Imaging Plate Research Articles

Devising a method for processing the image of a vehicle's license plate when shooting with a smartphone camera

This paper reports an improved method for processing the image of a vehicle's license plate when shooting with a smartphone camera. The method for processing the image of a vehicle's license plate includes the following stages: – enter the source data; – split the video streaming into frames; – preliminary process the image of a vehicle's license plate; – find the area of a vehicle's license plate; – refine character recognition using the signature of a vehicle's license plate; – refine character recognition using the combined results from frames in the streaming video; – obtain the result of processing. Experimental studies were conducted on the processing of images of a vehicle's license plate. During the experimental studies, the license plate of a military vehicle (Ukraine) was considered. The original image was the color image of a vehicle. The results of experimental studies are given. A comparison of the quality of character recognition in a license plate has been carried out. It was established that the improved method that uses the combined results from streaming video frames works out efficiently at the end of the sequence. The improved method that employs the combined results from streaming video frames operates with numerical probability vectors. The assessment of errors of the first and second kind in processing the image of a license plate was carried out. The total accuracy of finding the area of a license plate by known method is 61 % while the improved method's result is 76 %. It has been established that the minimization of errors of the first kind is more important than reducing errors of the second kind. If a license plate is incorrectly identified, these results would certainly be discarded at the character recognition stage.

Novel Fluorescence-Based High-Throughput FLIPR Assay Utilizing Membrane-Tethered Genetic Calcium Sensors to Identify T-Type Calcium Channel Modulators.

T-type voltage-gated Ca2+ channels have been implicated in many human disorders, and there has been increasing interest in developing highly selective and potent T-type Ca2+ channel modulators for potential clinical use. However, the unique biophysical properties of T-type Ca2+ channels are not conducive for developing high-throughput screening (HTS) assays to identify modulators, particularly potentiators. To illustrate, T-type Ca2+ channels are largely inactivated and unable to open to allow Ca2+ influx at -25 mV, the typical resting membrane potential of the cell lines commonly used in cellular screening assays. To address this issue, we developed cell lines that express Kir2.3 channels to hyperpolarize the membrane potential to -70 mV, thus allowing T-type channels to return to their resting state where they can be subsequently activated by membrane depolarization in the presence of extracellular KCl. Furthermore, to simplify the HTS assay and to reduce reagent cost, we stably expressed a membrane-tethered genetic calcium sensor, GCaMP6s-CAAX, that displays superior signal to the background compared to the untethered GCaMP6s or the synthetic Ca2+ sensor Fluo-4AM. Here, we describe a novel GCaMP6s-CAAX-based calcium assay utilizing a high-throughput fluorometric imaging plate reader (Molecular Devices, Sunnyvale, CA) format that can identify both activators and inhibitors of T-type Ca2+ channels. Lastly, we demonstrate the utility of this novel fluorescence-based assay to evaluate the activities of two distinct G-protein-coupled receptors, thus expanding the use of GCaMP6s-CAAX to a wide range of applications relevant for developing cellular assays in drug discovery.

Application of edge detection based on hexagonal image structure to delamination detection of carbon fiber reinforced polymer material

Delamination is one of the most critical damage modes in carbon fiber reinforced polymer (CFRP) materials, and is invisible to the naked eye. Although delamination defect images of CFRP materials can be obtained through ultrasonic tomography, it is still difficult to accurately identify the position and contours of defect images. In this study, four different delamination defect images of CFRP composite plates were obtained through a finite element simulation and fan-beam ultrasonic tomography. A hexagon algorithm based on interpolation is proposed that transforms the reconstructed defect images from square pixels into hexagonal pixels. The interpolation algorithm is based on the overlap between the square and hexagonal pixels. As the experiment results verify, the use of the hexagonal structure-based morphological method for edge detection significantly reduces the recognition error from 7.74% to 0.148% in comparison with a traditional quadrilateral structure. The experimental results also showed that this method can effectively reduce the artifact interference and obtain complete and accurate target edge information more effectively than a square structure.

CR Image Optimization and Radiation Dose Limitation With Collimation Adjusting on Cervical Radiography

Background: Radiological examination in cervical services, Radiographers often opening the collimation wider of the size of the imaging plate for fear of being cut. This is done because it is supported by the image cropping facility on the CR. So there is a habit of opening the collimation width as wide as the imaging plate, which is often done with the assumption that it is better to widen the collimation than cut off the object. Obviously this will increase the radiation dose in patients with basic limitations. According to ICRP, patient safety must refer to the principle of radiation protection ALARA namely in the shortest possible time to get quality radiographs and patients receive the most minimum radiation exposure as possible.Methods: This research is a descriptive analytic research conducted with a cross sectional approach. The study was conducted by varying the radiation field area to be used for cervical examination. There are 10 variations of collimation area with 3 (three) exposures in each collimation. Then measured contrast, noise and radiation dose. Radiation dose is calculated by means of a babyline device.Results: The results showed that the collimation setting did not affect contrast radiography in CR because collimation was not the main factor that could affect contrast radiography on CR. The collimation setting on cervical radiography has an effect on the radiation dose, so it is necessary to adjust the collimation to get the minimum dose possible. The most optimal collimation area on cervical radiographic examination was obtained at a size of 12 x 24 cm with the smallest noise reception and dose compared to others and had higher contrast. Average contrast value achieved was 0.45 with a dose of 39.23 µGy.Conclusion: The collimation setting on cervical radiography has an effect on the radiation dose, so it is very necessary to adjust the collimation to get the minimum dose possible. The most optimal collimation area on cervical radiographic examination is obtained at a size of 12 x 24 cm with the smallest noise reception and dose compared to others and has a higher contrast.

Neutron imaging with an inertial electrostatic confinement fusion neutron source.

A compact inertial electrostatic confinement (IEC) fusion neutron source was developed. Imaging tests using the cylindrical IEC neutron source were conducted with the indirect imaging plate (IP) method using dysprosium foil and an imaging plate. An array of B4C powder contained in a stainless-steel blade and Cd pins was successfully imaged. To calculate thermal neutron flux of the device, we tested multiple types of IP and IP scanners with the indirect IP method to create a calibration curve, and it was confirmed that dental IP scanners, which are more widely used than industrial ones, can be applied to the indirect IP method.

Hybridoma Screening by Antibody Capture: Detection of Cell-Surface Binding by Immunofluorescence.

If the antigen of interest is a cell-surface protein, immunofluorescence can be used to identify hybridomas secreting monoclonal antibodies to these proteins. They can be stained on microscope chamber slides, flat-bottomed cell culture plates (96, 48, or 24 well), or imaging plates (96 well). This protocol uses 96-well imaging plates.

An overview of tritium retention in dust particles from the JET-ILW divertor

Tritium (T) retention characteristics in dust collected from the divertor in JET with ITER-like wall (JET-ILW) after the third campaign in 2015–2016 (ILW-3) have been examined in individual dust particles by combining radiography (tritium imaging plate technique) and electron probe micro-analysis. The results are summarized and compared with the data obtained after the first campaign in 2011–2012 (ILW-1). The dominant component in ILW-1 dust was carbon (C) originating from tungsten-coated carbon fibre composite (CFC) tiles in JET-ILW divertor and/or legacy of C dust after the JET operation with carbon wall. Around 85% of the total tritium retention in ILW-1 dust was attributed to the C dust. The retention in tungsten (W) and beryllium (Be) dominated particles was 100 times smaller than the highest T retention in carbon-based particles. After ILW-3 the main component contributing to the T retention was W. The number of small W particles with T increased, in comparison to ILW-1, most probably by the exfoliation and fragmentation of W coatings on CFC tiles though T retention in individual W particles was smaller than in C particles. The detection of only very few Be-dominated dust particles found after ILW-1 and ILW-3 could imply stable Be deposits on the divertor tiles.

Effect of enveloping and disinfection methods on artefact formation on enveloped PSP plate images.

ObjectivesTo investigate the artefacts caused by different disinfection and protection methods that can be used for infection control of photostimulable phosphor (PSP) plates.MethodsThe plates that were enveloped with single or double envelopes were sprayed with an alcohol-containing solution or wiped with an alcohol-containing tissue. Four PSP groups with two plates in each group were formed (A = wiping single envelope, B = wiping double envelopes, C = spraying onto single envelope, and D = spraying onto double envelopes). Any artefacts (1 = no artefact, 2 = presence of artefact less than 0.5 cm wide, and 3 = presence of artefact larger than 0.5 cm wide) on the 12 acquired images were evaluated.ResultsArtefacts (score-3) occurred on the images of Group C-2 PSP plate after the 3rd exposure. According to the Kruskal–Wallis test the difference between the artefact score of the four methods was found significant. Post-hoc comparisons showed the Group C artefact scores were higher than the others and the difference was significant. The total time of enveloping and disinfection processes for groups was; A = 6.30 min, B = 7.58 min, C = 5.48 min, and D = 7.14 min.ConclusionsRegardless of the number of envelopes, wiping with a tissue was less likely to cause artefacts, while spraying was reliable only when double envelopes were used. However, using a tissue and double envelopes, which are less risky in terms of artefact, causes time loss, difficulty in manipulation, environmental pollution and high cost.

C-type Natriuretic Peptide–induced PKA Activation Promotes Endochondral Bone Formation in Hypertrophic Chondrocytes

Longitudinal bone growth is achieved by a tightly controlled process termed endochondral bone formation. C-type natriuretic peptide (CNP) stimulates endochondral bone formation through binding to its specific receptor, guanylyl cyclase (GC)-B. However, CNP/GC-B signaling dynamics in different stages of endochondral bone formation have not been fully clarified, especially in terms of the interaction between the cyclic guanine monophosphate (cGMP) and cyclic adenosine monophosphate (cAMP) pathways. Here, we demonstrated that CNP activates the cAMP/protein kinase A (PKA) pathway and that this activation contributed to the elongation of the hypertrophic zone in the growth plate. Cells of the chondrogenic line ATDC5 were transfected with Förster resonance energy transfer (FRET)–based cGMP and PKA biosensors. Dual-FRET imaging revealed that CNP increased intracellular cGMP levels and PKA activities in chondrocytes. Further, CNP-induced PKA activation was enhanced following differentiation of ATDC5 cells. Live imaging of the fetal growth plate of transgenic mice, expressing a FRET biosensor for PKA, PKAchu mice, showed that CNP predominantly activates the PKA in the hypertrophic chondrocytes. Additionally, histological analysis of the growth plate of PKAchu mice demonstrated that CNP increased the length of the growth plate, but coadministration of a PKA inhibitor, H89, inhibited the growth-promoting effect of CNP only in the hypertrophic zone. In summary, we revealed that CNP-induced cGMP elevation activated the cAMP/PKA pathway, and clarified that this PKA activation contributed to the bone growth–promoting effect of CNP in hypertrophic chondrocytes. These results provide insights regarding the cross-talk between cGMP and cAMP signaling in endochondral bone formation and in the physiological role of the CNP/GC-B system.

Analysis of photostimulable phosphor image plate artifacts and their prevalence.

BACKGROUND Digital radiography has recently been used in dentistry as a substitute for conventional film radiography worldwide. Digital imaging has many advantages and provides new possibilities for recording and interpreting radiographic data. This system uses different types of digital receptors.AIMTo detect the frequency, type, and reasons behind the appearance of intraoral image artifacts acquired by photostimulable phosphor plates (PSP).METHODSThis retrospective descriptive study was conducted in the oral and maxillofacial radiology unit of the dental clinics of the College of Dentistry, Princess Nourah University (PNU). All intraoral digital radiographs were acquired using (Gendex Expert DC., United States) an intraoral X-ray machine with 7 -mA, 65-kVP using a PSP system (Soredex DIGORA Optime imaging plate) and laser scanners (Soredex DIGORA Optime), which can house all sizes of reusable intraoral PSP sensor plates with image acquisition software (MIPACS Dental Enterprise viewer 3.2.2). A total of 50000 intraoral radiographs were retrieved from the clinical database from April 2018 to April 2020 to evaluate the reason, type, and solutions to these image artifacts.RESULTSOverall, 50000 intraoral digital radiographs were acquired in a two-year-period; that is, from April 2018 to April 2020. Of these, 3550 (7.1%) retakes were performed due to the presence of image artifacts. Periapical radiographs were the most common image type of intraoral retakes (80.8%). Imaging artifacts were divided into three categories: operator, plate and scanning errors. Out of 3550 retakes, 5%, 1.37%, and 0.73% were related to operator, plate, and scanning errors, respectively. The cone cut was the most common operator error (988 images), Bite marks were the most common plate error (276 images), and delayed scanning artifacts were the most common scanning errors (145 images). The calculated kappa value for interobserver reliability was 0.99, indicating almost perfect interobserver agreement. CONCLUSIONOur study discussed intraoral image artifacts that were characteristic of PSP, where the most common artifacts were bitemarks, image size reduction, scratches, and delayed scanning.

Ursodesoxycholic acid is an FFA4 agonist and reduces hepatic steatosis via FFA4 signaling

Ursodeoxycholic acid (UDCA) is a safe bile acid effective in reducing hepatic steatosis in non-alcoholic fatty liver disease (NAFLD). However, the mechanism of action linked to this effect is poorly defined. In the present study, we identified that UDCA acted as a free fatty acid receptor 4 (FFA4) agonist with EC50 of 10.4 ± 0.7 μM, and its activity was determined by dynamic mass redistribution, fluorometric imaging plate reader, inositol monophosphate and bioluminescence resonance energy transfer assays. Moreover, UDCA showed FFA4 selectivity over eleven other G protein-coupled receptors. Real-Time PCR and immunocytochemistry analyses showed that FFA4 was abundantly expressed in human hepatocytes HuH-7 cells. In an in vitro model of NAFLD induced by oleic acid (OA), UDCA downregulated lipid accumulation in HuH-7 cells and suppressed sterol-regulatory element binding protein-1c (SREBP-1c) mRNA expression. This suppression of SREBP-1c was restored when FFA4 expression was knocked down in siRNA assay. In a mouse model of hepatic steatosis, db/db mice were exposed to a high-fat diet (HFD), and treatment of UDCA or docosahexaenoic acid (DHA, an endogenous FFA4 agonist) effectively prevented body weight gain and hepatic fat deposition and reduced triglyceride (TG) levels in serum and liver. This study not only identified a new skeleton of FFA4 agonists, but also demonstrated that FFA4 signal was accounting for the protective effects of UDCA in the NAFLD treatment.

Surface defect detection of vehicle light guide plates based on an improved RetinaNet

Aiming at the vehicle navigation light guide plate (LGP) image characteristics, such as complex and gradient textures, uneven brightness, and small defects, this paper proposes a visual inspection method for LGP defects based on an improved RetinaNet. First, we use ResNeXt50 with higher accuracy under the same parameters as the backbone network, and propose the lightweight module Ghost_module to replace the 1 × 1 convolution in the lower half of the ResNeXt_block. This can reduce the resource parameters and consumption, and speed up training and inference. Second, we propose and use an improved feature pyramid network module to improve the feature fusion network in RetinaNet. It can more effectively fuse the shallow semantic information and high-level semantic information in the backbone feature extraction network, and further improve the detection ability of small target defects. Finally, the defect detection dataset constructed based on the vehicle LGP images collected at a industrial site, and experiments are performed on the vehicle LGP dataset and Aluminum Profile Defect Identification dataset (Aluminum Profile DID). The experimental results show that the proposed method is both efficient and effective. It achieves a better average detection rate of 98.6% on the vehicle LGP dataset. The accuracy and real-time performance can meet the requirements of industrial detection.

Automated food intake tracking requires depth-refined semantic segmentation to rectify visual-volume discordance in long-term care homes

Malnutrition is a multidomain problem affecting 54% of older adults in long-term care (LTC). Monitoring nutritional intake in LTC is laborious and subjective, limiting clinical inference capabilities. Recent advances in automatic image-based food estimation have not yet been evaluated in LTC settings. Here, we describe a fully automatic imaging system for quantifying food intake. We propose a novel deep convolutional encoder-decoder food network with depth-refinement (EDFN-D) using an RGB-D camera for quantifying a plate’s remaining food volume relative to reference portions in whole and modified texture foods. We trained and validated the network on the pre-labelled UNIMIB2016 food dataset and tested on our two novel LTC-inspired plate datasets (689 plate images, 36 unique foods). EDFN-D performed comparably to depth-refined graph cut on IOU (0.879 vs. 0.887), with intake errors well below typical 50% (mean percent intake error: -4.2%). We identify how standard segmentation metrics are insufficient due to visual-volume discordance, and include volume disparity analysis to facilitate system trust. This system provides improved transparency, approximates human assessors with enhanced objectivity, accuracy, and precision while avoiding hefty semi-automatic method time requirements. This may help address short-comings currently limiting utility of automated early malnutrition detection in resource-constrained LTC and hospital settings.

Intraoperative Imaging in Total Hip Arthroplasty Is Cost-Effective Regardless of Surgical Approach

BackgroundComponent positioning in total hip arthroplasty (THA) may be improved with utilization of intraoperative imaging. The purpose of this study is to determine if intraoperative imaging during THA is cost-effective. MethodsA break-even analysis was used as a model for cost-effectiveness, which incorporates cost of imaging (including direct charges and the additional time required for imaging), rate of revision surgery, and cost of revision surgery, yielding a final revision rate that needs to be achieved with use of intraoperative imaging in order for its use to be cost-effective. Absolute risk reduction (ARR) is determined by the difference between the initial revision rate and final revision rate. ResultsAt an anticipated institutional cost of $120 and requiring 4 additional minutes, intraoperative fluoroscopy would be cost-effective if the baseline rate of revision due to component mispositioning (0.62%) is reduced to 0.46%. Intraoperative flat plate radiographs ($127) are cost-effective at an ARR of 0.16%. Cost-effectiveness is achieved with lower ARR in the setting of lower imaging costs ($15, ARR 0.02%), and higher ARR with higher imaging costs ($225, ARR 0.29%). ARR for cost-effectiveness is independent of baseline revision rate, but varies with the cost of revision procedures. ConclusionAt current revision rates for component malpositioning, only 1 revision among 400 THAs needs to be prevented for the utilization of fluoroscopy (or 1 in 385 THAs with flat plate imaging), to achieve cost-effectiveness.

Evaluation of the microstructure, chemical composition, and image quality of different PSP receptors.

This study aimed to assess the microstructure, chemical composition, and image quality of different photostimulable phosphor plates (PSP). Four PSP systems, Express®, Digora®, VistaScan®, and Apixia,® were assessed. Five radiographs of a homogeneous acrylic phantom were obtained with the PSP of each system, to acquire a total of 20 images. The images were objectively evaluated for uniformity using mean grey and standard deviation (SD) of their grey values. PSP receptors were analyzed using scanning electron microscopy (SEM) to determine the thickness of the granule layer and the size of the granules. The chemical composition of the PSP receptors was analyzed using total reflection X-ray fluorescence (TXRF). VistaScan showed more uniform and higher density images than the other tested systems (p < 0.05), as well as the lowest SD of grey values (p < 0.05). Regarding the microstructure of the receptors, Digora and VistaScan had thicker granule layers than Express and Apixia, and VistaScan had smaller granules than Digora and Express (p < 0.05). Fourteen chemical elements were detected in the receptors, with barium being the element with the highest concentration in all PSP systems. The microstructure, chemical composition, and image quality varied among all four PSP receptors studied. VistaScan receptors showed the smallest variation in granule size, one of the thickest granule layers, and the most uniform and least noisy images.

Structure-From-Motion With Varying Principal Point

We consider the problem of structure-from-motion (SfM) for images with fixed calibration but varying principal point. This scenario occurs for archival imagery taken using historic glass plate and film cameras without fiducial markers, when images have been inconsistently cropped or when image plates are broken into multiple fragments.We derive initialisation and pose estimation methods and regularisation penalties tuned specifically for this scenario leading to a complete archival imagery SfM pipeline. We illustrate the performance of our methods on challenging real world examples from image archives. Specifically, we use archival images of the East coast of Greenland from the British Arctic Air Route Expedition (BAARE). This is of particular glaciological interest for measuring historic ice loss. We use a modern digital elevation model (ArcticDEM), masked to stable regions, as ground truth to evaluate our method.

Designing a toroidal crystal for monochromatic X-ray imaging of a laser-produced He-like plasma

Abstract In this study, a toroidal quartz ( $20\overline{2}3$ ) crystal is designed for monochromatic X-ray imaging at 72.3°. The designed crystal produces excellent images of a laser-produced plasma emitting He-like Ti X-rays at 4.75 keV. Based on the simulations, the imaging resolutions of the spherical and toroidal crystals in the sagittal direction are found to be 15 and 5 μm, respectively. Moreover, the simulation results show that a higher resolution image of the source can be obtained by using a toroidal crystal. An X-ray backlight imaging experiment is conducted using 4.75 keV He-like Ti X-rays, a 3 × 3 metal grid, an imaging plate and a toroidal quartz crystal with a lattice constant of 2d = 0.2749 nm. The meridional and sagittal radii of the toroidal α-quartz crystal are 295.6 and 268.5 mm, respectively. A highly resolved image of the microgrid, with a spatial resolution of 10 μm, is obtained in the experiment. By using similar toroidal crystal designs, the application of a spatially resolved spectrometer with high-resolution X-ray imaging ability is capable of providing imaging data with the same magnification ratio in the sagittal and meridional planes.

Preliminary evaluation of dual imaging plate intraoral radiography.

This study evaluated the contrast-to-noise ratio (CNR), spatial resolution, and subjective quality of dual imaging plates (DIP) intraoral radiography. The DIP and conventional single IP (CSIP) methods both used YCR DT-1 imaging plates (Yoshida Co.). The DIP, comprising a front IP (FIP) and back IP (BIP), was constructed. DIP images were synthesized from the FIP and BIP images. An aluminum step phantom was used to measure the CNR. A line pair gauge was used to measure the spatial imaging resolution. A phantom comprising a porcine mandible embedded in acrylic resin was used for subjective evaluation. The CNR of the DIP image was 32% higher than that of the FIP image. The spatial resolution achieved using the FIP, DIP, and CSIP was highly comparable except above 4 line pairs/mm, where that of the CSIP was highest. In subjective evaluation, the noise in the DIP images was significantly lower than in those obtained using the FIP and CSIP. The CNR of the DIP was higher than that of the FIP. The decrease in spatial resolution of the DIP was limited. The subjective image quality of the DIP was higher than that of the FIP.

Clinical comparison of intraoral CMOS and PSP detectors in terms of time efficiency, patient comfort, and subjective image quality.

PurposeThis study compared the effectiveness of complementary metal-oxide semiconductors (CMOS) and photostimulable phosphor (PSP) plates as intraoral imaging systems in terms of time efficacy, patient comfort, and subjective image quality assessment in real clinical settings.Materials and MethodsFifty-eight patients (25 women and 33 men) were included. Patients were referred for a full-mouth radiological examination including 1 bitewing radiograph (left and right) and 8 periapical radiographs for each side (left maxilla/mandible and right maxilla/mandible). For each patient, 1 side of the dental arch was radiographed using a CMOS detector, whereas the other side was radiographed using a PSP detector, ensuring an equal number of left and right arches imaged by each detector. Clinical application time, comfort/pain, and subjective image quality were assessed for each detector. Continuous variables were summarized as mean±standard deviation. Differences between detectors were evaluated using repeated-measures analysis of variance. P<0.05 was accepted as significant.ResultsThe mean total time required for all imaging procedures with the CMOS detector was significantly lower than the mean total time required for imaging procedures with PSP (P<0.05). The overall mean patient comfort scores for the CMOS and PSP detectors were 4.57 and 4.48, respectively, without a statistically significant difference (P>0.05). The performance of both observers in subjectively assessing structures was significantly higher when using CMOS images than when using PSP images for all regions (P<0.05).ConclusionThe CMOS detector was found to be superior to the PSP detector in terms of clinical time efficacy and subjective image quality.

Compression of Pulsed Infrared Thermography Data With Unsupervised Learning for Nondestructive Evaluation of Additively Manufactured Metals

Additive manufacturing (AM) of high-strength metals, which is typically based on laser powder bed fusion (LPBF), can introduce microscopic pores in the AM metal. Pulsed Infrared Thermography (PIT) offers several advantages for nondestructive imaging of subsurface defects in AM structures because the method is one-sided, non-contact and scalable to structures of arbitrary size. However, high-resolution PIT imaging results in the generation of a large volume of thermography data (&#x007E;TB), which creates challenges for the storage and transmission of data. Compression of thermography data requires an approach that achieves high data compression ratio while preserving weak thermal features corresponding to microscopic material defects. We investigate thermography data compression using several unsupervised learning (UL) algorithms, which include Principal Component Analysis (PCA), Independent Component Analysis (ICA), Exploratory Factor Analysis (EFA), Sparse Dictionary Learning (SDL), and a novel lightweight Thermography Compressive Sparse Autoencoder (TCSA). Algorithms are benchmarked using PIT experimental data obtained from imaging of a stainless steel plate with calibrated porosity defects imprinted with AM process. For all algorithms, we obtain compression ratio &#x003E;30 (highest compression of 46 is achieved with TCSA), and peak signal-to-noise ratio for reconstruction accuracy &#x003E;73dB. Compared to existing methods, advantages of UL algorithms include achieving high compression ratio while preserving weak features to allow extraction of microscopic material defects from images. UL-based methods have general applicability because they are adaptable to compression of different data types, and allow for memory-efficient training and rapid on-line augmentation of the model.