Color Space Conversion for Sea Target Detection," International Journal of Computer and Electrical Engineering vol. 2, no. 1, pp. 111-118, 2010.