Implementasi Metode Convolutional Neural Network Menggunakan Arsitektur LeNet-5 untuk Pengenalan Doodle

Abstract

Recognition of objects to date has been widely applied in various fields, for example in handwritten recognition. This research utilizes the ability of CNN to use LeNet-5 architecture for the introduction of doodle types with 5 object images, namely clothes, pants, chairs, butterflies and bicycles. Each doodle object consists of 30 images with a total dataset of 150 images. The test results show that the first, second and fourth scenarios of bicycle objects are more recognized with an accuracy value of 93% - 98%, recall 86% - 93% and precision 81% - 93%, clothes objects are more recognized in the third scenario with an accuracy value of 94%, 86% recall, and 83% precision.

Deteksi Dini Kanker Payudara Menggunakan Metode Convolution Neural Network (CNN)

“…Sehingga dapat dikatakan bahwa CNN tidak jauh berbeda dengan JST yang lain seperti perceptron maupun backpropagation. Arsitektur jaringan pada CNN secara umum dibagi menjadi dua yaitu feature extraction dan classification (Alwanda et al, 2020;Tobias et al, 2016).…”

Classification of Human Activity Recognition Utilizing Smartphone Data of CNN-LSTM

“…Convolutional layers becomes the first layer consisting of several filters for feature extraction from input data by applying convolution operations to combine information sets [17] [18].…”

“…Pool layer serves to reduce the input spatially (reducing the number of parameters) from the convolution feature so as to reduce the required computational resources to process the data and to accelerate the computing process [17] [18].…”

“…Fully connected layer contains a layer to perform the transformations on the data dimensions for linear classification [18], connecting each neuron in one layer to other neurons in another layerin order to classify inputs into classes, based on the trained dataset [17].…”

Classification of Calligraphy Writing Types Using Convolutional Neural Network Method (CNN)

“…Setiap citra yang digunakan akan melewati beberapa proses seperti convolutional layer, pooling yang digunakan untuk pengekstrasian fitur dari citra, dan proses terakhir adalah dengan fully connected layer untuk menyelesaikan pengklasifikasian. Sehingga algoritma CCN yang digunakan untuk pengolahan citra bisa dibilang cukup efektif [2] [3]. Beberapa penelitian yang terkait dalam pengklasifikasian jenis tulisan kaligrafi yang telah dilakukan seperti Arabic handwriting recognition system using convolutional neural network.…”