Abstract
Most of the existing synthetic aperture radar (SAR) image superpixel generation methods are designed based on the raw SAR images or artificially designed features. However, such methods have the following limitations: (1) SAR images are severely affected by speckle noise, resulting in unstable pixel distance estimation. (2) Artificially designed features cannot be well-adapted to complex SAR image scenes, such as the building regions. Aiming to overcome these shortcomings, we propose a multitask learning-based superpixel generation network (ML-SGN) for SAR images. ML-SGN firstly utilizes a multitask feature extractor to extract deep features, and constructs a high-dimensional feature space containing intensity information, deep semantic informantion, and spatial information. Then, we define an effective pixel distance measure based on the high-dimensional feature space. In addition, we design a differentiable soft assignment operation instead of the non-differentiable nearest neighbor operation, so that the differentiable Simple Linear Iterative Clustering (SLIC) and multitask feature extractor can be combined into an end-to-end superpixel generation network. Comprehensive evaluations are performed on two real SAR images with different bands, which demonstrate that our proposed method outperforms other state-of-the-art methods.
Highlights
Synthetic aperture radar (SAR) has been widely used in many fields due to the capability of providing unique and useful information in all-weather and multi-climate conditions, such as sea monitoring [1], agricultural development [2], urban planning [3]
We evaluate the performance of the proposed method for SAR image superpixel generation
We propose an end-to-end SAR image superpixel generation method, named the multitask learning-based superpixel generation network (ML-SGN)
Summary
Synthetic aperture radar (SAR) has been widely used in many fields due to the capability of providing unique and useful information in all-weather and multi-climate conditions, such as sea monitoring [1], agricultural development [2], urban planning [3]. Xiang et al [17] defined the distance to measure both the feature similarity and spatial proximity, and combined local K-means clustering and Ncut to generate SAR image superpixels. It is worth noting that the above superpixel generation methods are designed based on the raw SAR data or artificially designed features to measure the similarity between pixels. We propose a novel SAR image superpixel generation network based on multitask learning (ML-SGN). We construct a high-dimensional feature space including deep semantic information, spatial information, and intensity information, and define an effective distance measure between pixels based on the high-dimensional feature space. We construct a high-dimensional feature space containing deep semantic information, intensity information and spatial information, and define an effective pixel distance measure based on this high-dimensional feature space.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have