Search

The removal of plastic contamination from cotton lint is an issue of top priority to the U.S. cotton industry. One of the main sources of plastic contamination showing up in marketable cotton bales is plastic used to wrap cotton modules produced by John Deere round module harvesters. Despite diligent efforts by cotton ginning personnel to remove all plastic encountered during module unwrapping, plastic still finds a way into the cotton gin’s processing system. To help mitigate plastic contamination at the gin, a machine-vision detection and removal system was developed that utilizes low-cost color cameras to see plastic coming down the gin-stand feeder apron, which upon detection, blows plastic out of the cotton stream to prevent contamination. This paper presents the software design of this inspection and removal system. The system was tested throughout the entire 2019 cotton ginning season at two commercial cotton gins and at one gin in the 2018 ginning season. The focus of this report is to describe the software design and discuss relevant issues that influenced the design of the software.

The removal of plastic contamination in cotton lint is an issue of top priority for the U.S. cotton industry. One of the main sources of plastic contamination appearing in marketable cotton bales is plastic used to wrap cotton modules on cotton harvesters. To help mitigate plastic contamination at the gin, automatic inspection systems are needed to detect and control removal systems. Due to significant cost constraints in the U.S. cotton ginning industry, the use of low-cost color cameras for detection of plastic contamination has been successfully adopted. However, some plastics of similar color to background are difficult to detect when utilizing traditional machine learning algorithms. Hence, current detection/removal system designs are not able to remove all plastics and there is still a need for better detection methods. Recent advances in deep learning convolutional neural networks (CNNs) show promise for enabling the use of low-cost color cameras for detection of objects of interest when placed against a background of similar color. They do this by mimicking the human visual detection system, focusing on differences in texture rather than color as the primary detection paradigm. The key to leveraging the CNNs is the development of extensive image datasets required for training. One of the impediments to this methodology is the need for large image datasets where each image must be annotated with bounding boxes that surround each object of interest. As this requirement is labor-intensive, there is significant value in these image datasets. This report details the included image dataset as well as the system design used to collect the images. For acquisition of the image dataset, a prototype detection system was developed and deployed into a commercial cotton gin where images were collected for the duration of the 2018–2019 ginning season. A discussion of the observational impact that the system had on reduction of plastic contamination at the commercial gin, utilizing traditional color-based machine learning algorithms, is also included.

Automating inspection of critical infrastructure such as sewer systems will help utilities optimize maintenance and replacement schedules. The current inspection process consists of manual reviews of video as an operator controls a sewer inspection vehicle remotely. The process is slow, labor-intensive, and expensive and presents a huge potential for automation. With this work, we address a central component of the next generation of robotic inspection of sewers, namely the choice of 3D sensing technology. We investigate three prominent techniques for 3D vision: passive stereo, active stereo, and time-of-flight (ToF). The Realsense D435 camera is chosen as the representative of the first two techniques wheres the PMD CamBoard pico flexx represents ToF. The 3D reconstruction performance of the sensors is assessed in both a laboratory setup and in an outdoor above-ground setup. The acquired point clouds from the sensors are compared with reference 3D models using the cloud-to-mesh metric. The reconstruction performance of the sensors is tested with respect to different illuminance levels and different levels of water in the pipes. The results of the tests show that the ToF-based point cloud from the pico flexx is superior to the output of the active and passive stereo cameras.

This paper presents an advanced methodology for the detection of damage in aircraft composite materials based on the sensor fusion of two image-based non-destructive evaluation techniques. Both of the techniques, phased-array ultrasonics and infra-red thermography, are benchmarked on an aircraft-grade painted composite material skin panel with stringers. The sensors systems for carrying out the inspections have been developed and miniaturized for being integrated on a vortex-robotic platform inspector, in the framework of a larger research initiative, the Horizon-2020 ‘CompInnova’ project.

Manual inspection of workpieces in highly flexible production facilities with small lot sizes is costly and less reliable compared to automated inspection systems. Reinforcement Learning (RL) offers promising, intelligent solutions for robotic inspection and manufacturing tasks. This paper presents an RL-based approach to determine a high-quality set of sensor view poses for arbitrary workpieces based on their 3D computer-aided design (CAD). The framework extends available open-source libraries and provides an interface to the Robot Operating System (ROS) for deploying any supported robot and sensor. The integration into commonly used OpenAI Gym and Baselines leads to an expandable and comparable benchmark for RL algorithms. We give a comprehensive overview of related work in the field of view planning and RL. A comparison of different RL algorithms provides a proof of concept for the framework’s functionality in experimental scenarios. The obtained results exhibit a coverage ratio of up to 0.8 illustrating its potential impact and expandability. The project will be made publicly available along with this article., Ministry of Economic Affairs of the state Baden-Württemberg

This survey presents an in-depth overview of the last 25 years of research within the field of image-based automation of Closed-Circuit Television (CCTV) and Sewer Scanner and Evaluation Technology (SSET) sewer inspection. The survey investigates both the algorithmic pipeline, and the datasets and corresponding evaluation protocols. As a result of the in-depth survey, several trends within the research field are revealed, discussed, and future research directions are proposed. Based on the conducted survey, we put forth a set of three recommendations, which we believe will further improve and open the research field, as well as make the future research more reproducible: 1) The introduction of free and public benchmark datasets, 2) Standardized evaluation metrics, and 3) Open-sourcing the associated code.

We propose a novel approach for surface finish classification of digitally fabricated structures using an industrial depth camera. Data collected at different viewpoints are jointly processed to derive the spatial distribution of features describing the reflectance, which is in turn related to the surface finish. The features can be used to classify the surfaces according to their finish e.g., for assessing the homogeneity or conformance. We apply the method to four sprayed plaster specimens of similar visual appearance but different roughness. Using nearest neighbor classification we achieve an accuracy of 97% for the plaster samples. The approach is a contribution towards real-time quality inspection in digital fabrication., Automation in Construction, 129, ISSN:0926-5805

Increasing the inspection quality and speed is essential in manufacturing applications, especially for large structures (e.g. modern aircrafts). Traditional ultrasonic manual scanning can be comprehensive, but lacks repeatability and is time-consuming. Several robotic non-destructive testing systems have been developed in recent years. Although high inspection rates have been achieved by the use of robotic arms, there is the need to furtherly increase the inspection speeds, to cope with the current industrial demands. For systems delivering robotic ultrasonic inspection through phased array probes, the current bottleneck is given by the time required to electrically fire all elements of the phased array probes, which limits the maximum scanning speed of the automated manipulators. This paper discusses the development of a multi-aperture beamforming method to focus the beam with multiple focusing points at a single firing. This work investigates this approach and the influence of different aperture excitations on the data quality. Experiments have been carried out using a 5MHz 32-element phased array probe manipulated by a KUKA robot. The results highlight the possibility to significantly improve the speed of automated inspection compared to linear beamforming, without compromising the inspection quality.Increasing the inspection quality and speed is essential in manufacturing applications, especially for large structures (e.g. modern aircrafts). Traditional ultrasonic manual scanning can be comprehensive, but lacks repeatability and is time-consuming. Several robotic non-destructive testing systems have been developed in recent years. Although high inspection rates have been achieved by the use of robotic arms, there is the need to furtherly increase the inspection speeds, to cope with the current industrial demands. For systems delivering robotic ultrasonic inspection through phased array probes, the current bottleneck is given by the time required to electrically fire all elements of the phased array probes, which limits the maximum scanning speed of the automated manipulators. This paper discusses the development of a multi-aperture beamforming method to focus the beam with multiple focusing points at a single firing. This work investigates this approach and the influence of different aperture excitat...

In large civil aircraft manufacturing, a time-consuming post-production process is the non-destructive inspection of wing panels. This work aims to address this challenge and improve the defects&rsquo, detection by performing automated aerial inspection using a small off-the-shelf multirotor. The UAV is equipped with a wide field-of-view camera and an ultraviolet torch for implementing non-invasive imaging inspection. In particular, the UAV is programmed to perform the complete mission and stream video, in real-time, to the ground control station where the defects&rsquo, detection algorithm is executed. The proposed platform was mathematically modelled in MATLAB/SIMULINK in order to assess the behaviour of the system using a path following method during the aircraft wing inspection. In addition, two defect detection algorithms were implemented and tested on a dataset containing images obtained during inspection at Airbus facilities. The results show that for the current dataset the proposed methods can identify all the images containing defects.

Physical and operational inspection of sewer pipelines is critical to sustaining an acceptable level of system serviceability. Emerging inspection tools in addition to developments in sensor and lens technologies have facilitated sewer condition assessment and increased the quality and consistency of provided data. Meanwhile, sewer networks are too vast to be adequately investigated manually so the development of innovative computer vision techniques for automation applications has become an interest point of recent studies. This review paper presents the current state of inspection technology practices in sewer pipelines. An overall inspection tool comparison was conducted and the advantages and disadvantages of each method were discussed. This was followed by a comprehensive review of recent studies on visual inspection automation using computer vision and machine learning techniques. Finally, current achievements and limitations of existing automation methods were debated to outline open challenges and future research for both infrastructure management and computer science researchers.

A combination of multiple nondestructive testing (NDT) methods speeds up the assessment of concrete and increases the precision. This is why the UIR-Scanner was developed at Warsaw University of Technology. The scanner uses an Impact-Echo (IE) method with a unique arrangement of multiple transducers. This paper presents the development of the IE module using numerical models validated with experimental testing. It was found that rectangular arrangement of four transducers with the impactor in the middle is optimal for quick scanning of area for faults and discontinuities, changing the method from punctual to volumetric. A numerical study of void detectability depending on its position with respect to the IE module is discussed as well. After confirmation of the findings of models using experimental tests, the module was implemented into the scanner.

International audience; Automated inspection of manufactured parts is increasingly getting attention as it helps to make a rapid decision on product conformity. In this context, the aim of this paper is to present a new scanner path planning method for part inspection using an industrial 6-axis robot The novelty of the approach is to generate a scan path with the control of the overlap between 2 adjacent scanning paths based on the use of the least-squares conformal maps, which stretches a 3D mesh surface on a 2D plane. Equidistant paths calculated in the 2D space are then transformed into equidistant paths in the 3D space. The effective performance of controlling the overlap can improve digitizing quality and save digitizing time by managing the coverage of the laser beam. Furthermore, the digitizing quality is also ensured by keeping a constant scanning distance and executing a continuous control of the scanner orientation relatively to the part surface for all the driven points of the scan path. An experimental application of this new approach is proposed for a laser-plane scanner mounted on an industrial robot with 6 degrees of freedom, which demonstrates the interest of such an approach.

A photonic imaging method that gives the possibility to measure egg quality was applied. Since the method is non-contact and non-destructive we believe that this photonic imaging method may be successfully integrated in the automated inspection systems in the poultry industry. The method involves scanning an invisible infrared light beam over the eggshell, allowing to detect possible cracks and reveal information about the structure of the eggshell. The high resolution, high quality measurements obtained through optical coherence tomography (OCT) make it feasible to be utilized as part of an automated inspection system. In this paper we present an OCT scan image of the egg tip and reconstructed volumetric images of the eggshell surface. The method enables the detection of small cracks on eggs and reveals the detailed inner structures of the cracks.

Non-destructive ultrasound inspection of metallic structures is a perpetual high-interest area of research because of its well-known benefits in industrial applications, especially from an economic point of view, where detection and localisation of defects in their most initial stages can help maintain high production capabilities for any enterprise. This paper is aimed at providing further validation regarding a new technique for detecting and localising defects in metals, the Matched Filter-based Dual Symmetric Path Inspection (MF-DSPI). This validation consists in demonstrating the consistency of the useful ultrasound echoes, within the framework of the MF-DSPI. A description of the MF-DSPI method and the related work of the authors with it are presented in this paper, along with an experimental setup used to obtain the data with which the useful echo consistency was studied. The four proposed methods are: signal envelope analysis, L2-norm criterion, correlation coefficient criterion and sliding bounding rectangle analysis. The aim of this paper is to verify the useful echo consistency (with the help of these four approaches), as the MF-DSPI method strongly relies on this feature. The results and their implications are discussed in the latter portion of this study.

Unstable movement in roller bearings like cage or roller slip can lead to damages or eventually even to an early break of the bearing. To prevent slip, inadequate operating states should be avoided. Therefore, it is necessary to study the dynamic behavior of the bearing. Unfortunately, there is only a limited range of measurement methods for the dynamic of bearing components. Two possible approaches are using solely a high-speed camera or the combination of an optomechanical image derotator and a high-speed camera. This work focuses on a proposal which is suitable for both. Initially, the influence of the rotational velocity in the images is eliminated. In the next step the measurement data is reduced to a region of interest which displays a particular rolling-element. A rolling element is equipped with a linear marker which, in the next stage, is segmented by a thresholding method to multiple regions. The region representing the marker is extracted from the background and the position is calculated by a Principle Component Analysis. Depending on the shift of the angular position and the lag time between two images, the rotational velocity of the rolling element is calculated. Thus, it is possible to determine whether the rolling element is operating under ideal conditions. In conclusion, it can be said that this approach enables a simple and flexible non-invasive method to depict the occurrence of roller slip in roller bearings. © 2017 SPIE.

In this article we introduce a visual measurement system intended for an automatic determination of material segregation in steel wires as a feedback quality control. Our system is based on an image processing which represents a non-destructive field in measurement area. Until our automated inspection system has been deployed, the measurement of segregation level was accomplished by a human inspector who represents inexact and subjective evaluation. Moreover our automated classification of individual steel wires is based on a certified official regulation. The mentioned automated classification of wires into pre-defined classes is carried out by a sequence of image processing steps. All these steps are progressively described in this article in-depth. In this article we introduce a visual measurement system intended for an automatic determination of material segregation in steel wires as a feedback quality control. Our system is based on an image processing which represents a non-destructive field in measurement area. Until our automated inspection system has been deployed, the measurement of segregation level was accomplished by a human inspector who represents inexact and subjective evaluation. Moreover our automated classification of individual steel wires is based on a certified official regulation. The mentioned automated classification of wires into pre-defined classes is carried out by a sequence of image processing steps. All these steps are progressively described in this article in-depth.

The increasing number of skyscrapers along with the large number of tall bridges throughout the world also increases the demand of a robust, automated and remotely controlled health monitoring system for civil architectures. It is very difficult and sometimes not feasible to inspect the structures whose heights are beyond the limit of an average traditional structure of the same type. Therefore, in this paper an unmanned aerial vehicle is utilized to provide real time images of the structural site. A gradient of temporal range of images is used for such applications but the uncertainties caused by the camera locations make it quite difficult to evaluate the images from a same position on the structure to reveal any apparent structural damage. These images are, therefore, pre-processed for registration and are then classified automatically. A Speeded Up Robust Features (SURF) based feature detection algorithm is the heart of the scheme presented here in order to determine its performance in image registration and classification for civil structures. Also, the damage detection has been shown, which is achieved using the complete algorithm presented here. Scopus

In the last few years the need for methodologies capable of performing an automated geometric inspection has increased. These methodologies often use 3D high-resolution optical digitisers to acquire points from the surface of the object to be inspected. It is expected that, in the near future, geometric inspection will be requiring more and more the use of these instruments. At present geometric inspection is not profiting from all the opportunities attainable by 3D high-resolution optical scanners or from the numerous tools which can be used for processing the point cloud acquired from the inspected product. For some years now, these authors have been working on a new methodology for automatic tolerance inspection working from a 3D model acquired by optical digitisers. In this paper all the information recognisable in a scanned object is organised into a new data structure, called Recognised Geometric Model (RGM). The final aim is to define a representation of the inspected object for the automatic evaluation of the non-idealities pertaining to the form, orientation and location of the non-ideal features of the acquired object. The key concept of the proposed approach is the capability to recognise some intrinsic nominal properties of the acquired model. These properties are assumed as references to evaluate the non-idealities of the inspected object. With this approach the references of geometric inspection are searched for in the inspected object independently of a tolerance specification and of the availability of a 3D nominal representation. The high-level geometric information within RGM depends on the rules used for its identification. The capability to recognise specific categories of nominal references offers the possibility of introducing new tolerances to be specified. The proposed approach has been implemented in original software by means of which a specific test case has been analysed.

Due to increasing consumer demand for healthy, high quality, and safe food, food processing industries are exploring/adopting new non-destructive, fast, and accurate techniques for quality and safety inspection of their products. Optical sensing and spectroscopic techniques have high potential for automated real-time quality and safety inspection of agricultural and food products. These techniques have already been adopted successfully in several food processing units for automated quality monitoring of their products. This chapter discusses various optical sensing techniques under three broad categories namely: spectroscopic, fiber optic and image sensing. Working principles, instrumentation, advantages, disadvantages, and limitations of these techniques have been described and various applications for quality monitoring in the agriculture and food processing industries have been reviewed. Refereed/Peer-reviewed

International audience; In this paper the authors present an original methodology aiming at the automation of the geometric inspection, starting from a high-density acquired surface. The concept of intrinsic nominal reference is herein introduced in order to evaluate geometric errors. Starting from these concepts, a new specification language, which is based on recognizable geometric entities, is defined. This work also proposes some surface differential properties, such as the intrinsic nominal references, from which new categories of form errors can be introduced. Well-defined rules are then necessary for the unambiguous identification of these intrinsic nominal references. These rules are an integral part of the tolerance specification. This new approach requires that a recognition process be performed on the acquired model so as to automatically identify the already-mentioned intrinsic nominal references. The assessable errors refer to recognizable geometric entities and their evaluation leaves the nominal reference specification aside since they can be intrinsically associated with a recognized geometric shape. Tolerance specification is defined based on the error categories which can be automatically evaluated and which are an integral part of the specification language.

Industrial inspection of micro-devices is often a very challenging task, especially when those devices are produced in large quantities using micro-fabrication techniques. In the case of microlenses, millions of lenses are produced on the same substrate, thus forming a dense array. In this article, we investigate a possible automation of the microlens array inspection process. First, two image processing methods are considered and compared: reference subtraction and blob analysis. The criteria chosen to compare them are the reliability of the defect detection, the processing time required per frame, as well as the sensitivity to image acquisition conditions, such as varying illumination and focus. Tests performed on a real-world database of microlens array images led to select the blob analysis method. Based on the selected method, an automated inspection software module was then successfully implemented. Its good performance allows to dramatically reduce the inspection time as well as the human intervention in the inspection process.

Optimization of design is an important step in obtaining tissue engineering scaffolds with appropriate shapes and inner microstructures. Different shapes and sizes of scaffolds are modeled using UGS NX 6.0 software with variable pore sizes. The quality issue we are concerned is the scaffold porosity, which is mainly caused by the fabrication inaccuracies. Bone scaffolds are usually characterized using a scanning electron microscope, but this study presents a new automated inspection and classification technique. Due to many numbers and size variations for the pores, the manual inspection of the fabricated scaffolds tends to be error-prone and costly. Manual inspection also raises the chance of contamination. Thus, non-contact, precise inspection is preferred. In this study, the critical dimensions are automatically measured by the vision camera. The measured data are analyzed to classify the quality characteristics. The automated inspection and classification techniques developed in this study are expected to improve the quality of the fabricated scaffolds and reduce the overall cost of manufacturing.

In recent years, there have been a number of reported studies on the use of non-destructive technique to evaluate and determine mango maturity and ripeness levels. However, most of these reported works were conducted using single-modality sensing systems, either using an electronic nose (e-nose), acoustics, CCD, IR sensor or by other non-destructive measurements. This paper presents the work on the classification of mangoes (Magnifera Indica cv. Harumanis) maturity and ripeness levels using data fusion of the electronic nose (e-nose) and acoustic sensor and combine with CCD and IR sensor. A Fourier-based shape separation method was developed from CCD camera images to grade mango by its shape and able to correctly classify 100%. Colour intensity from infrared image was used to distinguish and classify the level of maturity and ripeness of the fruits. The finding shows 92% correct classification of maturity levels by using infrared vision Three groups of samples each from two different harvesting times (week 7 and week 8) were evaluated by the e-nose and then followed by the acoustic sensor. By applying a low level data fusion technique on the e-nose and acoustic data, the classification for maturity and ripeness levels using LDA was improved.