Evaluating transfer learning approaches for image information mining applications

The recent explosion of data from various Earth observation (EO) systems requires new ways to rapidly harness the information and synthesize it for decision making. Currently several image information mining (IIM) systems have some form of supervised statistical learning models that relate the image content to the various semantic classes. However, this kind of approach is constrained by the paucity of training information in several EO domains due to limited ground truth. Although, semi-supervised learning methods alleviate this problem to a certain extent by using unlabelled data from various spatial databases, however these methods require that the training data and future unseen data should conform to the same statistical distribution and feature space. To overcome this problem a more recent approach is focused on using small amounts of labeled information from closely related or similar learning task and somehow adapt that information in developing new semantic models. The above methodology called transfer learning can be applied in several processes of supervised and unsupervised learning. In this paper, we propose Transfer learning methods for IIM and discuss various techniques and their implications for content-based retrieval in the EO domain. Specifically, we explore transfer learning application in a rapid disaster response scenarios during coastal events. The adopted methodology for knowledge transfer is based on harnessing prior knowledge from similar concepts to learn new ones and uses a modified weighted least squares support vector machine (SVM).