Using Vector Agents to Implement an Unsupervised Image Classification Algorithm

Abstract

Unsupervised image classification methods conventionally use the spatial information of pixels to reduce the effect of speckled noise in the classified map. To extract this spatial information, they employ a predefined geometry, i.e., a fixed-size window or segmentation map. However, this coding of geometry lacks the necessary complexity to accurately reflect the spatial connectivity within objects in a scene. Additionally, there is no unique mathematical formula to determine the shape and scale applied to the geometry, being parameters that are usually estimated by expert users. In this paper, a novel geometry-led approach using Vector Agents (VAs) is proposed to address the above drawbacks in unsupervised classification algorithms. Our proposed method has two primary steps: (1) creating reliable training samples and (2) constructing the VA model. In the first step, the method applies the statistical information of a classified image by k-means to select a set of reliable training samples. Then, in the second step, the VAs are trained and constructed to classify the image. The model is tested for classification on three high spatial resolution images. The results show the enhanced capability of the VA model to reduce noise in images that have complex features, e.g., streets, buildings.