Context-Guided Discrimination Feature Learning in Color Space for Aircraft Detection in SAR Images

Aircraft often manifest as highly aspect/pose-sensitive, disjointed blobs made of pixels with fluctuating levels of brightness in classic grayscale SAR images, which makes the aircraft annotation task challenging even for human experts. As more and more high-resolution colorized SAR images acquired by the latest commercial imaging modes are released for open access, both the academia and the industry started to notice the benefits of color-coded SAR images. However, since the large-scale datasets in the area of SAR aircraft detection feature grayscale SAR images, research on discrimination feature learning in color space for SAR aircraft detection is very limited. To embrace the opportunity brought by the new generation of highly informative colored SAR images, we propose the Phase-Aware Clustering Enhanced Detector (PACE-Det), which consists of three main components: the Phase-Orientation Color Encoder (POCE) module, the core detection network, and the Multi-Space Clustering Constraint (MSCC) module. The front-end POCE module generates pseudo-color SAR images based on Phase Congruency (PC). The color representations are fed into the core detection network for feature extraction, where the phase-aware alignment loss is introduced in addition to the classification and regression losses in the standard object detection task. The initial predictions generated by the core detection network are further refined by the contextual information extracted by the post-processing MSCC module based on the image segmentation result in L∗a∗b∗ color space, where anisotropic objects like aircraft and isotropic scatterers like impervious surfaces exhibit distinct spatial distributions and color features. Experiments based on the Composite SAR Aircraft Dataset (CSAD), which is constructed by fusing airport scenes with SAR aircraft target patches, show that the proposed PACE-Det achieves a mAP75 of 0.948 and mAP90 of 0.561, which are higher than many state-of-the-art networks.