Search Results for image processing algorithms

This article reviews the essential parts of the complex process of quantitative image analysis to assist automatic image analysis in laboratories. It describes the basic difference between the bias of classical manual stereological analysis and quantitative image analysis. The article concentrates on the basic properties of digital measurements that are the core of quantitative image analysis. It provides a brief description of the specimen and apparatus preparation as well as the image acquisition. The article explains how to evaluate stereological parameters and provides the general rules and guidelines for optimization of image processing algorithms from the viewpoint of shape quantification. It concludes with examples that demonstrate the usefulness of automatic image analysis in comparison to manual methods.

Ultrasound is an ideal modality for nondestructive evaluation (NDE) because it enables the interior of objects to be assessed without the safety and access issues associated with radiography. This article summarizes the history of array usage in NDE and its relationship to medical applications. It discusses the mathematics behind classical beamforming, full matrix capture, and total focusing methods of imaging. The article shows how ultrasonic array data can be simulated by direct numerical methods (most commonly using finite-element methods), analytical methods, or hybrid methods. It also considers various methods of comparing the performance of arrays and imaging algorithms. The article provides a comparison of various advanced and nonlinear imaging algorithm and looks at some practical industrial applications of arrays. It concludes with some future perspectives for arrays in NDE.

Process optimization is the discipline of adjusting a process to optimize a specified set of parameters without violating engineering constraints. This article reviews data-driven optimization methods based on genetic algorithms and stochastic models and demonstrates their use in powder-bed fusion and directed energy deposition processes. In the latter case, closed-loop feedback is used to control melt pool temperature and cooling rate in order to achieve desired microstructure.

Machine vision, also referred to as computer vision or intelligent vision, is a means of simulating the image recognition and analysis capabilities of the human eye and brain system with digital techniques. The machine vision functionality is extremely useful in inspection, supervision, and quality control applications. This article presents a variety of machine vision functions for different purposes and provides a comparison of machine and human vision capabilities in a table. It discusses the processes of a machine vision system: image acquisition, image preprocessing, image analysis, and image interpretation. The article provides information on the uses of machine vision systems in three categories of manufacturing applications: visual inspection, identification of parts, and guidance and control applications.

In situ process monitoring includes any technologies that monitor or inspect during an additive manufacturing (AM) process. This article presents the types, process considerations, and challenges of in situ monitoring technologies that can be implemented during an AM process. The types include system health monitoring, melt pool monitoring, and layer monitoring. The article discusses data analysis, and provides an overview of the integration of sensors into AM machines.

Additive manufacturing (AM) is a revolutionary technology that fabricates parts layerwise and provides many advantages. This article discusses polymer AM processes such as material extrusion, vat photopolymerization (VPP), powder-bed fusion (PBF), binder jetting (BJ), material jetting (MJ), and sheet lamination (SL). It presents the benefits of online monitoring and process control for polymer AM. It also introduces the respective monitoring devices used, including the models and algorithms designed for polymer AM online monitoring and control.

Three-dimensional microscopy can be used to reveal the shape, distribution, and connectivity of three-dimensional (3D) features that lie buried within an opaque material. This article discusses several experimental techniques that can be used to generate 3D images. These include serial sectioning, focused ion beam tomography, atom probe tomography, and X-ray microtomography. Nine case studies are presented that represent the work of the various research groups currently working on 3D microscopy using serial sectioning and illustrate the variants of the basic experimental techniques. The article also discusses the techniques for reconstruction and visualization of 3D microstructures with advanced computer software and hardware.

Powder-bed additive manufacturing (AM) processes are some of the most commonly used techniques, necessitating the accurate measurement of powder flowability properties. This article discusses some powder flow tests that occur in powder-bed AM machines. These include the Hall/Carney flow test, bulk/tap density, rheometer, and the revolving or rotating drum technique. The three categories of powder properties that are available from rheometer experiments are discussed: bulk, dynamic flow, and shear properties. The article also describes the basic principles and applications of micro-X-ray computed tomography in studying powder porosity characteristics nondestructively.

The use of additive manufacturing (AM) is increasing for high-value, critical applications across a range of disparate industries. This article presents a discussion of high-valued engineering components predominantly used in the aerospace and medical industries. Applications involving metal AM, including methods to identify pores and voids in AM materials, are the focus. The article reviews flaw formation in laser-based powder-bed fusion, summarizes sensors used for in situ process monitoring, and outlines advances made with in situ process-monitoring data to detect AM process flaws. It reviews investigations of ML-based strategies, identifies challenges and research opportunities, and presents strategies for assessing anomaly detection performance.

This article covers in-line process monitoring of the metal additive manufacturing (AM) methods of laser and electron beam (e-beam) powder-bed fusion (PBF) and directed-energy deposition (DED). It focuses on methods that monitor the component directly throughout the build process. This article is organized by the type of AM process and by the physics of the monitoring method. The discussion covers two types of monitoring possible with the PBF process: monitoring the area of the powder bed and component and monitoring the melt pool created by the laser or e-beam. Methods for layer monitoring include optical and thermal methods that monitor light reflected or emitted in the visible and infrared wavelengths, respectively. Monitoring methods for laser directed-energy deposition (DED) discussed are those that measure the size and shape of the melt pool, the temperature of the melt pool, and the plasma generated by the laser as it interacts with the molten metal.

This article examines how cellular automaton (CA) can be applied to the simulation of static and dynamic recrystallization. It describes the steps involved in the CA simulation of recrystallization. These include defining the CA framework, generating the initial microstructure, distributing nuclei of recrystallized grains, growing the recrystallized grains, and updating the dislocation density. The article concludes with information on the developments in CA simulations.

Optimization plays a key role in the design of any structure or system, and electromagnetic devices are no exception. This article provides a description of the formulation of a design problem, and provides a review of the Paretian optimality. It focuses on nondominating sorting algorithms and multiobjective evolutionary strategy algorithms associated with evolutionary computing. The article provides information on field-based optimization problems. It also discusses the design of the pancake inductor that implies the solution of coupled electromagnetic and thermal fields, along with the use of optimal design procedures, to identify the best possible device or process.

This article presents the analytics challenges in additive manufacturing. It discusses the types and applications of data analytics. Data analytics can be classified into four types: descriptive, diagnostic, predictive, and prescriptive. The diverse applications of data analytics and machine learning include design, process-structure-properties (PSP) relationships, and process monitoring and quality control. The article also presents tools used for data analytics.

Computed tomography (CT) is an imaging technique that generates a three-dimensional (3-D) volumetric image of a test piece. This article illustrates the basic principles of CT and provides information on the types, applications, and capabilities of CT systems. A comparison of performance characteristics for film radiography, real-time radiography, and X-ray computed tomography is presented in a table. A functional block diagram of a typical computed tomography system is provided. The article discusses CT scanning geometry that is used to acquire the necessary transmission data. It also provides information on digital radiography, image processing and analysis, dual-energy imaging, and partial angle imaging, of a CT system.

This article describes the fundamental aspects of three nondestructive evaluation (NDE) methods of solid-state welds in terms of operation principles. These methods are radiography, ultrasound, and eddy current methods. The article provides examples of these NDE techniques performed on various types of flaws resulting from solid-state welding processes.

A number of nondestructive evaluation (NDE) methods, such as radiography, ultrasound, and eddy current, are available to detect flaws in solid materials. This article describes the fundamental aspects of these NDE methods in terms of operation principles. It presents some examples of the methods performed on various types of flaws resulting from solid-state welding processes.

This article discusses tools that are used for the systematic optimization of engineering designs. It focuses on the practical application of optimization technology in a computer-aided engineering environment. The article presents numerical optimization algorithms and provides some background on how these algorithms make decisions when searching for the optimal design. It also provides information on structural optimization, topology optimization, materials processing optimization, multidisciplinary optimization, and global optimization.

This article surveys common additive manufacturing (AM) data-acquisition methods, covering preprocess materials characterization in the lab, machine calibration in the field, in-process monitoring during a build, and the postprocess part inspections and tests. The focus is to identify acquisition-related metadata for AM data sets to improve data usability and reusability. Also included in the article are exemplar metadata definitions for a data set acquired from light-scattering-based particle size analysis.

This article reviews the characterization methods for producing 3-D microstructural data sets. The methods include serial sectioning by mechanical material removal method and focused ion beam tomography method. The article describes how these data sets are used in realistic 3-D simulations of microstructural evolution during materials processing and materials response. It also explains how the 3-D experimental data are actually input and used in the simulations using phase-field modeling and finite-element modeling.

The potential for introducing defects during processing becomes greater as the relative density of pressed and sintered powder metallurgy (PM) parts increases and more multilevel parts with complex geometric shapes are produced. This article discusses the potential defects in pressed and sintered PM parts: density variations, compaction and ejection cracks, microlaminations, poor degree of sintering, and voids from prior lubricant agglomerates. It describes the various methods applicable to green compacts: direct-current resistivity testing, radiographic techniques, computed tomography, and gamma-ray density determination. The article also discusses the methods for automated nondestructive testing of pressed and sintered PM parts: acoustic methods-resonance testing, eddy current testing, magnetic bridge comparator testing, ultrasonic techniques, radiographic techniques, gamma-ray density determination, and visual inspection.