Boyu Zhang

This work focuses on solving two challenging problems in pavement crack detection: (1) noises caused by complicated pavement textures and intensity inhomogeneity cannot be removed effectively, which makes crack extraction difficult; and (2) sealed cracks and cracks with similar intensity and width cannot be separated correctly, which makes data analysis and budgeting inaccurate. Here, a unified crack and sealed crack detection approach is proposed that can detect and separate both cracks and sealed cracks under the same framework. It trains a deep convolutional neural network to preclassify a pavement image into crack, sealed crack, and background regions. A blockwise thresholding method is developed to segment the crack/sealed crack pixels efficiently and effectively. Finally, tensor voting–based curve detection is applied to extract the crack/sealed crack. The proposed approach is validated using 800 images (each 2,000×4,000 pixels); the experimental results demonstrate that this approach accurately distinguishes cracks from sealed cracks and achieves very good detection performance (recall=0.951; precision=0.847).

Abstract The emergence of cesium lead iodide (CsPbI 3 ) perovskite solar cells (PSCs) has generated enormous interest in the photovoltaic research community. However, in general they exhibit low power conversion efficiencies (PCEs) because of the existence of defects. A new all‐inorganic perovskite material, CsPbI 3 :Br:InI 3 , is prepared by defect engineering of CsPbI 3 . This new perovskite retains the same bandgap as CsPbI 3 , while the intrinsic defect concentration is largely suppressed. Moreover, it can be prepared in an extremely high humidity atmosphere and thus a glovebox is not required. By completely eliminating the labile and expensive components in traditional PSCs, the all‐inorganic PSCs based on CsPbI 3 :Br:InI 3 and carbon electrode exhibit PCE and open‐circuit voltage as high as 12.04% and 1.20 V, respectively. More importantly, they demonstrate excellent stability in air for more than two months, while those based on CsPbI 3 can survive only a few days in air. The progress reported represents a major leap for all‐inorganic PSCs and paves the way for their further exploration in order to achieve higher performance.

This article describes a worm-like soft robot capable of operating in complicated tubular environments, such as the complex pipeline with different diameters, water, oil, and gas environments, or the clinical application in natural orifice transluminal endoscopic surgery. The robot is completely soft and robust, and consists of one multidegree of freedom (DoF) extension module and two clampers for locomotion and steering. The multi-DoF extension module is able to adjust the heading direction in the three-dimensional space. The clamper has a basic expansion module structure and detachable sucking module structure. The combined clamping principle for sticking to the inner wall can be reconfigurable to adapt the tubes with multiple tubular scales and super elastic materials. For fabrication of the mechanical structure, a low-cost and time-efficient method is proposed in this article. Based on our proposed robot, a series of phantom and application experiments are performed. The results demonstrate that the soft robot can freely bend and elongate with the entire soft body, and pass through tubes with changing diameters or branches, dry tubes, liquid environments, hard surfaces, and even soft deformable tubes. It has the ability to remove a load of >10 times its own weight. In addition, an additional visualization unit, biopsy, and electromagnetic sensor are mounted on the robot tip for the real-time image inspection, manipulation, and robot tracking. The proposed worm-like soft robot is compact, flexible-actuated, and sufficiently safe, as well as extensible. Its ability to move in the complex unstructured environment shows a great potential for search and medical applications.