Local adaptive receptive field self-organizing map for image color segmentation

摘要

A new self-organizing map with variable topology is introduced for image segmentation. The proposed network, called a Local Adaptive Receptive Field Self-organizing Map (LARFSOM), is a fast convergent network capable of color segmenting images satisfactorily, which has optimum self-adaptive topology and achieves good PSNR values. LARFSOM is compared to SOM, FS-SOM and GNG, self-organizing maps used for color segmentation. LARFSOM reached a higher color palette variance and a better 3D RGB color space distribution of learned data from the training images than the other models. LARFSOM was tested to segment images with different degrees of complexity and has given promising results.