Abstract: To further enhance the contrast of infrared image, improve the definition and suppress the noise, an adaptive enhancement method in Contourlet domain based on chaotic Particle Swarm Optimization (PSO) is proposed. First, Contourlet transform of the infrared image is performed. The proportion of low-pass image and detail image in the original image is adjusted, and the contrast is enhanced by linear gray stretch. Then, the coefficients of noisy bandpass directional subbands are adjusted by nonlinear gain function. An integrated quantitative evaluation function is used as the fitness of the chaotic PSO. In this evaluation function three indexes are taken into account, i.e. contrast, definition and signal-to-noise ratio. The optimal parameters, involved in the enhancement method in spatial domain and the nonlinear gain function for adjustment of coefficients of bandpass directional subbands base on Contourlet, are obtained by chaotic PSO algorithm. Experimental results for a large number of infrared images show that, compared with four existing image enhancement methods, the proposed method improves the contrast of enhanced infrared image, increases the definition, reduces the noise, and has a better overall visual effect.

Key words: information processing technology, infrared image enhancement, Contourlet transform, adaptive enhancement, niche chaotic mutation particle swarm optimization, nonlinear gain function