Matches in SemOpenAlex for { <https://semopenalex.org/work/W4285327270> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4285327270 abstract "The demand for seafood has grown rapidly over the last few decades. Given this growth, there is an increasing need for more efficient and sustainable fishing practices. A prime challenge in this regard is monitoring fisheries to prevent harmful fishing practices. Therefore, commercial fishing boats have started carrying electronic monitoring (EM) systems to mitigate these practices. However, as the volume of video footage from EM grows, reviewing the footage manually becomes infeasible. Therefore, we seek to address this limitation by developing a convolutional neural network model to classify images of fish in EM footage of fisheries. To make our model efficient and effective, we utilize transfer learning, wherein pre-trained image classification models are built upon to improve model training speed, and data augmentation, wherein images are modified to create a larger training dataset. For evaluation, we utilize the FishNet Open Image Database with over 85,000 images from EM footage of fisheries in the western and central Pacific Ocean. We train two models to classify 10 species of fish: one without and one with data augmentation. Our results show that using data augmentation improves our model’s ability to deal with more difficult fish image classification scenarios. By using deep learning, we can achieve a high classification accuracy and quickly identify species in EM footage." @default.
- W4285327270 created "2022-07-14" @default.
- W4285327270 creator A5031540690 @default.
- W4285327270 creator A5060766682 @default.
- W4285327270 date "2021-12-01" @default.
- W4285327270 modified "2023-10-16" @default.
- W4285327270 title "Fish Species Classification with Data Augmentation" @default.
- W4285327270 cites W2015918764 @default.
- W4285327270 cites W2520286314 @default.
- W4285327270 cites W2615126350 @default.
- W4285327270 cites W2739651313 @default.
- W4285327270 cites W2805544603 @default.
- W4285327270 cites W2809371442 @default.
- W4285327270 cites W2809598685 @default.
- W4285327270 cites W2945972832 @default.
- W4285327270 cites W2954996726 @default.
- W4285327270 cites W2963246747 @default.
- W4285327270 cites W3003025382 @default.
- W4285327270 cites W3034429256 @default.
- W4285327270 cites W3106306218 @default.
- W4285327270 doi "https://doi.org/10.1109/csci54926.2021.00307" @default.
- W4285327270 hasPublicationYear "2021" @default.
- W4285327270 type Work @default.
- W4285327270 citedByCount "4" @default.
- W4285327270 countsByYear W42853272702022 @default.
- W4285327270 countsByYear W42853272702023 @default.
- W4285327270 crossrefType "proceedings-article" @default.
- W4285327270 hasAuthorship W4285327270A5031540690 @default.
- W4285327270 hasAuthorship W4285327270A5060766682 @default.
- W4285327270 hasConcept C108583219 @default.
- W4285327270 hasConcept C115961682 @default.
- W4285327270 hasConcept C119857082 @default.
- W4285327270 hasConcept C150899416 @default.
- W4285327270 hasConcept C153294291 @default.
- W4285327270 hasConcept C154945302 @default.
- W4285327270 hasConcept C205649164 @default.
- W4285327270 hasConcept C2777211547 @default.
- W4285327270 hasConcept C2909208804 @default.
- W4285327270 hasConcept C41008148 @default.
- W4285327270 hasConcept C505870484 @default.
- W4285327270 hasConcept C514101110 @default.
- W4285327270 hasConcept C75294576 @default.
- W4285327270 hasConcept C81363708 @default.
- W4285327270 hasConcept C86803240 @default.
- W4285327270 hasConceptScore W4285327270C108583219 @default.
- W4285327270 hasConceptScore W4285327270C115961682 @default.
- W4285327270 hasConceptScore W4285327270C119857082 @default.
- W4285327270 hasConceptScore W4285327270C150899416 @default.
- W4285327270 hasConceptScore W4285327270C153294291 @default.
- W4285327270 hasConceptScore W4285327270C154945302 @default.
- W4285327270 hasConceptScore W4285327270C205649164 @default.
- W4285327270 hasConceptScore W4285327270C2777211547 @default.
- W4285327270 hasConceptScore W4285327270C2909208804 @default.
- W4285327270 hasConceptScore W4285327270C41008148 @default.
- W4285327270 hasConceptScore W4285327270C505870484 @default.
- W4285327270 hasConceptScore W4285327270C514101110 @default.
- W4285327270 hasConceptScore W4285327270C75294576 @default.
- W4285327270 hasConceptScore W4285327270C81363708 @default.
- W4285327270 hasConceptScore W4285327270C86803240 @default.
- W4285327270 hasFunder F4320306076 @default.
- W4285327270 hasLocation W42853272701 @default.
- W4285327270 hasOpenAccess W4285327270 @default.
- W4285327270 hasPrimaryLocation W42853272701 @default.
- W4285327270 hasRelatedWork W2996856019 @default.
- W4285327270 hasRelatedWork W3012393889 @default.
- W4285327270 hasRelatedWork W3018421652 @default.
- W4285327270 hasRelatedWork W3021430260 @default.
- W4285327270 hasRelatedWork W3091976719 @default.
- W4285327270 hasRelatedWork W3192840557 @default.
- W4285327270 hasRelatedWork W4220996320 @default.
- W4285327270 hasRelatedWork W4285149559 @default.
- W4285327270 hasRelatedWork W4312200629 @default.
- W4285327270 hasRelatedWork W4382286161 @default.
- W4285327270 isParatext "false" @default.
- W4285327270 isRetracted "false" @default.
- W4285327270 workType "article" @default.