Surface Detection Method Research Articles

A Binocular Color Line-Scanning Stereo Vision System for Heavy Rail Surface Detection and Correction Method of Motion Distortion.

Thanks to the line-scanning camera, the measurement method based on line-scanning stereo vision has high optical accuracy, data transmission efficiency, and a wide field of vision. It is more suitable for continuous operation and high-speed transmission of industrial product detection sites. However, the one-dimensional imaging characteristics of the line-scanning camera cause motion distortion during image data acquisition, which directly affects the accuracy of detection. Effectively reducing the influence of motion distortion is the primary problem to ensure detection accuracy. To obtain the two-dimensional color image and three-dimensional contour data of the heavy rail surface at the same time, a binocular color line-scanning stereo vision system is designed to collect the heavy rail surface data combined with the bright field illumination of the symmetrical linear light source. Aiming at the image motion distortion caused by system installation error and collaborative acquisition frame rate mismatch, this paper uses the checkerboard target and two-step cubature Kalman filter algorithm to solve the nonlinear parameters in the motion distortion model, estimate the real motion, and correct the image information. The experiments show that the accuracy of the data contained in the image is improved by 57.3% after correction.

The effect of MoS2 concentration on the thickness and tribology performance of electro-codeposited Ni–Cr–MoS2 coatings

A Ni–Cr–MoS2 coating with trivalent Cr was prepared on the surface of the cylinder liner by the composite co-deposition method to modify its tribology performance. The effects of MoS2 concentration on the properties of the coatings and tribological properties of cylinder liners were studied by surface detection methods and reciprocating wear tests. The results show that Ni–Cr–MoS2 coating has a smaller grain size, higher micro-hardness, and greater coating thickness, compared with Ni–Cr coating. Doping by MoS2 effectively reduces the friction coefficient of Ni–Cr coating from 0.142 to 0.107, the wear amount of coating decreases from 7.9 mg to 7.1 mg, and the scuffing time increases from 19 min to 56 min. It is found that the doping of MoS2 provided more nucleation sites and promoted the deposition of ions, which resulted in the refined grain and thicker coating. The formation of solid lubrication film by MoS2 during the wear process and more coating thickness are mainly responsible for the modification of the tribology performance of cylinder liners. This research provides a practical approach to prolong the service life of friction pairs.

Robust Airport Surface Object Detection Based on Graph Neural Network

Accurate and robust object detection is of critical importance in airport surface surveillance to ensure the security of air transportation systems. Owing to the constraints imposed by a relatively fixed receptive field, existing airport surface detection methods have not yet achieved substantial advancements in accuracy. Furthermore, these methods are vulnerable to adversarial attacks with carefully crafted adversarial inputs. To address these challenges, we propose the Vision GNN-Edge (ViGE) block, an enhanced block derived from the Vision GNN (ViG). ViGE introduces the receptive field in pixel space and represents the spatial relation between pixels directly. Moreover, we implement an adversarial training strategy with augmented training samples generated by adversarial perturbation. Empirical evaluations on the public remote sensing dataset LEVIR and a manually collected airport surface dataset show that: 1. our proposed method surpasses the original model in precision and robustness; 2. defining the receptive field in pixel space performs better than that on representation space.

Multi-View Dynamic Reflection Prior for Video Glass Surface Detection

Recent research has shown significant interest in image-based glass surface detection (GSD). However, detecting glass surfaces in dynamic scenes remains largely unexplored due to the lack of a high-quality dataset and an effective video glass surface detection (VGSD) method. In this paper, we propose the first VGSD approach. Our key observation is that reflections frequently appear on glass surfaces, but they change dynamically as the camera moves. Based on this observation, we propose to offset the excessive dependence on a single uncertainty reflection via joint modeling of temporal and spatial reflection cues. To this end, we propose the VGSD-Net with two novel modules: a Location-aware Reflection Extraction (LRE) module and a Context-enhanced Reflection Integration (CRI) module, for the position-aware reflection feature extraction and the spatial-temporal reflection cues integration, respectively. We have also created the first large-scale video glass surface dataset (VGSD-D), consisting of 19,166 image frames with accurately-annotated glass masks extracted from 297 videos. Extensive experiments demonstrate that VGSD-Net outperforms state-of-the-art approaches adapted from related fields. Code and dataset will be available at https://github.com/fawnliu/VGSD.

Revolutionizing Robotic Depalletizing: AI-Enhanced Parcel Detecting with Adaptive 3D Machine Vision and RGB-D Imaging for Automated Unloading.

Detecting parcels accurately and efficiently has always been a challenging task when unloading from trucks onto conveyor belts because of the diverse and complex ways in which parcels are stacked. Conventional methods struggle to quickly and accurately classify the various shapes and surface patterns of unordered parcels. In this paper, we propose a parcel-picking surface detection method based on deep learning and image processing for the efficient unloading of diverse and unordered parcels. Our goal is to develop a systematic image processing algorithm that emphasises the boundaries of parcels regardless of their shape, pattern, or layout. The core of the algorithm is the utilisation of RGB-D technology for detecting the primary boundary lines regardless of obstacles such as adhesive labels, tapes, or parcel surface patterns. For cases where detecting the boundary lines is difficult owing to narrow gaps between parcels, we propose using deep learning-based boundary line detection through the You Only Look at Coefficients (YOLACT) model. Using image segmentation techniques, the algorithm efficiently predicts boundary lines, enabling the accurate detection of irregularly sized parcels with complex surface patterns. Furthermore, even for rotated parcels, we can extract their edges through complex mathematical operations using the depth values of the specified position, enabling the detection of the wider surfaces of the rotated parcels. Finally, we validate the accuracy and real-time performance of our proposed method through various case studies, achieving mAP (50) values of 93.8% and 90.8% for randomly sized and rotationally covered boxes with diverse colours and patterns, respectively.

Fast surface signal extraction method for photon point clouds with strong background noise without prior altitude information.

It is extremely challenging to rapidly and accurately extract target echo photon signals from massive photon point clouds with strong background noise without any prior geographic information. Herein, we propose a fast surface detection method realized by combining the improved density-dimension algorithm (DDA) and Kalman filtering (KF), termed the DDA-KF algorithm, for photon signals with a high background noise rate (BNR) to improve the extraction of surface photon signals from spacecraft platforms. The results showed that the algorithm exhibited good adaptability to strong background noise and terrain slope variations, and had real-time processing capabilities for massive photon point clouds in large-scale detection range without prior altitude information of target. Our research provides a practical technical solution for single-photon lidar applications in deep space navigation and can help improve the performance in environments characterized by strong background noise.

Comparison of Two Contact Detection Methods for Ground Reaction Forces and Moment Estimation During Sidestep Cuts, Runs, and Walks.

Force platforms often limit the analysis of human movement to the laboratory. Promising methods for estimating ground reaction forces and moments (GRF&M) can overcome this limitation. The most effective family of methods consists of minimizing a cost, constrained by the subject's dynamic equilibrium, for distributing the force over the contact surface on the ground. The detection of contact surfaces over time is dependent on numerous parameters. This study proposes to evaluate two contact detection methods: the first based on foot kinematics and the second based on pressure sole data. Optimal parameters for these two methods were identified for walking, running, and sidestep cut tasks. The results show that a single threshold in position or velocity is sufficient to guarantee a good estimate. Using pressure sole data to detect contact improves the estimation of the position of the center of pressure (CoP). Both methods demonstrated a similar level of accuracy in estimating ground reaction forces.

Single-lens dynamic z-scanning for simultaneous in situ position detection and laser processing focus control

Existing auto-focusing methods in laser processing typically include two independent modules, one for surface detection and another for z-axis adjustment. The latter is mostly implemented by mechanical z stage motion, which is up to three orders of magnitude slower than the lateral processing speed. To alleviate this processing bottleneck, we developed a single-lens approach, using only one high-speed z-scanning optical element, to accomplish both in situ surface detection and focus control quasi-simultaneously in a dual-beam setup. The probing beam scans the surface along the z-axis continuously, and its reflection is detected by a set of confocal optics. Based on the temporal response of the detected signal, we have developed and experimentally demonstrated a dynamic surface detection method at 140–350 kHz, with a controlled detection range, high repeatability, and minimum linearity error of 1.10%. Sequentially, by synchronizing at a corresponding oscillation phase of the z-scanning lens, the fabrication beam is directed to the probed z position for precise focus alignment. Overall, our approach provides instantaneous surface tracking by collecting position information and executing focal control both at 140–350 kHz, which significantly accelerates the axial alignment process and offers great potential for enhancing the speed of advanced manufacturing processes in three-dimensional space.

An Effective and Efficient Method for Detecting Defective Areas of Auto Parts based on CNNs

We propose a method for identifying the flaws on the product surface using a deep Convolutional Neural Network (CNN), making them more efficient and effective. Our approach – Gradient-weighted Flaw Detecting (GWFD), uses the gradients of the significant layers (identify the flaws in a CNN) to produce a localization map identifying the flaw areas on the product surface in the image for detecting industrial products. The problem with manual inspection is that it requires human and material resources, and accuracy cannot be guaranteed. The crux of identifying the flaws on the product surface is that the method can extract the features of the flaws and achieve a precise location. Unlike previous surface detection methods, GWFD combines an attention mechanism with edge detection to precisely focus on the flaw area of the product surface. At the same time, simple heuristic fine-tuning and reinforcement learning are applied during the training stage. Experiments show that GWFD achieves the precise location and detection of the flaws compared with other methods such as SIFT, KAZE, SSD MobileNet (V3), and SINet. GWFD helps users successfully inspect the flawed area of auto parts and has always been an industry pioneer for identifying flawed areas on the product surface. We hope that our work will promote and complete the practical application in more production lines.

Inspection Robot and Wall Surface Detection Method for Coal Mine Wind Shaft

The coal mine wind shaft is an important ventilation channel in coal mines, and it is of great significance to ensure its long-term safety. At present, the inspection of wind shafts still depends on manual work, which has low reliability and high risk. There are two main problems in the shaft wall detection of ventilation shafts: (1) The humidity and dust concentration in ventilation shafts are high, which makes imaging difficult; (2) the cracks on the shaft wall are long and irregular, so it is impossible to acquire the information of the whole crack from a single photo. Firstly, the mapping analysis between the concentration of water vapor and dust in the wind shaft and the image definition is determined by experiments. Then, the inspection robot is designed to move along the axial and circumferential directions to get close to the shaft wall, and the rack-and-rail drive design is adopted to ensure the real-time position feedback of the robot. Then, through the crack parameter detection method based on depth learning, the movement direction of the robot is controlled according to the crack direction so as to ensure that the complete crack parameters are obtained. Finally, the crack detection algorithm is verified by experiments.

Study on the Reliability of X-ray Detection of Dissimilar Metal Welded Joints

Different steel welding joints are widely used in special equipment and often bear the role of high temperature and high pressure, and corrosive media. Because dissimilar steel welded joints are prone to residual stress and defects during welding, cracks may occur during use. Due to the tissue structure of heterogeneous steel and austenite, ultrasonic detection and surface detection methods are poor. This paper puts forward the ray detection method, including artificial defects, using the ray detection method. Detection results show that the heterogeneous steel welding joint internal defects and that the ray detection method has a good detection effect.

Illumination modelling for reconstructing the machined surface topography

Illumination estimation is crucial for surface detection methods based on visual image reconstruction. The luminance of the machined surface topography with desertification fractal structure is difficult to accurately describe using existing illumination models, which depend on the macrosurface geometric features and reflection characteristics. Therefore, machined surface topography illumination model under coaxial light microscopic vision is proposed in this paper for the online detection of complex surfaces and special functional surface topography in the manufacturing field. Based on light scattering theory, the luminance of the surface topography is studied under the coaxial light microscopic vision; the influencing factors of luminance of surface topography are revealed and the illumination model of machined surface topography is established. The experimental results show that less calculation error occurs when using the proposed illumination model, which can be used to more accurately describe the luminance of machined surface topography. The research results will lay a foundation for online detection and monitoring of machined surfaces.

Encapsulation for Sensing Element and Its Application in Asphalt Road Monitoring

The internal pavement structure is a “black box”; an accurate strain response for the pavement interlayer structure under vehicle load is hard to obtain by conventional road surface detection methods. This is due to the true strain field of the pavement structure, which means that the service state of the pavement cannot be accurately evaluated. This paper proposes an innovative strain sensor based on a carbon nanotube and epoxy (CNT/EP) composite to solve the current strain monitoring problem in asphalt pavement health monitoring. The CNT/EP composite encapsulation method was proposed, and the I-shaped strain sensor for asphalt pavement structure was developed. The strain–resistance response characteristics of the self-developed sensor were further investigated using a universal testing machine. The encapsulated sensor was used to monitor the strain of the asphalt mixture by means of a laboratory asphalt concrete beam and a practical pavement field. The results showed that the encapsulation method proposed in the study is suitable for CNT/EP material, which could guarantee the survivability and monitoring effectiveness of the self-developed sensor under the harsh environment of high temperature and pressure of asphalt mixture paving. The resistance of encapsulated sensor presents a linear relationship with strain. The laboratory and practical paving verified the feasibility of the self-sensor for strain monitoring of asphalt pavement. Compared to other post-excavating buried sensors, the self-developed sensor can be embedded in the pavement interlayer as the asphalt mixtures paving process, which can obtain the real strain response of pavement structure and reduce the perturbation of the sensor to the dynamic response of the pavement.

Surface detection method of glass fibre composites based on computer vision

Surface detection method of glass fibre composites based on computer vision

Energy efficient RANSAC algorithm for flat surface detection in point clouds

Mobile robots control systems achieve greater efficiency through the use of robust environmental analysis algorithms based on data collected from optical sensors such as depth cameras, Light Detection and Ranging sensors (LIDARs). These data sources provide information about control object environment in point cloud. The work of such algorithms, as a rule, is aimed at detecting the objects of interest and searching for the specified objects, as well as relocating its own position on the scene. There are many different approaches for solving object detection problem in point clouds, but most of them require high computational resources. In this work, many variations of the random sample consensus (RANSAC) method are analyzed for objects defined by a mathematical model of an analytical form. Statistical characteristics of data analysis were used to compare the methods. The results demonstrate the most energy efficient flat surface detection method that processes 60 RGB-D camera frames per second.

Surface detection method of glass fibre composites based on computer vision

Considering the high cost, low efficiency and poor real-time performance of manual inspection methods in detecting surface defects such as glass fibre imprints, resin build-up and wrinkles, in this paper, a machine vision-based method is proposed to detect surface defects of glass fibre composites. The method designs an automatic inspection platform using two high-resolution line scan cameras for image acquisition. The eight directional templates of the Kirsch operator are used to convolve the derivatives of the image pixel points respectively, and the largest template is selected to determine its edge direction, and the detection of surface defects is achieved by combining with the Canny operator. The experimental results show that the proposed algorithm can well suppress noise interference, improve the accuracy of edge localisation and detection, and well retain edge information while avoiding pseudo-edges.

Detecting geothermal anomalies using Landsat 8 thermal infrared remote sensing data in the Ruili Basin, Southwest China.

With the recent increase in global focus on green energy, the application of thermal infrared remote sensing data for the detection of geothermal anomalies has attracted wide attention as it can overcome the difficulty of using only ground surveying. This study aimed to highlight areas of geothermal anomalies with land surface temperature (LST) time series data in winter derived from thermal infrared remote sensing. To extract LST anomaly areas in the Ruili Basin for geothermal prospecting, nine types of data on the study area in winter during 2014 ~ 2021 from Landsat 8 were analyzed. Landsat 8 LST inversion data based on the mono-window algorithm (MWA) can be used to identify hot springs, volcanoes, and other heat-related phenomena. Superimposing LST anomalies for each cycle through drilling data, excluding the heat island effect, geothermal anomaly regions could be plotted. The results show that the accuracy of MWA LST varied within 2K, which is acceptable for geothermal energy and higher than those of the radiative transfer equation (RTE) algorithm and MODIS LST products. Three high-LST regions in the southeast of the study area were identified as geothermal anomaly areas (A, B, and C), and region B was further verified through a comprehensive field investigation of geothermal wells, supplemented by the temperature gradient (TG) method. The findings reveal that the distribution of geothermal anomaly areas and high-LST areas are highly consistent with the northeast trending fault structure; faults act as thermal channels and help in accurately detecting local LST anomalies. Overall, the infrared remote sensing method proved to be a valid technique for detecting LST anomalies. Considering the synergy between thermal infrared surface detection and subsurface exploration methods, the identification of known geothermal fields (B) and other possible areas (A and C) has significance in the upscaling of local geologic information to regional prospecting, thus providing a direction for future geothermal research.

Optical Rail Surface Crack Detection Method Based on Semantic Segmentation Replacement for Magnetic Particle Inspection.

Railway damage detection is of great significance in ensuring railway safety. The cracks on the rail surface play a key role in studying the formation and development process of rail damage, predicting the occurrence of rail defects, and then improving the service life of the rail. However, due to the small shape of the cracks, the typical detection method is relatively complicated, and the speed is quite slow. Although traditional magnetic particle inspection technology is fairly accurate at detection, it is costly and inconvenient to carry and install, while also limiting the detection speed and affecting the system's operation. In this paper, a semantic segmentation detection method is developed by using various collected rail surface crack data and deep learning through a neural network. By comparing the inspection of the same rail surface with magnetic particle inspection technology, only inexpensive cameras are used and the inspection speed is increased while maintaining relatively high accuracy. In addition, the method can achieve fast detection speeds if it is extended to be combined with high-frequency cameras. It is an economical, efficient, and environmentally friendly method for future rail surface detection.

Recent Advancements in Electrochemical Biosensors for Monitoring the Water Quality.

The release of chemicals and microorganisms from various sources, such as industry, agriculture, animal farming, wastewater treatment plants, and flooding, into water systems have caused water pollution in several parts of our world, endangering aquatic ecosystems and individual health. World Health Organization (WHO) has introduced strict standards for the maximum concentration limits for nutrients and chemicals in drinking water, surface water, and groundwater. It is crucial to have rapid, sensitive, and reliable analytical detection systems to monitor the pollution level regularly and meet the standard limit. Electrochemical biosensors are advantageous analytical devices or tools that convert a bio-signal by biorecognition elements into a significant electrical response. Thanks to the micro/nano fabrication techniques, electrochemical biosensors for sensitive, continuous, and real-time detection have attracted increasing attention among researchers and users worldwide. These devices take advantage of easy operation, portability, and rapid response. They can also be miniaturized, have a long-life span and a quick response time, and possess high sensitivity and selectivity and can be considered as portable biosensing assays. They are of special importance due to their great advantages such as affordability, simplicity, portability, and ability to detect at on-site. This review paper is concerned with the basic concepts of electrochemical biosensors and their applications in various water quality monitoring, such as inorganic chemicals, nutrients, microorganisms’ pollution, and organic pollutants, especially for developing real-time/online detection systems. The basic concepts of electrochemical biosensors, different surface modification techniques, bio-recognition elements (BRE), detection methods, and specific real-time water quality monitoring applications are reviewed thoroughly in this article.

Ground surface detection method using ground penetrating radar signal for sugarcane harvester base-cutter control

Sugarcane harvesters, which can automatically adjust the base-cutter height according to the topography, have attracted increasing interest. Existing ground probing approaches mainly rely on contact machineries, which exhibit slow responses and damage sugarcane plants. This paper presents a rapid and non-contact terrain sensing method using ground penetrating radar (GPR) to realise the automatic height control of the base-cutter. In this study, a 1.6-GHz air-coupled GPR antenna was first mounted on a knockdown roller to measure the vertical distance between the ground and the antenna. The geographic relationship between the detection height and cutter cylinder length was established, based on which the base-cutter height was adjusted by changing the cylinder length. The obtained GPR signal was analysed to locate the ground surface. Post-process techniques were used to enhance the GPR signal contrast between the air and ground layer. To evaluate the penetrating ability and accuracy of the GPR method, sugarcane stalks were densely placed on the soil surface to mimic the extreme conditions of sugarcane fields. The influences of the height/depth of respective highlands/potholes, slope of the ground, and the moisture content of the soil on the GPR imaging were systematically investigated. The numerical and laboratory results confirmed the high accuracy of the proposed GPR method. Hence, a field experiment was conducted, and the soil surface was clearly visualised by GPR imaging. This study demonstrates the significant potential of the GPR-based method for the detection of a sugarcane field surface, which verifies its feasibility for use in manufactured automatic sugarcane harvesters. • A ground penetrating radar (GPR) was proposed to sense sugarcane terrain. • Simulation and lab experiments were used to investigate the GPR penetration ability. • The GPR can clearly visualise the ground layer, even when covered with stalks. • A 1.6-GHz GPR antenna achieved high ground detection accuracy in a field survey.