Applying transfer learning for marine animal image classification using pre-trained convolutional neural network [thesis] /
Alejado, Alexander S.
Applying transfer learning for marine animal image classification using pre-trained convolutional neural network [thesis] / Alexander S. Alejado. Ellaine T. Gabotero, and Naomi P. Ezaki. - Dumaguete City : Foundation University, 2019. - v, 53 leaves : ill. (col.) 275 cm.
Includes bibliographical references and appendices.
This study aims to classify each image into its class of marine animals and to determine the accuracy pf the image classifier. The image classifier applies Transfer Learning using a pre-trained Convolutional Neural Network. Transfer learning is using existing labeled data gained from solving one problem and then apply it to a different but related problem. In applying transfer learning, the researchers use a pre-trained Convolutional Neural Network. This machine learning algorithm is used because of its benefit in detecting patterns from images in order to recognize objects. In addition, they eliminate the need for manual feature extraction and they can be trained for new recognition tasks which let the researchers enable the use of pre-trained models.
Artificial intelligence.
Thesis Writing 2--Computer Science, Bachelor of Science.--CS 416
LG 221 D35 C66 / A5 C66 .2019 Al252
Applying transfer learning for marine animal image classification using pre-trained convolutional neural network [thesis] / Alexander S. Alejado. Ellaine T. Gabotero, and Naomi P. Ezaki. - Dumaguete City : Foundation University, 2019. - v, 53 leaves : ill. (col.) 275 cm.
Includes bibliographical references and appendices.
This study aims to classify each image into its class of marine animals and to determine the accuracy pf the image classifier. The image classifier applies Transfer Learning using a pre-trained Convolutional Neural Network. Transfer learning is using existing labeled data gained from solving one problem and then apply it to a different but related problem. In applying transfer learning, the researchers use a pre-trained Convolutional Neural Network. This machine learning algorithm is used because of its benefit in detecting patterns from images in order to recognize objects. In addition, they eliminate the need for manual feature extraction and they can be trained for new recognition tasks which let the researchers enable the use of pre-trained models.
Artificial intelligence.
Thesis Writing 2--Computer Science, Bachelor of Science.--CS 416
LG 221 D35 C66 / A5 C66 .2019 Al252