Perbandingan Metode EfficientNetB3 dan MobileNetV2 Untuk Identifikasi Jenis Buah-buahan Menggunakan Fitur Daun

Zaelani, Fiqry; Miftahuddin, Yusup

doi:10.33197/jitter.vol9.iss1.2022.911

Cited by 5 publications

(6 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Arsitektur Efficientnet B3 mempunyai spesifikasi yang terdiri dari 2 convolution layer, 7 mobile bottleneck convolution layer, 1 pooling layer, dan 1 fully connected layer, menghasilkan 12M parameter. Struktur layer Efficientnet B3 pada dasarnya menggunakan struktur layer Efficientnet B0 hanya saja yang berbeda adalah jumlah layer-nya [11].…”

Section: Efficientnetb3unclassified

Penerapan Transfer Learning Efficientnetb3 Untuk Pengenalan Senjata Tradisional Sumatera Barat Menggunakan Convolutional Neural Network (Cnn)

Herlangga

2024

JITET

View full text Add to dashboard Cite

Penelitian ini mengimplementasikan metode Transfer Learning dengan memanfaatkan model EfficientNetB3 untuk mengenali senjata tradisional khas Sumatera Barat. Tiga jenis senjata tradisional, yaitu kerambit, keris pustaka, dan sumpitan, digunakan sebagai dataset dalam pengujian. Dengan menggunakan dataset citra sebanyak 819 untuk pelatihan dan 123 untuk validasi, serta 82 untuk pengujian, model CNN berhasil mencapai akurasi 100%. Evaluasi model menunjukkan nilai sempurna dalam Precision, Recall, dan F1-score untuk setiap kelas. Dengan demikian, model ini dapat dijadikan sebagai alat untuk mengklasifikasikan senjata tradisional Sumatera Barat secara efektif.

show abstract

Section: Efficientnetb3unclassified

Penerapan Transfer Learning Efficientnetb3 Untuk Pengenalan Senjata Tradisional Sumatera Barat Menggunakan Convolutional Neural Network (Cnn)

Herlangga

2024

JITET

View full text Add to dashboard Cite

show abstract

“…The flipping used in this research is a horizontal flip and a vertical flip. Horizontal flip is a form of transformation by reflecting the Y axis, while vertical flip is a form of geometric transformation by reflecting the X axis (Zaelani & Miftahuddin, 2022).…”

Section: ) Flipping Imagementioning

confidence: 99%

Combination of Image Enhancement and U-Net Architecture for Cervical Cell Semantic Segmentation

Rudiansyah,

Iryani,

Kesuma

et al. 2024

JITE

View full text Add to dashboard Cite

Cervical cancer is the second leading cause of death in women and ranks fourth as a disease that occurs in women worldwide. Cervical cancer is a disease that is difficult to detect and can be detected when it is in an advanced stage. This requires early prevention by carrying out a pap-smear examination. Pap-smear examination manually requires a relatively long time, so a tool is needed by segmentation. Segmentation is image processing by performing perfection between the intended object and the background. One of the CNN methods commonly used in medical image segmentation is the U-Net architecture. Segmentation in this study was carried out on the nucleus and cytoplasm of the Herlev dataset using the U-Net architecture combined with data augmentation and image enhancement. In the learning process, this research resulted in a fairly high IoU value of 78% and an RMSE close to 20%. The results of this study also yielded an accuracy value of 89%, with an average precision, recall and F1 score of 89%, 89% and 88.67%, respectively. This shows that the combination of the CNN U-Net architecture with image quality improvement and data augmentation is quite good at segmenting cervical cells for the nucleus and cytoplasm

show abstract

“…MobileNetV2 is one architecture from the usual convolutional neural network methodology also known as CNN stands for which can used For overcome need will be excessive computing resources or need computing tall [13] .…”

Section: Mobilenetv2mentioning

confidence: 99%

Design And Development Of Karo Traditional Music Instrument Recognition Application Based On Digital Image Using Convolutional Neural Network Method

Tarigan,

Achmad Fauzi

2023

j. of artif. intell. and eng. appl.

View full text Add to dashboard Cite

In this modern era, multimedia technology has role important in various field like education, health, and publications. Thesis This focused on development application introduction tool music traditional Karo tribe based digital image use Convolutional Neural Network (CNN) method. Karo music has characteristic typical unique and important for culture of North Sumatra. Study This use method Squeezenet and MobileNetV2 for compare accuracy in introduction object tool music Karo tribe. The Convolutional Neural Network (CNN) method is used in the introduction process. Steps covers convolution, function Activation ReLU, pooling, layers connected full, and the output layer with function Activation Soft max. Image data tool music Karo tribe gathered from Google Images. System This own steps like preprocessing image, model training with CNN method, and analysis results training. Research results This give comparison accuracy between method Squeezenet and MobileNetV2 in recognize tool music Karo tribe. Supporting data used in analysis, especially image tool music obtained Karo tribe from Google Images, with structured design and implementation of CNN, applications This succeed recognize type tool music with level good accuracy.

show abstract

Perbandingan Metode EfficientNetB3 dan MobileNetV2 Untuk Identifikasi Jenis Buah-buahan Menggunakan Fitur Daun

Cited by 5 publications

References 0 publications

Penerapan Transfer Learning Efficientnetb3 Untuk Pengenalan Senjata Tradisional Sumatera Barat Menggunakan Convolutional Neural Network (Cnn)

Penerapan Transfer Learning Efficientnetb3 Untuk Pengenalan Senjata Tradisional Sumatera Barat Menggunakan Convolutional Neural Network (Cnn)

Combination of Image Enhancement and U-Net Architecture for Cervical Cell Semantic Segmentation

Design And Development Of Karo Traditional Music Instrument Recognition Application Based On Digital Image Using Convolutional Neural Network Method

Contact Info

Product

Resources

About