Semi-supervised Segmentation Fusion of Multi-spectral and Aerial Images
This work addresses segmentation fusion for remote sensing and aerial imagery, but it appears incremental as it builds on existing unsupervised fusion with added semi-supervision.
The paper tackles the problem of achieving consensus among different segmentation outputs from multi-spectral and aerial images by proposing a semi-supervised segmentation fusion algorithm, which shows improved performance over individual state-of-the-art segmentation methods in experiments on artificial and real-world benchmarks.
A Semi-supervised Segmentation Fusion algorithm is proposed using consensus and distributed learning. The aim of Unsupervised Segmentation Fusion (USF) is to achieve a consensus among different segmentation outputs obtained from different segmentation algorithms by computing an approximate solution to the NP problem with less computational complexity. Semi-supervision is incorporated in USF using a new algorithm called Semi-supervised Segmentation Fusion (SSSF). In SSSF, side information about the co-occurrence of pixels in the same or different segments is formulated as the constraints of a convex optimization problem. The results of the experiments employed on artificial and real-world benchmark multi-spectral and aerial images show that the proposed algorithms perform better than the individual state-of-the art segmentation algorithms.