The time-decreasing learning rate and neighborhood function of the basic SOM (Self-Organizing Map) algorithm reduce its capability to adapt weights for a varied environment. In dealing with non-stationary input distributions and changing environments, we propose a modified SOM algorithm called “Time Adaptive SOM”, or TASOM, that automatically adjusts learning rate and neighborhood size of each neuron independently.The proposed TASOM is tested with stationary environments and its performance is compared with that of the basic SOM. It is also tested with non-stationary environments for representing the letter “L”, which may be translated, rotated, or scaled. Moreover, the TASOM is used for adaptive segmentation of images, which may have undergone gray-level transformation.