CNN for Image Identification of Hiragana Based on Pattern Recognition using CNN

Abstract

Hiragana is one of the letters in Japanese. In this study, CNN (Convolutional Neural Network) method used as identication method, while he preprocessing used thresholding. Then carry out the normalization stage and the filtering stage to remove noise in the image. At the training stage use maxpooling and danse methods as a liaison in the training process, wherea in testing stage using the Adam Optimizer method. Here, we use 1000 images from 50 hiragana characters with a ratio of 950: 50, 950 as training data and 50 data as testing data. Our experiment yield accuracy in 95%.

Deep-Learning Based Estimation of Dielectrophoretic Force

“…We have adopted a Convolutional Neural Network (CNN) based deep learning approach to estimate the applied DEP voltage from the pearl chain alignment of spherical microparticles. CNN-based deep learning approach has been used in image processing and visual identification [35][36][37][38][39], image super-resolution [40], image segmentation [41][42][43][44], and damage detection [45]. The authors in [44] created a CNN-based image processing method for the detection of bubble patterns in dense bubbly flows using the shadowgraph technique.…”

Tomato Leaf Diseases Classification using Convolutional Neural Networks with Transfer Learning Resnet-50

“…CNN is a pivotal deep learning architecture renowned for its proficiency in image recognition tasks, particularly when confronted with complex visual datasets such as disease-infected tomato leaves [5], [9]. In the context of this study, the utilization of the ResNet-50 model through transfer learning emerges as a strategic approach to harness the pre-trained knowledge embedded within this sophisticated neural network [10]- [12].…”