Matches in SemOpenAlex for { <https://semopenalex.org/work/W3149894872> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W3149894872 endingPage "101286" @default.
- W3149894872 startingPage "101286" @default.
- W3149894872 abstract "Deep learning has shown to achieve dramatic results in image classification tasks. However, deep learning models require large amounts of data to train. Most of the real-world datasets, generally insect classification data does not have large number of training dataset. These images have a large amount of noise and various differences. The paper proposes a novel architectural model which removes the background noise and classify the Tiger beetles. Here object location is identified using contours by converting the original coloured image to white on black background. Then the remaining background is eliminated using grabcut algorithm. Later the extracted images are classified using a modified SqueezeNet transfer learning model to identify the tiger beetle class up to genus level. Transfer learning models with fewer trainable parameters performed well than the total number of parameters in the original model. When evaluating results it was identified that by freezing uppermost layers of SqueezeNet model better accuracy can be gained while freezing lowermost layers will reduce the validation accuracy. The proposed model achieved more than 90% for the test set in 40 epochs using 701,481 trainable parameters by freezing the top 19 layers of the original model. Improving the pre-processing to localize insect has improved the accuracy." @default.
- W3149894872 created "2021-04-13" @default.
- W3149894872 creator A5002853978 @default.
- W3149894872 creator A5006497714 @default.
- W3149894872 creator A5066899248 @default.
- W3149894872 creator A5084039823 @default.
- W3149894872 date "2021-05-01" @default.
- W3149894872 modified "2023-10-06" @default.
- W3149894872 title "Deep learning approach to classify Tiger beetles of Sri Lanka" @default.
- W3149894872 cites W2006587375 @default.
- W3149894872 cites W2021005899 @default.
- W3149894872 cites W2036816374 @default.
- W3149894872 cites W2055728185 @default.
- W3149894872 cites W2061572659 @default.
- W3149894872 cites W2116562978 @default.
- W3149894872 cites W2124351162 @default.
- W3149894872 cites W2129860849 @default.
- W3149894872 cites W2140047820 @default.
- W3149894872 cites W2163345210 @default.
- W3149894872 cites W2165698076 @default.
- W3149894872 cites W2285671993 @default.
- W3149894872 cites W2398330969 @default.
- W3149894872 cites W2564288310 @default.
- W3149894872 cites W2598645336 @default.
- W3149894872 cites W2769019609 @default.
- W3149894872 cites W2770456481 @default.
- W3149894872 cites W2791630193 @default.
- W3149894872 cites W2820540054 @default.
- W3149894872 cites W2902114955 @default.
- W3149894872 cites W2919681993 @default.
- W3149894872 cites W2962949934 @default.
- W3149894872 cites W2964298670 @default.
- W3149894872 cites W2997849610 @default.
- W3149894872 cites W3014860971 @default.
- W3149894872 cites W3022859973 @default.
- W3149894872 cites W3123241418 @default.
- W3149894872 doi "https://doi.org/10.1016/j.ecoinf.2021.101286" @default.
- W3149894872 hasPublicationYear "2021" @default.
- W3149894872 type Work @default.
- W3149894872 sameAs 3149894872 @default.
- W3149894872 citedByCount "7" @default.
- W3149894872 countsByYear W31498948722021 @default.
- W3149894872 countsByYear W31498948722022 @default.
- W3149894872 crossrefType "journal-article" @default.
- W3149894872 hasAuthorship W3149894872A5002853978 @default.
- W3149894872 hasAuthorship W3149894872A5006497714 @default.
- W3149894872 hasAuthorship W3149894872A5066899248 @default.
- W3149894872 hasAuthorship W3149894872A5084039823 @default.
- W3149894872 hasConcept C108583219 @default.
- W3149894872 hasConcept C11413529 @default.
- W3149894872 hasConcept C115961682 @default.
- W3149894872 hasConcept C119857082 @default.
- W3149894872 hasConcept C150899416 @default.
- W3149894872 hasConcept C153180895 @default.
- W3149894872 hasConcept C154945302 @default.
- W3149894872 hasConcept C41008148 @default.
- W3149894872 hasConcept C91451191 @default.
- W3149894872 hasConcept C99498987 @default.
- W3149894872 hasConceptScore W3149894872C108583219 @default.
- W3149894872 hasConceptScore W3149894872C11413529 @default.
- W3149894872 hasConceptScore W3149894872C115961682 @default.
- W3149894872 hasConceptScore W3149894872C119857082 @default.
- W3149894872 hasConceptScore W3149894872C150899416 @default.
- W3149894872 hasConceptScore W3149894872C153180895 @default.
- W3149894872 hasConceptScore W3149894872C154945302 @default.
- W3149894872 hasConceptScore W3149894872C41008148 @default.
- W3149894872 hasConceptScore W3149894872C91451191 @default.
- W3149894872 hasConceptScore W3149894872C99498987 @default.
- W3149894872 hasFunder F4320327149 @default.
- W3149894872 hasFunder F4320335353 @default.
- W3149894872 hasLocation W31498948721 @default.
- W3149894872 hasOpenAccess W3149894872 @default.
- W3149894872 hasPrimaryLocation W31498948721 @default.
- W3149894872 hasRelatedWork W2889705046 @default.
- W3149894872 hasRelatedWork W2960456850 @default.
- W3149894872 hasRelatedWork W3192840557 @default.
- W3149894872 hasRelatedWork W4223943233 @default.
- W3149894872 hasRelatedWork W4312200629 @default.
- W3149894872 hasRelatedWork W4317565044 @default.
- W3149894872 hasRelatedWork W4360585206 @default.
- W3149894872 hasRelatedWork W4380075502 @default.
- W3149894872 hasRelatedWork W4382286161 @default.
- W3149894872 hasRelatedWork W4386213806 @default.
- W3149894872 hasVolume "62" @default.
- W3149894872 isParatext "false" @default.
- W3149894872 isRetracted "false" @default.
- W3149894872 magId "3149894872" @default.
- W3149894872 workType "article" @default.