Optimized design of top-hat filter based on algorithms of support vector machine

A novel method for self adapting morphological top-hat operator was presented. The structural elements of opening top-hat are trained by utilizing the support vector machine from a mass of sample sets. Once the machine is trained by the samples of structural elements, it can implement background suppressing in the following frames. Practical infrared image sequence background suppressing shows that the proposed method based on support vector machine was fit for lower signal noise ratio image. Experimental results of the actual measurement show that using the improved top-hat operator, the detection probability of image with sbr (sbrap0.9 dB) is higher than 96%, compared with fixed top-hat filter, the detection probability is improved by nearly 7%.