http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/108 Fri, 17 Apr 2026 17:31:24 GMT 2026-04-17T17:31:24Z http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10043 Title: Behaviour of concrete specimens retrofitted with bio-based polyurethane coatings under dynamic loads Authors: Somarathna, H.M.C.C.; Raman, S.N.; Mohotti, D.; Mutalib, A.A.; Badrie, K.H. Abstract: Experimental investigationwasconductedusingconcretespecimenstoassesstheeffectivenessofbiobasedpolyurethane(PU)coatingsynthesizedfrompalmkerneloilinenhancingthedynamicmechanical responseofconcretespecimensunderquasi-staticanddynamicloads.Thedynamicloadingcondition wassimulatedbyconductingthree-pointbendingtestsatastrainrateof0.067s 1,andsimultaneously, under quasi-static loading(strainrateof0.00033s 1)conditions .The application of PU layer(s)(either on the impact face, reface ,or on both faces of the concrete specimens)increases the dynamic resistance of theconcreteelement,whichcanbeenhancedbyincreasingcoatingthicknessoneitherfaceoftheconcreteelement.Underdynamicconditions,with10%of total coating thickness compared to the beam depth, strainduringultimatefailure,andstrainenergydensitywereenhancedsignificantlywithmarginalenhancementintheultimateflexuralstrength.PUcoatingdoesnotdebondduringultimatefailure of the test specimens which implies good adhesion characteristics, and even with minimum coating thickness(2.5%),drasticfragmentationeffectscanbereduced.Bio-basedPUisagreenmaterialandapplicationof PU coating provides available and sustainable technique for protecting concrete structures against dynamic loads. Fri, 01 Jan 2021 00:00:00 GMT http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10043 2021-01-01T00:00:00Z http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10042 Title: Automated gastrointestinal abnormalities detection from endoscopic images Authors: Gowtham, P.; Niranjan, M.; Kaneswaran, A. Abstract: Impressive high performance reported in the use of machine learning on computer vision problems is often due to the availability of very large datasets with which deep neural networks can be trained. With inference from medical images, however, this is not the case and available data is often only a small fraction in size in comparison to benchmark natural scene recognition problems. To circumvent this problem, transfer learning is often applied, where a model trained on a large natural image corpus is adapted, or pre-trained, to model the medical problem. In this work, we consider transfer learning applied to a specific medical diagnostics problem, that of abnormality detection in the gastrointestinal tract of a human body using images obtained during endoscopy. We carry out a search over several image recognition architectures and adapt pretrained models to the endoscopy problem. Using the benchmark KVASIR dataset, we show that transfer learning is effective in outperforming previously reported results, at an accuracy of 98.5±0.27. Sat, 01 Jan 2022 00:00:00 GMT http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10042 2022-01-01T00:00:00Z http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10024 Title: Smart Shopping Trolley Authors: Vinoth Kumar, L.; Poornima, B.; Nithusiga, S.; Anitha valli, R. Abstract: This paper targeted to reduce the Queue at a billing counter in a shopping complex. The system does the same by displaying the price of the product inside the cart. In this way the customer can directly pay the amount at the billing counter and leave with the commodities he/she has bought. It eliminates the traditional scanning of products at the counter and in turn speeds up the entire process of shopping, also with this system the customer shall know the total amount to be paid and hence can accordingly plan his shopping by only buying the essential commodities resulting in enhanced savings. Since the entire process of billing is automated it reduces the possibility of human error substantially. Also the system has a feature to delete the scanned products to further optimize the shopping experience of the customer. The hardware for the test run is based on the Arduino platform and zigbee modules, as both are very popular in small-scale research and wireless automation solution. Also the amount can be paid through online using QR code scanner. Wed, 01 Jan 2020 00:00:00 GMT http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10024 2020-01-01T00:00:00Z http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10023 Title: Artificial Intelligence-Assisted Loop Mediated Isothermal Amplification(AI-LAMP) for Rapid Detection of SARS-CoV-2 Authors: Rohaim, M.A.; Clayton, E.; Sahin, I.; Julianne, V.; Khalifa, M.E.; Al-Natour, M.Q.; Bayoumi, M.; Poirier, M.C.; Branavan, M.; Mukunthan, T.; Chaudhry, N.S.; Ravinder, S.; Amy, B.; Peter, B.; Hacking, W.; Botham, J.; Boyce, J.; Wilkinson, H.; Williams, C.; Whittingham-Dowd, J.; Shaw, E.; Matt Hodges; Lisa Butler; Bates, M.D.; Roberto La, R.; Balachandran, V.; Anil, F.; Munir, M. Abstract: Untilvaccinesandeffectivetherapeuticsbecomeavailable,thepracticalsolutiontotransit safelyoutofthecurrentcoronavirusdisease19(CoVID-19)lockdownmayincludetheimplementation ofaneffectivetesting,tracingandtrackingsystem.However,thisrequiresareliableandclinically validateddiagnosticplatformforthesensitiveandspecificidentificationofSARS-CoV-2.Here, wereportonthedevelopmentofadenovo,high-resolutionandcomparativegenomicsguided reverse-transcribedloop-mediatedisothermalamplification(LAMP)assay.Tofurtherenhancethe assayperformanceandtoremoveanysubjectivityassociatedwithoperatorinterpretationofresults, weengineeredanovelhand-heldsmartdiagnosticdevice.Therobustdiagnosticdevicewasfurther furnishedwithautomatedimageacquisitionandprocessingalgorithmsandthecollateddatawas processedthroughartificialintelligence(AI)pipelinestofurtherreducetheassayruntimeandthe subjectivityofthecolorimetricLAMPdetection.ThisadvancedAIalgorithm-implementedLAMP (ai-LAMP)assay,targetingtheRNA-dependentRNApolymerasegene,showedhighanalytical sensitivityandspecificityforSARS-CoV-2.Atotalof~200coronavirusdisease(CoVID-19)-suspected NHSpatientsamplesweretestedusingtheplatformanditwasshowntobereliable,highlyspecific andsignificantlymoresensitivethanthecurrentgoldstandardqRT-PCR.Therefore,thissystemcould provideanefficientandcost-effectiveplatformtodetectSARS-CoV-2inresource-limitedlaboratories. Wed, 01 Jan 2020 00:00:00 GMT http://repo.lib.jfn.ac.lk/ujrr/handle/123456789/10023 2020-01-01T00:00:00Z