From flight to insight: An end-to-end aerial inspection framework for 3D modeling and defect detection of infrastructures

Abstract. With the development of real 3D model production technology and the expansion of application field, people pay more and more attention to the quality of real 3D model. However, how to measure the quality of today's real 3D model have been bothering its producers and users. In this paper, we analysed the quality model of real scene 3D model based on oblique photography from the perspective of the third party. Our analysis is guided by the application requirements of real scene 3D model, combined with the existing production technology level. Our analysis is guided by the application requirements of real scene 3D model, combined with the existing production technology level, we established the quality framework of real scene 3D model. This quality framework of real 3D model includes nine quality elements. Using this quality framework, we made a quality evaluation test in Yingjing County, Sichuan Provence. The test results show that the quality framework can fully reflect the quality of the real scene 3D model. The quality framework of real scene 3D model established in this paper solves the problem that it is difficult to evaluate the quality of real scene 3D model. The quality framework provides a basis for comprehensive and objective evaluation of real scene 3D model quality.

A conceptual framework for automatic modelling and conflict detection of 3D land-use regulation restrictions to support issuing planning permits

Three-dimensional (3D) defect detection provides an effective method for improving industrial production efficiency. However, the 3D dataset is scarce, which is valuable for the industrial production field. This study proposes a new approach for detecting defect point clouds, which can provide an end-to-end 3D defect detection model. A self-attention mechanism is used to enrich the semantic relationships between local neighborhood features and global features based on the connection between them. Through adding multi-channel features, the rich structural features of the target point cloud are obtained, and the defect areas are accurately segmented to finally complete the 3D point cloud defect detection task. Furthermore, the multi-feature fusion in the model makes the segmented defect regions closer to the ground truth. Our method outperforms four state-of-the-art point cloud segmentation methods in terms of both segmentation region accuracy and defect detection point cloud accuracy. In the field of 3D defect detection, it provides an effective method to detect 3D information of industrial products.

This paper presents a general texture mapping framework for image-based 3D modeling. It aims to generating seamless texture map for 3D model created by real-world photos under uncontrolled environment. Our proposed method addresses two challenging problems: 1) texture discontinuity due to system error in 3D modeling from self-calibration; 2) color/lighting difference among images due to real-world uncontrolled environments. The general framework contains two stages to resolve these problems. The first stage globally optimizes the registration of texture patches and triangle faces with Markov Random Field (MRF) to optimize texture mosaic. The second stage does local radiometric correction to adjust color difference between texture patches and then blend texture boundaries to improve color continuity. The proposed method is evaluated on several 3D models by image-based 3D modeling, and demonstrates promising results.

Damage assessment using the Lamb wave factorization method

Dark-field structured illumination microscopy for highly sensitive detection of 3D defects in optical materials

We present a framework for 3D spatial gesture design and modeling. A wearable input device that facilitates the use of visual sensors and body sensors is proposed for gesture acquisition. We adapted two different pattern matching techniques, Dynamic Time Warping (DTW) and Hidden Markov Models (HMMs), to support the registration and evaluation of 3D spatial gestures as well as their recognition. One key ingredient of our framework is a concept for the convenient gesture design and registration using HMMs. DTW is used to recognize gestures with a limited training data, and evaluate how the performed gesture is similar to its template gesture. In our experimental evaluation, we designed 18 example gestures and analyzed the performance of recognition methods and gesture features under various conditions. We discuss the variability between users in gesture performance.

Real-Time Quality Control in Additive Manufacturing: An AI-driven Internet of Things framework for 3D Printing Defect Detection and Prediction

Three-dimensional (3D) printing techniques are being rapidly adopted for prototyping and product development across fields as scientifically diverse as wind energy and regenerative medicine. Through materials processing advancements, the incorporation of nanomaterials within 3D printed parts and structures has begun to enable enhanced material functionalities. In this work, the optical properties of gold nanoparticles are harnessed via the development of functionalized printer filament to detect defects and missing print layers in 3D printed parts. Gold nanoparticles are incorporated within a poly(lactic acid) polymer host matrix, and filament compatible with stock 3D printers is fabricated. Consistent with Beer–Lambert’s Law for nanoparticles in solution, a linear relationship between absorbance intensity and the total number of print layers is observed. By analyzing changes in absorbance intensity, the presence, location, and extent of material defects as small as 0.2 mm are identified through a nondest...

A 1D model of glow low-pressure CO2 discharges is developed. In the framework of this model, simulation of stationary and repetitively pulsed discharges at pressure ranging from 0.5 to 5 Torr and current from 10 to 50 mA is performed. The obtained plasma characteristics are compared with the available experimental results and with the data evaluated based on the approximate 0D approach. The results of 0D and 1D calculations agree for most of plasma parameters, except for the molar fraction of CO molecules produced at CO2 dissociation by electron impact. Agreement between the measured and calculated, in the framework of the 1D model, values of the CO molar fraction is provided by modifying the expression of the dissociation rate constant vs the reduced electric field.

In electrical engineering and automated manufacturing systems, high-precision, non-contact defect detection and geometric measurement are key to ensuring reliable equipment operation and product quality control. The transition from qualitative defect detection to quantitative three-dimensional (3D) characterization remains a significant challenge in automated industrial inspection. To address these issues, this paper proposes a 3D visual weak-texture defect detection and quantification method for industrial inspection applications in electrical automation. This method is based on a binocular vision perception framework, integrating pixel-level semantic segmentation and robust stereo matching to achieve precise defect localization and 3D metrology. The system first employs a Multi-Scale Edge-Aware Segmentation Network (MSEA-Net) to extract defect regions with exceptional boundary fidelity. Subsequently, a Deformable Convolution and Attention-guided Stereo Matching Network (DCASM-Net) is proposed to reconstruct dense and accurate 3D point clouds from surfaces with weak or repetitive textures, a common challenge in industrial settings. The 2D segmentation masks are precisely mapped to the 3D coordinate space, enabling the direct computation of key physical dimensions such as length, width, depth, and projected area. Comprehensive experiments on a dedicated industrial defect dataset demonstrate that our system achieves state-of-the-art performance in both defect segmentation (F1-score: 96.2%, IoU: 93.1%) and stereo matching (≥ 3px error rate: 2.1% in non-occluded areas). The proposed framework can serve as a visual perception and metrology module in electrical automation systems and intelligent manufacturing equipment, providing a reliable technical path for automated 3D defect detection in industry.

An automated optical inspection (AOI) platform for three-dimensional (3D) defects detection on glass micro-optical components (GMOC)

An intelligent and automated 3D surface defect detection system for quantitative 3D estimation and feature classification of material surface defects

Modeling transportation infrastructure assets in three dimensions (3D) is becoming increasingly necessary for good management. Condition assessment, maintenance, operations, and construction activities are exploiting 3D models for improved visualization, communications, and process control. Acquiring 3D models rapidly can improve safety and productivity and is becoming feasible through approaches based on sparse range point clouds; however, although this approach has contextual advantages, it is ultimately limited in speed. Emerging Flash laser detection and ranging (LADAR) technology is opening up the possibility of 3D modeling at rates better than 1 Hz (real time). A framework for 3D modeling is presented that includes the dimension of time. In particular, the performance of the Flash LADAR technology is examined, and potential applications are explored. Technologies such as Flash LADAR will play an important role in real-time modeling of infrastructure assets in the near future.

Modeling transportation infrastructure assets in three dimensions (3D) is becoming increasingly necessary for good management. Condition assessment, maintenance, operations, and construction activities are exploiting 3D models for improved visualization, communications, and process control. Acquiring 3D models rapidly can improve safety and productivity and is becoming feasible through approaches based on sparse range point clouds; however, although this approach has contextual advantages, it is ultimately limited in speed. Emerging Flash laser detection and ranging (LADAR) technology is opening up the possibility of 3D modeling at rates better than 1 Hz (real time). A framework for 3D modeling is presented that includes the dimension of time. In particular, the performance of the Flash LADAR technology is examined, and potential applications are explored. Technologies such as Flash LADAR will play an important role in real-time modeling of infrastructure assets in the near future.