Machine learning techniques for sea-ice identification and classification

Abstract

Sea-ice identification and classification are essential processes for safety critical navigation support of surface vessels in polar waters. Semantic segmentation has drawn much attention as an enabling technique for fast detection of objects in a scene including sea-ice conditions. Identifying sea-ice is a challenging problem, especially in the presence of raindrops. The raindrop alters the boundaries of the objects in the scene, and thus, degrades the identification performance. In this work, deep-learning (DL) semantic segmentation networks are trained to classify the scene of sea-ice images including the VGG-16, fully convolutional network, pyramid scene parsing network, and conditional generative adversarial network (cGAN) semantic segmentation model. Two datasets are utilized to train the cGAN model. The images in the first dataset capture four classes: sea-ice, open water, sky, and vessel. The images in the second dataset capture first year sea-ice, new sea-ice, and gray sea-ice in addition to the open water, sky, and vessel classes. Data augmentation operations are applied to both datasets to change the camera mounting angle and location and enhance the datasets’ size. Transfer learning is also implemented to improve the training process. A raindrop removing framework based on morphological operations is developed to enhance the classification performance in rainy weather. Results illustrate that data augmentation operations enhance the performance of the DL models. Moreover, the raindrop removing framework improves the models’ performance. Results also show that the cGAN model outperforms the other DL models and it provides remarkable performance with rainy weather images, in which the performance of the other models degrades due to the raindrop.