Superpixel Segmentation for Polarimetric SAR Imagery Using Local Iterative Clustering

Abstract

The simple linear iterative clustering (SLIC) algorithm shows good performance in superpixel generation for optical imagery. However, SLIC can perform poorly when there is too much noise in the image. To solve this problem, we have improved the cluster center initialization step and the postprocessing step, and then introduce the SLIC superpixel segmentation algorithm to the polarimetric synthetic aperture radar (PolSAR) image processing field. Experiments using AirSAR and ESAR L-band PolSAR data show that the improved SLIC algorithm can overcome the effect of speckle noise in PolSAR imagery, and it shows a better performance in detail preservation than the original SLIC algorithm and the normalized cuts superpixel segmentation algorithm.Index Terms-Polarimetric synthetic aperture radar (SAR) (PolSAR), SAR, segmentation, simple linear iterative clustering (SLIC), superpixel.

Superpixel Segmentation for Polsar Images With Local Iterative Clustering and Heterogeneous Statistical Model

“…Similarly, in (Song et al 2015;Fachao, Jiming, and Fengkai 2015;Feng, Cao, and Pi 2014), this parameter is also chosen to be a constant to balance the polarimetric and spatial similarity. This parameter is usually set manually by trial and error, which might cause over-or under-superpixel segmentation in some spatially complicated areas.…”

“…2 (b). In our experiment, two superpixel segmentation methods for PolSAR data, i.e., Qin's method in (Fachao, Jiming, and Fengkai 2015) and Liu's method in (Liu et al 2013), are utilized for comparison. The former is a modified version of SLIC and the latter is designed based on Normalized cuts.…”

“…For PolSAR data, Liu et al (Liu et al 2013) incorporated the revised Wishart distance and edge map into the Normalized cuts algorithm to produce superpixels. On the basis of SLIC, Feng et al (Feng, Cao, and Pi 2014), Song et al (Song et al 2015), and Qin et al (Fachao, Jiming, and Fengkai 2015) utilized the symmetric revised Wishart distance, Bartlett distance and revised Wishart distance respectively as the similarity measures instead of the original one to generate superpixels. It can be seen that these methods are all designed based on the assumption of Wishart distribution, which can well describe the backscatters of natural areas.…”

Identification of changes in urban land cover type using fully polarimetric SAR data

“…The SLIC method has been successfully applied for fully PolSAR images segmentation purposes (Feng, Cao, and Pi 2014;Qin, Guo, and Lang 2015). The basic idea of SLIC is iteratively assigning pixels to the nearest superpixels (Feng, Cao, and Pi 2014).…”

“…The basic idea of SLIC is iteratively assigning pixels to the nearest superpixels (Feng, Cao, and Pi 2014). Firstly, superpixels are initialized by placing a set of seeds on the image domain according to the superpixel's expected size (S × S), and to avoid seeding a centre on an edge or on a noisy pixel, the centres are moved to locations corresponding to the lowest gradient position in a 3 × 3 neighbourhood (Qin, Guo, and Lang 2015). Then, the algorithm is implemented with two alternating steps: (1) fix superpixel centres and assign each pixel to the nearest superpixel according to a distance measure D; and (2) update superpixel centres and repeat step (1).…”