Avancerad sökning

Hittade 1 uppsats som matchar ovanstående sökkriterier.

1. 1. Character Recognition in Natural Images Utilising TensorFlow

   Kandidat-uppsats, KTH/Skolan för datavetenskap och kommunikation (CSC)

   Författare :Alexander Viklund; Emma Nimstad; [2017]
   Nyckelord :character recognition; natural images; TensorFlow; data augmentation; neural networks; Chars74K; convolutional; teckenigenkänning; naturliga bilder; TensorFlow; dataaugmentering; neurala nätverk; neuronnät; Chars74K;

   Sammanfattning : Convolutional Neural Networks (CNNs) are commonly used for character recognition. They achieve the lowest error rates for popular datasets such as SVHN and MNIST. Usage of CNN is lacking in research about character classification in natural images regarding the whole English alphabet. LÄS MER