Despeckling of sonar images based on a naive homogeneity index
Rithu James, M. H. Supriya · 2016
Despeckling is a preprocessing step required for application like segmentation or registration of target from sonar images. The detection and classification of objects in the sonar images is adversely affected by the presence of multiplicative speckle noise . A blind denoising of sonar images based on sonar image heterogeneity of sonar image patches is proposed. The homomorphic approach is used to convert the multiplicative signal model into an additive one. The noise level of the single noisy sonar image is estimated using a patch based algorithm and then non blind denoising is done. The sonar image heterogeneous patch classification is based on a new proposed naive homogeneity index which depends on the texture strength of the patches and despeckling filters can be adjusted to these patches. Experimental results substantiate the effectiveness of the despeckling method. The different non reference image performance evaluation criterion are used to evaluate the proposed method.