Inspection Requirements Research Articles

Improved YOLOv5l-based Detection of Surface Defects in Hot Rolled Steel Strips

To address the problems of complex background, different sizes and easy to miss and mis-detect in the detection of surface defects in hot-rolled strip, an improved YOLOv5l-based method for detecting surface defects in hot-rolled strip is proposed. Firstly, by adding the SimAM attention mechanism module to the aggregation network, the important information is focused with high weights to improve the recall rate of the original algorithm; secondly, by replacing all C3 modules in the YOLOv5l structure with C2F, a richer gradient of information flow is obtained to improve the accuracy rate of the original algorithm. The experimental results show that the average detection accuracy using the improved YOLOv5l improves by 5.3% and the accuracy rate by 8.3% compared to the original network, resulting in higher detection accuracy and lower error and miss detection rates, meeting the requirements of hot-rolled strip steel inspection in industrial manufacturing.

Research on inspection of machine room by mobile robot

The machine room inspection robot, as a special robot, can locate and navigate in the machine room independently, and carry sensors to detect equipment faults and record instrument data instead of the machine room staff. In this paper, the relevant standards and existing inspection robots at home and abroad were investigated and the overall scheme was determined based on the actual machine room environment. The monitoring background employs B/S architecture, allowing staff to deliver inspection tasks and manage robots from any computer. The motion map is created using the GMapping algorithm, and the topology map for inspection is created based on the inspection task and feasible path. For raster map localization, the adaptive Monte Carlo localization algorithm is used. The functional layer software for the robot is designed based on the inspection function to be performed by the robot and the man-machine interaction with the monitoring background. The findings of this paper's research indicate that the inspection robot system can meet the requirements of computer room autonomous inspection.

Research on SLAM and Path Planning Method of Inspection Robot in Complex Scenarios

Factory safety inspections are crucial for maintaining a secure production environment. Currently, inspections are predominantly performed manually on a regular basis, leading to low efficiency and a high workload. Utilizing inspection robots can significantly improve the reliability and efficiency of these tasks. The development of robot localization and path planning technologies ensures that factory inspection robots can autonomously complete their missions in complex environments. In response to the application requirements of factory inspections, this paper investigates mapping, localization, and path planning methods for robots. Considering the limitations of cameras and laser sensors due to their inherent characteristics, as well as their varying applicability in different environments, this paper proposes SLAM application systems based on multi-line laser radar and visual perception for diverse scenarios. To address the issue of low efficiency in inspection tasks, a hybrid path planning algorithm that combines the A-star algorithm and time elastic band method is introduced. This approach effectively resolves the problem of path planning becoming trapped in local optima in complex environments, subsequently enhancing the inspection efficiency of robots. Experimental results demonstrate that the designed SLAM and path planning methods can satisfy the inspection requirements of robots in complex scenarios, exhibiting excellent reliability and stability.

Research on Insulator Defect Detection System Based on Edge Intelligence

The paper aims at the problems of the high occupancy rate of communication resources and heavy server load caused by the image data taken during the inspection of the substation and sent to the remote server for processing. An edge intelligent inspection data processing architecture is constructed, and an insulator defect detection model based on transfer learning and YOLO v5 is proposed. The model can meet the training of a small sample network model in the case of scarcity of power equipment defect data. The proposed algorithm is compared with typical target detection algorithms, and the experimental results show that the proposed algorithm has an excellent performance in model recognition accuracy and recall rate. The trained insulator defect detection model is further deployed on the developed edge intelligent terminal for testing, and its image processing speed can meet the image processing requirements for intelligent inspection of substations. The system has good application value.

A Deep Learning-Based Surface Defects Detection and Color Classification Method for Solar Cells

In recent years, solar photovoltaic-based power generation technology has become the key planning direction of many countries around the world. In the process of making solar cells, the quality inspection requirements are very particular, such as physical damages, surface scratches, broken grids and microcracks. In traditional factory production, the detection of the above defects requires professional inspectors to carry out visual inspection, which often leads to low detection efficiency, subjective assumption and fatigue, as well as some detection errors. In recent years, the rapid development of computer vision makes it possible to be used to detect the defects in solar cells. To overcome existing barriers, this paper proposes a method for detecting surface defects in solar cells based on deep neural network. Specifically, a specified image segmentation model named U-Net is developed for this purpose. By automatically segmenting little objects using the proposed recognition approach, surface defects detection can be realized. At last, we use a set of experiments on images from real scenes to verify the proposed method.

A balanced walking-clamp mechanism for inspection robot of transmission line

PurposeThis paper aims to design a walking-clamp mechanism for the inspection robot of transmission line. The focus for this design is on climbing ability and obstacle-crossing ability with a goal to create a novel walking-clamp mechanism that can clamp not only the line but also the obstacle.Design/methodology/approachA novel clamping jaw used in the walking-clamp mechanism is proposed. The clamping wheel is mounted on the lower end of clamping jaw to reduce the friction between the clamping jaw and the line, and the top end of clamping jaw is designed as a hook structure to clamp the obstacle. The working principle and force states of the walking-clamp mechanism clamping the line and obstacle are analyzed, and the simulation and prototype experiments are carried out.FindingsThe experimental results show that this mechanism can clamp the obstacle steadily, and the clamping forces of the front and back pairs of clamping jaws are almost equal during robot walking along the catenary-shaped line. It is in agreement with the theoretical analysis, and it demonstrates that this mechanism can meet the working requirements of inspection robot.Practical implicationsThis novel mechanism can be used for inspection robot of transmission line, and it is beneficial for robot to complete long-distance inspection works.Social implicationsIt stands to reduce costs related to inspection and improve the inspection efficiency.Originality/valueInnovative features include its structure, working principle and force states.

Continuous monitoring with a permanently installed high-resolution ultrasonic phased array

High-resolution defect detection is an integral part of non-destructive testing (NDT). Ultrasonic array transducers have been developed to respond to this need and provide high-resolution images for such applications. At present, high-resolution imaging is achieved by ultrasonic array design, such as enhanced transducer materials, and optimised element geometry to fulfil inspection requirements and miniaturise imaging artefacts, etc. This paper presents a novel ultrasonic array and controller system design strategy that can miniaturise the array and reduce manufacturing cost of the entire array system. This then enables the permanent or semi-permanent installation of arrays onto structures and long-term continuous structural health monitoring (SHM). The array geometry is designed as a compromise between the required imaging performance for the desired application and the fabrication cost for permanent installation. A prototype array transducer is constructed from piezoceramic elements bonded to a flexible printed circuit (FPC) substrate and then permanently installed on example structures to demonstrate long-term monitoring. Improved long-term sensitivity to defect growth is achieved by applying a compensation strategy for velocity and phase changes with temperature. Results show that a 5 MHz, 18-element prototype array can detect changes as small as 0.1 mm in the diameter of a side-drilled hole (SDH) defect after 3 months of continuous monitoring.

A Hybrid Method of Frequency and Spatial Domain Techniques for TFT-LCD Circuits Defect Detection

Defect detection is a crucial but challenging task in thin film transistor liquid crystal display (TFT-LCD) manufacturing. Existing vision-based methods focus on either spatial domain or frequency domain with unsatisfactory detection. In view of that, this paper proposes a hybrid template matching method by drawing benefits from both frequency and spatial domain techniques. Under proposal, frequency domain template matching method and saliency detector method are firstly adopted separately to obtain two candidate frequency components associated with defects. In template matching process, a novel selection criterion is taken to improve identifying components associated with local spatial anomaly. Subsequently, the inverse Fourier transform is applied on the intersection of the two candidates to reconstruct the defect regions. In final steps, image entropy in the spatial domain is employed to filter out false detection regions to improve accuracy. To meet industry’s real-time inspection requirement, image partition and multi-threading calculation techniques are introduced into the proposed methodology. Experimental results have shown practical viability of the proposed approach to increase the yield rate of panels through robust defect detection in TFT-LCD manufacturing.

In-motion 3D reconstruction of high dynamic range surfaces.

Efficient and refined three-dimensional (3D) reconstruction of industrial parts has become an urgent need in the field of advanced manufacturing, and it's a great challenge when facing in-motion and online inspection requirements of high dynamic range (HDR) surfaces that have large reflectivity variations. This paper proposes a method using RGB line-scan cameras to realize in-motion multiple-shot 3D shape measurements with RGB channel fusion to increase the measurement dynamic range. First, multi-channel one-dimensional background-normalized Fourier transform profilometry (MC-1DBNFTP) is proposed as an effective in-motion HDR method. Second, for HDR surfaces with strongly overexposed areas, we propose a solution that obtains 6 results of different dynamic ranges for fusion with only 5 projected patterns, which further extends the measurement dynamic range while ensuring the small projection period. Third, we develop a fusion method based on reliability evaluation, which is more reliable than the existing methods in fringe projection systems. In addition, colored textures can be mapped to the reconstructed surfaces. Experimental results prove that the proposed method realizes accurate and reliable in-motion 3D reconstruction of HDR surfaces.

High-performance and space-qualified hybrid integrated circuit based on multi-group homogeneous gradient eutectic in quality conformance inspection

The quality and reliability of high integration electronic components has become one of the main issues restricting the advancement of aerospace products. During the development of a military-standard microwave hybrid integrated circuit, the penetration rate of the eutectic welding of the thin film substrates, thick film substrates, and carrier plates in the cavity directly affects the significant performance characteristics, such as gain and working current. By using of traditional process methods, the hybrid integrated circuit cannot pass the experiments (group A, group B, group C and group D) which presents the different types of grouping categories, test parameters, along with acceptance criteria that the quality conformance inspection (QCI) stipulated in the GJB 2438B General Specification for Hybrid Integrated Circuits, which is due to the low rate of finished products. We propose and experimentally demonstrate a multi-group homogeneous gradient eutectic-based method with penetration rate of >92 %, which ensured the grounding performance and reliability inspection requirements of the individual circuit unit. When the hybrid integrated circuit works under the temperature range −55~+125 °C and other environmental test conditions, the delta limits of working current and gain are kept in the extreme range of ±10 mA and ±2 dB which shows high stability and reliability. Then we finally achieved that the aerospace product successfully passed the QCI. Our research progress can expand the process applications of Class K(Space-Qualified) hybrid integrated circuits with the maximum level of assurance for use in extreme environment applications such as space where reliability is paramount.

RTLSeg: A novel multi-component inspection network for railway track line based on instance segmentation

The condition monitoring of railway track line is one of the fundamental tasks to ensure the safety of the railway transportation system. Railway track line is mainly made up of tracks, fasteners, bolts, backing plates, and so on. Given the requirements for rapid and accurate inspection, an innovative and intelligent method for multi-component identification and common defect detection of railway track line is investigated based on instance segmentation. More specifically, a railway track line image (RTL-I) dataset is constructed and annotated manually in this paper. After that, based on the work of YOLACT and YOLACT++, combined with prior knowledge, a railway track line image segmentation model (RTLSeg for short) is proposed. Firstly, taking the characteristics of the objects in the RTL-I dataset, preset anchors are redesigned and a feature enhanced module is introduced in the prediction head to improve the detection and segmentation accuracy of the model. Secondly, to strengthen the internal information propagation within the model, PaFPN (path aggregation feature pyramid network) is applied instead of FPN in RTLSeg. Thirdly, with the help of CoordConv, Coord-Protonet is presented to add position awareness explicitly to the model for more robust and higher quality prototype masks. Finally, to further improve the model performance, the attention mechanism is explored and a novel spatial attention-guided bounding box branch is employed in the enhanced prediction head. Both quantitative and qualitative experimental results show that the proposed method is feasible in detecting and segmenting multi-component and common defects of railway track line, and outperforms the compared baseline models. In particular, RTLSeg is able to achieve 91.35 bbox mAP and 91.60 mask mAP with the customized dataset. Meanwhile, the average inference speed reaches 13.07 fps. The average detection accuracy and recall are 100% and 99.83%, respectively. Furthermore, the effectiveness of each optimized part of the proposed RTLSeg model is demonstrated by additional ablation study.

Detailed infrared diagnostics of photovoltaic panels for higher safety and optimal energy generation

The Photovoltaic (PV) systems are one of the key renewable energy sources that are becoming increasingly popular, but they still have many drawbacks compared to conventional energy sources. Their main disadvantages are that they require a lot of installation space and they have low investment return ration. Both could be overcome by local installation of the PV plants as close as possible to the end consumer, and the best case is the PV panels to be installed on the buildings, but this could lead to higher fire risk for the residents. Different PV panels degrade with different speeds and work under slightly different conditions. These problems require at least periodic PV panel diagnostics and the thermal infrared (IR) inspection seems to be the best solution. The prerequisites for correct IR diagnostics are presented. One of the requirements for detailed IR inspection is the thermal camera to capture each PV cell with at least 5 by 5 pixels in horizontal and vertical directions. A methodology for pixel size calculation and IR monitoring system design is developed and presented in this work.

Photovoltaic Cell Panels Soiling Inspection Using Principal Component Thermal Image Processing

Intended for good productivity and perfect operation of the solar power grid a failure-free system is required. Therefore, thermal image processing with the thermal camera is the latest non-invasive (without manual contact) type fault identification technique which may give good precision in all aspects. The soiling issue, which is major productivity affecting factor may import from several reasons such as dust on the wind, bird mucks, etc. The efficient power production sufferers due to accumulated soil deposits reaching from 1%–7% in the county, such as India, to more than 25% in middle-east countries country, such as Dubai, Kuwait, etc. This research offers a solar panel soiling detection system built on thermal imaging which powers the inspection method and mitigates the requirement for physical panel inspection in a large solar production place. Hence, in this method, solar panels can be verified by working without disturbing production operation and it will save time and price of recognition. India ranks 3rd worldwide in the usage use age of Photovoltaic (PV) panels now and it is supported about 8.6% of the Nation’s electricity need in the year 2020. In the meantime, the installed PV production areas in India are aged 4–5 years old. Hence the need for inspection and maintenance of installed PV is growing fast day by day. As a result, this research focuses on finding the soiling hotspot exactly of the working solar panels with the help of Principal Components Thermal Analysis (PCTA) on MATLAB Environment.

KEC 기준을 적용한 저압 전기설비의 접지저항값에 따른 접지계통 안전성 분석

The KEC(Korea Electro-technical Code) is applied in accordance with the international standard conformity requirement for installation and inspection of domestic electrical equipments. The main differences between KEC and the past standard are classification of grounding methods according to the type of grounding system, and protection against electric shock. Accordingly, in the standard of human body protection from electric shock, the allowable dangerous voltage has become more important than the grounding resistance value, but in the case of design and construction of low voltage electrical equipment, it is difficult to calculate that. Therefore, in this paper we model the types of grounding systems that can be applied in Korea through simulation, and the safety of the grounding system according to the grounding resistance value in terms of electric shock protection is reviewed by simulating human electric shock case study.

Reclassifying Fracture Critical Members (Non-Redundant Steel Tension Members) to System Redundant Members in a Steel Arch Bridge

The American Association of State Highway and Transportation Officials (AASHTO) Load and Resistance Factor Design Bridge Design Specifications define fracture critical members (FCMs) as steel primary members, which are subject to tension and whose failure would probably cause the collapse of a portion of or the entire bridge. FCMs not only lead to additional material procurement and fabrication quality requirements, but bridges with FCMs need to meet the requirements of special mandatory bi-annual hands-on inspections in additions to the routine planned bridge inspections per the National Bridge Inspection Standards. The special mandatory inspections for FCMs are a considerable budgetary constraint for asset owners while still not providing definite safety proofs within the period in between the inspections. Designation of FCMs should be carefully scrutinized in a new bridge constructed with high-quality materials and well-designed details. In 2012, the Federal Highway Administration published a memorandum introducing system redundant members (SRMs), which are identified through refined analysis and exempted from the inspection requirements for FCMs. A new AASHTO Guide Specification released in 2018 outlines a detailed procedure for the reclassification of FCMs as SRMs. This paper summarizes the analyses and results of a detailed FEA performed for a newly designed steel arch bridge. The analyses considered potential failure scenarios at the edge girder of the steel arch bridge and demonstrated the available redundancy in the system to distribute the loads by satisfying the performance criteria set forth in the AASHTO Guide Specification. Reclassification of the FCMs as SRMs could help the asset owner to reallocate their strained resources more efficiently.

Automated UAV path-planning for high-quality photogrammetric 3D bridge reconstruction

The bridge models reconstructed from unmanned aerial vehicle (UAV) images via photogrammetry are often reported to have quality issues (e.g., high noise, insufficient resolution and precision loss) and thus restrict their application in bridge inspection. To address this problem, this paper proposes a novel Building Information Model (BIM)-based 3D path planning method for improving the quality of photogrammetric bridge models by optimising the UAV flight plan. This method firstly uses a simple BIM model as input and considers inspection and photogrammetry requirements to generate effective UAV viewpoints. Then it adjusts inaccessible UAV viewpoints that are in occupied space or cannot meet the flight safety rules. Finally, a feasible flight trajectory is generated through all valid viewpoints. To evaluate the performance of the proposed method, two prototypes were developed to automate the path planning and on-site image acquisition, respectively, and were tested on a real girder bridge. The results showed that, compared with the common UAV flight plan used in current practice, the proposed method could: (1) generate a more precise bridge model with fewer noise points and higher visual quality to support damage detection; and (2) significantly improve the efficiency of photogrammetric 3D bridge reconstruction with reduced human interventions in image collection and processing.

Detection of Marine Oil Spills Based on HOG Feature and SVM Classifier

Oil spill accidents have gradually increased due to the continuous development of marine transportation and petroleum processing industries. Monitoring and managing marine oil spills present important economic, social, and practical implications in preventing offshore oil pollution and maintaining ecological balance. Unmanned aerial vehicle (UAV) has become a suitable carrier for low-altitude oil spill detection because of their fast deployment and low cost. Thermal infrared remote sensing images are used as the research object in this study. A method around histogram of gradient (HOG) features combined with a support vector machine (SVM) is proposed for identifying oil spills at sea to improve the accuracy of offshore low-altitude oil spill recognition and realize all-weather monitoring of offshore oil spills in offshore waters. Steps for extracting HOG features and basic principles of the SVM classification are first investigated. Image preprocessing is then performed on collected thermal infrared image data to produce samples. HOG features of samples are extracted, and the radial basis function is selected as the kernel function for training the SVM classifier. HOG features of the infrared image to be tested are calculated and then sent to the classifier for identifying the oil spills. In addition, the proposed method is compared with the back propagation(BP) neural network method and local binary pattern (LBP) combined with the SVM classification method for analysis. The results show that the oil film recognition method based on the HOG feature and SVM has a recognition accuracy of 91.3% in the environment of small infrared oil film samples, which is significantly better than the BP and LBP-SVM recognition methods, and obtains a shorter training time. The method proposed in this study has obvious advantages in terms of small sample size and processing efficiency, can meet the requirements of all-weather inspection of oil film pollutants by UAV in offshore port areas, and has great application potential in the field of maritime supervision informatization in the future.

Development of a monoclonal antibody-based enzyme-linked immunosorbent assay for determination of vomitoxin (DON) in Coicis Semen

Coicis Semen is a common bulk medicinal material used for both medicine and food, which has the effect of promoting diuresis, draining dampness, invigorating the spleen and checking diarrhea. It is derived from Coix lacryma-jobi var. ma-yuen of the family Poaceae, and is easily contaminated by fungi such as Fusarium graminearum and F. flavum due to climate reasons to produce vomitoxin. The guiding principles for determination of vomitoxin in Chinese medicinal materials in Chinese Pharmacopoeia are mainly HPLC and LC-MS, which have long detection period and are time-consuming and laborious, and thus cannot meet the requirements of on-site quality inspection of drugs. The complete antigen of vomitoxin-protein was obtained by chemical derivatization of vomitoxin. The monoclonal antibody against vomitoxin was prepared by classic monoclonal antibody preparation technology, and enzyme-linked immunosorbent assay(ELISA) method for detection of vomitoxin in Coicis Semen was established through methodological investigation. The IC_(50) based on the ELISA for vomitoxin in Coicis Semen was 3.88 μg·L~(-1), and the average recoveries and the RSD were 77.32%-93.73% and 4.6%-9.7%, respectively. With the established ELISA method, the vomitoxin residue in 14 batches of Coicis Semen samples were determined and validated by LC-MS, and the correlation between the two assays was found to be 0.997 8, indicating that the established ELISA method could be used for quantitative determination of vomitoxin residue in Coicis Semen and could achieve the rapid quantitative determination of the vomitoxin residue.

Application Research of Inspection System Based on Intelligent Wearable Device in Communication Power Inspection

In order to effectively reduce the difficulty in communication power inspection and further improve the inspection efficiency of inspectors, this paper designs an inspection system based on smart wearable devices. Firstly, based on the analysis of the inspection task requirements of the communication power inspection, the overall structure of the wearable device intelligent inspection system is constructed, and the application of the wearable device intelligent inspection system in power engineering is studied; secondly, according to the inspection situation of the inspection personnel in the traditional inspection method, the key technical composition of the multi-functional inspection intelligent observation wearable lens is designed, and the key texture feature extraction technology of the communication power supply is used to locate the specific image target in the inspection process, and detect the local area. The abnormal sound is discharged, abnormal sound training and identification are completed, and intelligent inspection of the communication power supply is realized. Finally, it is verified by test experiments that the inspection system based on smart wearable devices designed in this paper can realize information transmission, task configuration, and data access in communication power inspection, with high work efficiency.

Bovine Tuberculosis - Analysis of 10-year cases and impact of visual inspection in the surveillance at the slaughterhouse in Portugal

Bovine Tuberculosis (bTB) is a One Health issue that interconnects human health with animal and environmental health. Due to its importance, cattle (the main host) it is under a national eradication programme in Portugal. Within it, meat inspection is considered a cost-effective procedure in bTB surveillance. The Commission's Implementing Regulation (EU) 2019/627 changed the post-mortem meat inspection requirements applied to cattle, essentially reducing handling and incisions in young cattle (<8 months). To study the impact of these changes on the surveillance of bovine Tuberculosis, data (age, sex, fitness, breed, lesions, laboratory results and post-mortem decisions) on suspected and confirmed cases of bTB detected at routine slaughter of the period from October 2010 to January 2020 were analysed.From a total of 3,279,548 slaughtered animals, meat inspection detected 808 suspected cases. Of these, 381 cases were laboratory confirmed. The estimated global prevalence was 11.6 cases/100,000, increasing with age: 3.5, 5.9 and 27.3 cases/100,000 in, respectively, the groups <8 months, 8 to 19 months, and ≥ 20 months. The prevalence was higher in females (16.3 vs 7.3 cases/100,000 in males) and in beef cattle (16.3 vs 1.4 cases/100,000 in dairy cattle). Among the confirmed cases, mediastinal and bronchial lymph nodes were the most reported locations. Based on the results, there is no evidence of a negative impact on the application of Regulation (EU) 2019/627 on bTB surveillance in animals ≥8 months. For the <8 months group the change to a visual inspection protocol could lead to a failure in the detection of bTB cases. Considering the epidemiological situation, the incision of the mediastinal, bronchial and retropharyngeal lymph nodes in cattle <8 months should be maintained.