Doutorado em Engenharia Elétrica e da Computação (EMC)
URI Permanente para esta coleção
Navegar
Navegando Doutorado em Engenharia Elétrica e da Computação (EMC) por Por Orientador "Cruz Júnior, Gélson da"
Agora exibindo 1 - 2 de 2
Resultados por página
Opções de Ordenação
Item Detecção de imagens falsificadas baseada em descritores locais de textura e rede neural convolucional(Universidade Federal de Goiás, 2020-06-30) Ferreira, William Divino; Soares, Fabrizzio Alphonsus Alves de Melo Nunes; http://lattes.cnpq.br/7206645857721831; Cruz Júnior, Gélson da; http://lattes.cnpq.br/4370555454162131; Cruz Júnior, Gélson da; Pedrini, Hélio; Salvini, Rogério Lopes; Costa, Ronaldo Martins da; Lemos, Rodrigo PintoNowadays, digital image transformation has become a widespread activity. Hence, image copying, cloning, and resizing are easily performed, making it challenging to check image integrity and authenticity. Moreover, a criminal investigation from digital images becomes extremely hard, because using those images as proof demands to ensure its legitimately,under a risk to implicate the whole legal process.In this sense, this work develops a model for forged images based on local texture descriptors with convolutional neural networks. Henceforth, in this work, firstly, we evaluated fourteen local texture descriptors in five public image texture datasets, and then we selected descriptors with the best efficacy. Second, the selected descriptors are applied to four public datasets to extract texture features from forged and legit images. Finally, those features are used to train a residual convolutional neural network, and then, classifying images as authentic or forged with a Support Vector Machine Classifier. A result of the proposed model provides enthusiasm, mainly when applied to a dataset with a small number of images.Item Modelo baseado em redes neurais profundas com unidades recorrentes bloqueadas para legendagem de imagens por referências(Universidade Federal de Goiás, 2020-09-28) Nogueira, Tiago do Carmo; Vinhal, Cássio Dener Noronha; http://lattes.cnpq.br/9791117638583664; Cruz Júnior, Gélson da; http://lattes.cnpq.br/4370555454162131; Cruz Júnior, Gélson da; Ferreira, Deller James; Santos, Gilberto Antonio Marcon dos; Vinhal, Cássio Dener Noronha; Lemos, Rodrigo PintoDescribing images using natural language has become a challenging task for computer vision. Image captioning can automatically create descriptions through deep learning architectures that use convolutional neural networks (CNNs) and recurrent neural networks (RNNs). Image captioning has several applications, such as object descriptions in scenes to help blind people walk in unknown environments, and medical image descriptions for early diagnosis of diseases. However, architectures supported by traditional RNNs, in addition to problems of exploding and fading gradients, can generate non-descriptive sentences. To solve these difficulties, this study proposes a model based on the encoder-decoder structure using CNNs to extract the image characteristics and multimodal gated recurrent units (GRU) to generate the descriptions. The part-of-speech (PoS) and the likelihood function are used to generate weights in the GRU. The proposed method performs knowledge transfer in the validation phase using the k-nearest neighbors (kNN) technique. The experimental results in the Flickr30k and MS-COCO data sets demonstrate that the proposed PoS-based model is statistically superior to the leading models. It provides more descriptive captions that are similar to the expected captions, both in the predicted and kNN-selected captions. These results indicate an automatic improvement of the image descriptions, benefitting several applications, such as medical image captioning for early diagnosis of diseases.