Fabric defect detection is a necessary and essential step of quality control in the textile manufacturing industry. Traditional fabric inspections are usually performed by manual visual methods, which are low in efficiency and poor in precision for long-term industrial applications. In this paper, we propose an unsupervised learning-based automated approach to detect and localize fabric defects without any manual intervention. This approach is used to reconstruct image patches with a convolutional denoising autoencoder network at multiple Gaussian pyramid levels and to synthesize detection results from the corresponding resolution channels. The reconstruction residual of each image patch is used as the indicator for direct pixel-wise prediction. By segmenting and synthesizing the reconstruction residual map at each resolution level, the final inspection result can be generated. This newly developed method has several prominent advantages for fabric defect detection. First, it can be trained with only a small amount of defect-free samples. This is especially important for situations in which collecting large amounts of defective samples is difficult and impracticable. Second, owing to the multi-modal integration strategy, it is relatively more robust and accurate compared to general inspection methods (the results at each resolution level can be viewed as a modality). Third, according to our results, it can address multiple types of textile fabrics, from simple to more complex. Experimental results demonstrate that the proposed model is robust and yields good overall performance with high precision and acceptable recall rates.

Increasing deployment of optical fiber networks and the need for reliable high bandwidth make the task of inspecting optical fiber connector end faces a crucial process that must not be neglected. Traditional end face inspections are usually performed by manual visual methods, which are low in efficiency and poor in precision for long-term industrial applications. More seriously, the inspection results cannot be quantified for subsequent analysis. Aiming at the characteristics of typical defects in the inspection process for optical fiber end faces, we propose a novel method, “difference of min-max ranking filtering” (DO2MR), for detection of region-based defects, e.g., dirt, oil, contamination, pits, and chips, and a special model, a “linear enhancement inspector” (LEI), for the detection of scratches. The DO2MR is a morphology method that intends to determine whether a pixel belongs to a defective region by comparing the difference of gray values of pixels in the neighborhood around the pixel. The LEI is also a morphology method that is designed to search for scratches at different orientations with a special linear detector. These two approaches can be easily integrated into optical inspection equipment for automatic quality verification. As far as we know, this is the first time that complete defect detection methods for optical fiber end faces are available in the literature. Experimental results demonstrate that the proposed DO2MR and LEI models yield good comprehensive performance with high precision and accepted recall rates, and the image-level detection accuracies reach 96.0 and 89.3%, respectively.

The drilling length is an important parameter in the process of horizontal directional drilling (HDD) exploration and recovery, but there has been a lack of accurate, automatically obtained statistics regarding this parameter. Herein, a technique for real-time HDD length detection and a management system based on the electromagnetic detection method with a microprocessor and two magnetoresistive sensors employing the software LabVIEW are proposed. The basic principle is to detect the change in the magnetic-field strength near a current coil while the drill stem and drill-stem joint successively pass through the current coil forward or backward. The detection system consists of a hardware subsystem and a software subsystem. The hardware subsystem employs a single-chip microprocessor as the main controller. A current coil is installed in front of the clamping unit, and two magneto resistive sensors are installed on the sides of the coil symmetrically and perpendicular to the direction of movement of the drill pipe. Their responses are used to judge whether the drill-stem joint is passing through the clamping unit; then, the order of their responses is used to judge the movement direction. The software subsystem is composed of a visual software running on the host computer and a software running in the slave microprocessor. The host-computer software processes, displays, and saves the drilling-length data, whereas the slave microprocessor software operates the hardware system. A combined test demonstrated the feasibility of the entire drilling-length detection system.

For simulating CoalBed Methane (CBM) hydraulic fracturing using 3-D meshless method, this paper analyzed the hydraulic fracturing mechanism and cracking form for coal rock and established the geometric and mathematical models of hydraulic fracturing propagation in coal rock in terms of the Hillerborg model on crack opening displacement theory. With the theoretical basis of hydromechanics, the formulas for calculating hydraulic pressure inside the fracture by numerical simulation were deduced from the analysis on this fluid-structure interaction problem. The geometric and mathematical models established above were described by 3-D meshless Galerkin (EFG, Element-Free Galerkin) method and compiled into the numerical simulation program using VB and FORTRAN programming language to simulate the fracture propagation for an actual coal rock sample with a drilling hole as an example. Then the physical simulation experiment of hydraulic fracturing propagation of coal seam was conducted on the same coal rock sample. Through the direct observation with naked eyes and detection by advanced instruments of ESEM and Micro-CT, the shape and parameters of cracks on the surface of and inside the coal rock sample were achieved, which indicated that experimental results are reasonably consistent with numerical simulation results.

Horizontal auger boring (HAB) is a widely used trenchless technology for the high-accuracy installation of gravity or pressure pipelines on line and grade. Differing from other pipeline installations, HAB requires a more precise and automated guidance system for use in a practical project. This paper proposes an economic and enhanced automated optical guidance system, based on optimization research of light-emitting diode (LED) light target and five automated image processing bore-path deviation algorithms. An LED light target was optimized for many qualities, including light color, filter plate color, luminous intensity, and LED layout. The image preprocessing algorithm, direction location algorithm, angle measurement algorithm, deflection detection algorithm, and auto-focus algorithm, compiled in MATLAB, are used to automate image processing for deflection computing and judging. After multiple indoor experiments, this guidance system is applied in a project of hot water pipeline installation, with accuracy controlled within 2 mm in 48-m distance, providing accurate line and grade controls and verifying the feasibility and reliability of the guidance system.

Drill rod straightness has to be strictly controlled and the maximum bending degree detection needs to be used in the straightening process. The mechanical bending degree measurement depends on machinery instruments and workers’ experience, often with low efficiency and precision. While the optical inspection, as a non-contact detection method, with higher precision and lower installation accuracy requirements, is frequently applied in the online detection system. Based on this, an enhanced bending degree detection system for a drill rod is proposed in this paper. Compared to the existing detection system, the main progress is to use a two-dimensional laser to quickly obtain arc profile data and fit with ellipse. Segment inspection idea is also utilized is this system as the camera that could obtain the whole drill rod in one shot needs extremely high resolution and price. A specialized algorithm is designed to fit the cross-section shape and whole centerline displacement based on the least square method. Some laboratory tests are conducted to verify this detection system, findings of which are compared to manual measured results. The maximum bending degree error is 2.14 mm and the maximum position error is 8.21 mm, which are both within the tolerance of error. Those results show the feasibility and precision of this enhanced detection system.