Concrete bridge crack detection is critical to guaranteeing transportation safety. The introduction of deep learning technology makes it possible to automatically and accurately detect cracks in bridges. We proposed an end-to-end crack detection model based on the convolutional neural network (CNN), taking the advantage of atrous convolution, Atrous Spatial Pyramid Pooling (ASPP) module and depthwise separable convolution. The atrous convolution obtains a larger receptive field without reducing the resolution. The ASPP module enables the network to extract multi-scale context information, while the depthwise separable convolution reduces computational complexity. The proposed model achieved a detection accuracy of 96.37% without pre-training. Experiments showed that, compared with traditional classification models, the proposed model has a better performance. Besides, the proposed model can be embedded in any convolutional network as an effective feature extraction structure.

BACKGROUND: Optical coherence tomography (OCT) is a non-invasive diagnosing tool used in clinics. Due to its high resolution (<10um), it is appropriate for the early detection of tiny infections. It has been widely used in diagnosis and treatment of diseases, evaluation of therapeutic efficacy, and monitoring of various physiological and pathological processes. OBJECTIVE: To systemically review literature to summarize the clinic application of OCT in recent years. METHODS: For clinic applications that OCT has been applied, we selected studies that describe the most relevant works. The discussion included: 1) which tissue could be used in the OCT detection, 2) which character of different tissue could be used as diagnosing criteria, 3) which diseases and pathological process have been diagnosed or monitored using OCT imaging, and 4) the recent development of clinic OCT diagnosing. RESULTS: The literature showed that the OCT had been listed as a routine test choice for ophthalmic diseases, while the first commercial product for cardiovascular OCT detection had gotten clearance. Meanwhile, as the development of commercial benchtop OCT equipment and tiny fiber probe, the commercial application of OCT in dermatology, dentistry, gastroenterology and urology also had great potential in the near future. CONCLUSIONS: The analysis and discussions showed that OCT, as an optical diagnosing method, has been used successfully in many clinical fields, and has the potential to be a standard inspection method in several clinic fields, such as dermatology, dentistry and cardiovascular.

Bridge crack detection is essential to prevent transportation accidents. However, the surrounding environment has great interference with the detection of cracks, which makes it difficult to ensure the accuracy of the detection. In order to accurately detect bridge cracks, we proposed an end-to-end model named Skip-Squeeze-and-Excitation Networks (SSENets). It is mainly composed of the Skip-Squeeze-Excitation (SSE) module and the Atrous Spatial Pyramid Pooling (ASPP) module. The SSE module uses skip-connection strategy to enhance the gradient correlation between the shallow network and deeper network, alleviating the vanishing gradient caused by the deepening of the network. The ASPP module can extract multi-scale contextual information of images, while the depthwise separable convolution reduces computational complexity. In order to avoid destroying the topology of crack, we used atrous convolution instead of the pooling layer. The proposed SSENets achieved a detection accuracy of 97.77%, which performed better than the models we compared it with. The designed SSE module which used skip-connection strategy can be embedded in other convolutional neural networks (CNNs) to improve their performance.

Aiming at the problems of feature point calibration method of 3D light detection and ranging (LiDAR) and camera calibration that are calibration boards in various forms, incomplete information extraction methods and large calibration errors, a novel calibration board with local gradient depth information and main plane square corner information (BWDC) was designed. In addition, the "three-step fitting interpolation method" was proposed to select feature points and obtain the corresponding coordinates of feature points in the LiDAR coordinate system and camera pixel coordinate system based on BWDC. Finally, calibration experiments were carried out, and the calibration results were verified by methods such as incremental verification and reprojection error comparison. The calibration results show that using BWDC and the "three-step fitting interpolation method" can solve quite accurate coordinate transformation matrix and intrinsic and external parameters of sensors, which dynamically change within 0.2% in the repeatable experiments. The difference between the experimental value and the actual value in the incremental verification experiment is about 0.5%. The average reprojection error is 1.8312 pixels, and the value changes at different distances do not exceed 0.1 pixels, which also show that the calibration method is accurate and stable.

Cracks are one of the most serious defects that threaten the safety of bridges. In order to detect different forms of cracks in different collection environments quickly and accurately, we proposed a pixel-level crack segmentation network based on convolutional neural networks, which is called the Skip Connected Crack Detection Network (SCCDNet). The network is composed of three parts: the Encoder module with 13 convolutional layers pretrained in the VGG-16 network, the Decoder module with a densely connected structure, and the Skip-Squeeze-and-Excitation (SSE) module which connects the feature map shaving the same resolution in the Encoder and Decoder. We used depthwise separable convolution to improve the accuracy of crack segmentation while reducing the complexity of the model. In this paper, a dataset containing cracks collected in different scenes was established, and SCCDNet was trained and tested on this dataset. Compared with the advanced models, SCCDNet obtained the best crack segmentation performance, while F-score reached 0.7763.

The sparse CT (Computed Tomography), inspired by compressed sensing, means to introduce a prior information of image sparsity into CT reconstruction to reduce the input projections so as to reduce the potential threat of incremental X-ray dose to patients’ health. Recently, many remarkable works were concentrated on the sparse CT reconstruction from sparse (limited-angle or few-view style) projections. In this paper we would like to incorporate more prior information into the sparse CT reconstruction for improvement of performance. It is known decades ago that the given projection directions can provide information about the directions of edges in the restored CT image. ATV (Anisotropic Total Variation), a TV (Total Variation) norm based regularization, could use the prior information of image sparsity and edge direction simultaneously. But ATV can only represent the edge information in few directions and lose much prior information of image edges in other directions. To sufficiently use the prior information of edge directions, a novel MDATV (Multi-Direction Anisotropic Total Variation) is proposed. In this paper we introduce the 2D-IGS (Two Dimensional Image Gradient Space), and combined the coordinate rotation transform with 2D-IGS to represent edge information in multiple directions. Then by incorporating this multi-direction representation into ATV norm we get the MDATV regularization. To solve the optimization problem based on the MDATV regularization, a novel ART (algebraic reconstruction technique) + MDATV scheme is outlined. And NESTA (NESTerov’s Algorithm) is proposed to replace GD (Gradient Descent) for minimizing the TV-based regularization. The numerical and real data experiments demonstrate that MDATV based iterative reconstruction improved the quality of restored image. NESTA is more suitable than GD for minimization of TV-based regularization. MDATV regularization can sufficiently use the prior information of image sparsity and edge information simultaneously. By incorporating more prior information, MDATV based approach could reconstruct the image more exactly.