To cope with in the world, it is difficult to always manually inspect all of these problems, this paper proposes a unified framework them. Instead of the manual inspection, automatic inspection consisting of the following novel contributions (Table 1): is one of the prospective solutions for efficiently diagnosing 1. Crack Segmentation with Blind Super-Resolution these constructions. While such inspection can be achieved (CSBSR): As with Crack Segmentation with Super by several types of sensors such as the Falling Weight Deflectometer, Resolution (CSSR) proposed in our earlier conference the Pavement Density Profiler, and the Ground paper [60], CSBSR proposed in this paper connects "a Penetrating Radar, this paper focuses on crack segmentation network for Super Resolution (SR) accepting an input on images captured by generic cameras for visual inspection. LR image" in series to "a segmentation network" for Crack segmentation [31] is defined to be binary semantic end-to-end joint learning. We extend CSSR to CSBSR segmentation in the field of computer vision. While the with blind SR to handle realistically-blurred images.

Most deep learning-based super-resolution (SR) methods are not image-specific: 1) They are exhaustively trained on datasets synthesized by predefined blur kernels (\eg bicubic), regardless of the domain gap with test images. 2) Their model weights are fixed during testing, which means that test images with various degradations are super-resolved by the same set of weights. However, degradations of real images are various and unknown (\ie blind SR). It is hard for a single model to perform well in all cases. To address these issues, we propose an online super-resolution (ONSR) method. It does not rely on predefined blur kernels and allows the model weights to be updated according to the degradation of the test image. Specifically, ONSR consists of two branches, namely internal branch (IB) and external branch (EB). IB could learn the specific degradation of the given test LR image, and EB could learn to super resolve images degraded by the learned degradation. In this way, ONSR could customize a specific model for each test image, and thus could be more tolerant with various degradations in real applications. Extensive experiments on both synthesized and real-world images show that ONSR can generate more visually favorable SR results and achieve state-of-the-art performance in blind SR.