Search

This research classifies acoustic instruments using Convolutional Neural Network (CNN). We utilize a dataset from Kaggle containing audio recordings of piano, violin, drums, and guitar. The training set consists of 700 guitar, percussion, violin, and 528 piano samples. The test set contains 80 samples of each instrument. Features such as Mel spectrograms, MFCCs, and other spectral and non-spectral characteristics are extracted using the Librosa package. Three feature sets—spectral-only, non-spectral-only, and a combined set—are employed to evaluate the efficacy of CNN models. Various CNN configurations are tested by adjusting the number of convolutional filters, learning rates, and epochs. The combined feature set achieves the highest performance, with a validation accuracy of 71.8% and a training accuracy of 76.9%. In comparison, non-spectral features achieve a validation accuracy of 68.4%, and spectral-only features achieve 69.3%. These findings highlight the benefits of using a comprehensive feature set for accurate classification.

Tires are a very important component in a vehicle because they are related to driving safety. Defective tires often cause accidents ranging from minor to fatal accidents. Convolutional Neural Network (CNN) is a type of neural network that is used to detect and recognize objects in an image. CNN can imitate the image recognition system in the human visual cortex, making it suitable for identification and classification of image data. This research aims to develop and evaluate a CNN model that is able to classify vehicle tires as 'defective' or 'good'. Model uses a total of 1856 tire images from kaggle.com and is labeled 'defective' or 'good'. Dataset is split using four different data split ratios (60:40, 70:30, 80:20, and 90:10) to determine the optimal distribution that improves the generalization ability of the model. Model evaluation uses accuracy, precision and recall matrices, which are calculated based on the confusion matrix results from testing on 300 data samples. Research results show that the model achieves the best performance at a split ratio of 80:20, with an accuracy of 76.67%, precision of 77.33%, and recall of 76.32%.

Lung cancer has become one of the most frightening specters in the world of health, leading many people to death each year. Therefore, the classification of lung cancer types is very important to determine the appropriate treatment steps. Considering that lung cancer treatment in the early stages is far more effective and efficient, accurate classification is the key to improving survival rates. This research focuses on the classification of three common lung cancer types: Adenocarcinoma, Large Cell Carcinoma, and Squamous Cell Carcinoma. To achieve optimal results, this study utilizes the ResNet architecture, a deep neural network model that has demonstrated its capabilities in various fields. Before being used on the model, the dataset containing lung X-ray images of patients undergoes preprocessing. At this stage, each image is resized to 256x256 pixels to ensure uniformity and compatibility with the model. Furthermore, this research trains various ResNet models, ranging from ResNet50, ResNet101, to ResNet152, which is the model with the most parameters. By comparing the performance of each model, this study finds that all trained ResNet models are capable of producing good accuracy in classifying lung cancer types. Among these models, ResNet152 demonstrates the most superior performance with an accuracy of 89%. This result suggests that the ResNet architecture has great potential to be used as an aid in classifying lung cancer types with a high level of accuracy. This research makes a significant contribution to the effort to improve the diagnosis and treatment of lung cancer, paving the way for a brighter future for lung cancer patients.

Hallyu or Korean wave refers to the phenomenon of South Korean values and culture spreading to other countries, ultimately influencing global culture. South Korean culture, such as K-pop music, dramas, films, fashion, food, and lifestyle, has gained popularity in Indonesia since 2002. Because South Korean culture influences many aspects of life in Indonesia, responses to this Korean wave are widely discussed in social media, especially through X (Twitter) ranging from positive sentiment to negative sentiment. To gain a more in-depth and detailed understanding of public opinion, a classification process was conducted on the social media platform X (Twitter) using a deep learning algorithm based on the CNN method. The results of this classification provide more accurate and informative insight into the attitudes, opinions, and reactions of the Indonesian people towards the influence of South Korean culture in this country. The research was conducted using 717,998 tweet data resulting in an accuracy of 79%.

In synchronous online lectures, the lecturers often provide the lecture material directly through video conference technology. On the other hand, there are many students who do not pay attention to the lecturers when they are participating in online lectures. As a consequence, in this research, an application was developed to assist lecturers in gathering data regarding the degree to which students who participate in online lectures pay attention to the presented information. The application employed a convolutional neural network (CNN) model to recognize each student's facial expressions and place them into one of two classes: either engaged or disengaged. The captured student facial image was preprocessed to facilitate the classification process. The preprocessing stage consisted of image conversion to gray scale, face detection using the Haar-Cascade Classifier model, and a median filter to reduce noise. In the process of designing a CNN model, three different hyperparameter tuning scenarios were implemented. These tuning scenarios aimed to obtain the best possible CNN model by determining which CNN model hyperparameters were the most optimal. The results of the experiments indicate that the CNN model from the second scenario has the highest level of accuracy in terms of recognizing facial expressions, coming in at 86%. The results of this research have been tested to measure the level of student participation in online lectures. The trial results show that the proposed application can help lecturers evaluate student engagement during online lectures.

Cataract, a condition characterized by clouding of the eye's lens, leads to decreased vision and potentially blindness. In Indonesia, it is the predominant cause of blindness, accounting for 81.2% of cases. Given the rising life expectancy, the incidence of degenerative diseases like cataracts is expected to increase. This research aims to develop a cataract detection system capable of classifying eye images as either indicative of cataracts or normal. Utilizing Convolutional Neural Networks (CNN) and RGB-based image processing—including edge detection techniques such as Canny and Prewitt—the system identifies eye contours. This facilitates image segmentation to ascertain the eye's condition. Therefore, image collection and processing models play a crucial role in this study. The research findings indicate that the system functions effectively, with a 98% success rate in accurately processing normal eye images through the CNN model without detecting cataracts. When tested using grayscale imaging, cataract-afflicted eyes—characterized by red spots in the images—were also successfully identified by the CNN model. These test results demonstrate that the designed cataract detection system can accurately classify images into normal or cataract-afflicted eyes with high precision. This system shows promise for use in early cataract detection, potentially helping to reduce the prevalence of cataract-related blindness in Indonesia.

Mango is a fruit known as the "King of Fruit" due to its rich flavor, vast variability, and high nutritional value. Classifying mangoes based on their external appearance is the initial step in the process of identifying and categorizing mango types conventionally. The classification process can be performed by examining external features such as fruit color, shape, and size. Classifying different types of mango fruits accurately can assist researchers in developing superior varieties and also aid farmers for cultivation purposes, sales, distribution, and selecting the right varieties for local growth and weather conditions. This research conducts the classification of mango types based on color from mango images using machine learning. The study compares three methods, namely Random Forest, Support Vector Machine (SVM), and Convolutional Neural Network (CNN), to determine the best method for classifying mango types based on their images. The dataset underwent preprocessing, where image sizes were standardized to 300 x 300 pixels, and color was changed to grayscale. The dataset was then divided into training and testing data with a ratio of 70:30. Subsequently, the dataset was processed using three methods, and their accuracy results were compared. The findings indicate that the Random Forest method yielded the highest accuracy compared to the other methods, with an accuracy rate of 96%. The accuracy of the SVM method was 95%, and the accuracy of the CNN method was 33%. From these results, it can be concluded that the Random Forest method is highly effective for classifying mango types based on their image compared to SVM and CNN methods.

Penelitian ini mengeksplorasi penggunaan algoritma Grey Wolf Optimizer (GWO) untuk mengoptimalkan parameter pada Convolutional Neural Network (CNN) dalam mengenali ekspresi wajah manusia. Ekspresi wajah adalah aspek penting dalam komunikasi manusia, dan pengenalan ekspresi tersebut menjadi semakin vital dalam interaksi manusia-mesin dan bidang kesehatan psikologi. Metode deep learning, terutama CNN, telah terbukti efektif dalam mengklasifikasikan ekspresi manusia, meskipun masih menghadapi beberapa tantangan, seperti pengaturan parameter yang rumit dan kebutuhan akan data yang besar. Penelitian ini bertujuan untuk mencari parameter optimal untuk meningkatkan kinerja CNN dalam mengenali ekspresi wajah menggunakan algoritma GWO. Data yang digunakan adalah dataset Facial Expression Recognition 2013 (FER-2013), dengan total 600 citra wajah yang dibagi menjadi tiga kelas: happy, sad, dan angry. Pendekatan yang diusulkan mencakup preprocessing data, pencarian parameter arsitektur CNN menggunakan GWO, pembuatan model CNN, dan pengujian model menggunakan data testing. Hasil pengujian menunjukkan bahwa dengan parameter optimal, model CNN mencapai akurasi yang baik, dengan nilai akurasi 79% pada data training, 60% pada data validation, dan rata-rata akurasi 77% pada data testing. Penelitian ini menyoroti pentingnya penanganan yang cermat dalam menentukan parameter untuk memastikan hasil yang optimal dalam pengenalan ekspresi wajah manusia menggunakan CNN.

Pneumonia is an inflammation of the lung parenkime caused by bacteria, viruses, and other infections. Generally, the detection of pneumonia can be done by analyzing the x-ray image of the thorax after the reported symptoms and recommendations given by the physician. Previous studies said that Convolutional Neural Networks (CNN), part of Deep learning technology, can be used to analyze x-ray images. It also reported that this method could reduce independent parameters and handle the deformation of input images, such as translation, rotation, and scale. In this study, we report the implementation of CNN model VGG16 with varying epochs and pixels on chest x-ray images to classify pneumonia. The data used are 4000 x-ray images of the thorax taken from the Mendeley website. Final classification processes were done by using the softmax activation function. The results were tested using 20 batch sizes based on 2 image treatment parameters, namely resize and epoch. We reported that the higher image reduction size can increase the average calculation’s accuracy. It is found that the highest accuracy (87,54%) is obtained from the resizing of 300×300 pixels. The lowest average accuracy, 79.61%, is shown at a resize size of 100×100 pixels. The highest accuracy (94.39%) for the epoch variant on the resized image of 300×300 pixels is obtained on the 30th epoch

This study addresses the critical challenge of Diabetic Retinopathy (DR) detection and severity grading, aiming to advance the field of medical image analysis. The research problem focuses on the need for an accurate and efficient model to discern DR conditions, thereby facilitating early diagnosis and intervention. Employing a Convolutional Neural Network (CNN), our methodology is developed to strike a balance between precision and computational efficiency, a pivotal aspect in the context of healthcare applications. The research leverages the APTOS 2019 dataset, a comprehensive collection of fundus photographs, to evaluate the efficacy of our proposed model. The dataset allows for a thorough investigation into the model's performance in binary-class and multi-class classifications, providing a robust foundation for analysis. The most important result of our study manifests in the achieved accuracy rates of 98.67% and 87.81% for binary-class and multi-class classifications, respectively. These outcomes underscore the model's reliability and innovation, surpassing established machine learning algorithms and affirming its potential as a valuable tool for early DR detection and severity assessment. In conclusion, the study marks a significant advancement in leveraging deep learning for ophthalmic diagnoses, particularly in the nuanced landscape of DR. The implications of our findings extend to the broader realm of AI-driven healthcare solutions, presenting opportunities for enhanced clinical practices and early intervention strategies. Future research endeavors could explore further refinements to the model, considering additional datasets and collaborating with healthcare professionals for real-world validation, ensuring the continued progress of AI applications in the medical domain.

Kemajuan teknologi yang telah berkembang sangat pesat dapat membantu pekerjaan manusia dalam berbagai bidang. Salah satu kegiatan yang dapat diotomatisasi adalah aktivitas pengawasan kegiatan manusia dalam sebuah ruangan, misalnya bagi lansia ataupun orang cacat. Umumnya kegiatan pengawasan membutuhkan tenaga manusia untuk memantau kejadian dan aktivitas di lokasi tertentu secara live maupun direkam melalui kamera CCTV. Dengan perkembangan teknologi, seluruh kegiatan pengenalan terhadap aktivitas manusia dapat dilakukan secara otomatis. Sistem dapat mengenali aktivitas duduk, berdiri, belajar, mengangkat tangan, serta bertepuk tangan. Sistem dibuat dengan menggunakan metode Convolutional Neural Network (CNN). Eksperimen akan dilakukan dengan data uji yang berasal dari internet maupun rekaman video yang dilakukan oleh tim di lapangan. Hasil eksperimen menunjukkan bahwa sistem yang dikembangkan dapat mengenali aktivitas manusia dengan nilai akurasi sebesar 86,65% untuk data uji dari internet dan 96% untuk data uji dari IP camera yang terpasang pada sebuah lokasi. Selain itu, sistem juga dapat menghasilkan sebuah rangkuman catatan aktivitas terhadap seluruh kegiatan yang terjadi di dalam lokasi tersebut pada kisaran waktu yang ditetapkan.

Copyright of Sudo Jurnal Teknik Informatika is the property of Ilmu Bersama Center and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Techno.com is the property of Universitas Dian Nuswantoro, Fakultas Ilmu Komputer and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Jurnal Ilmu Komputer dan Agri-Informatika is the property of IPB University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Penelitian ini dalam bentuk Pengenalan jenis bunga menggunakan Convolutional Neural Network (CNN)untuk mengoptimalkan identifikasi jenis bunga berdasarkan spektrum warna. Spektrum warna bunga dapat bervariasi secara signifikan antar spesies maupun dalam satu spesies. Ini bisa menjadi tantangan dalam pengembangan model yang mampu mengidentifikasi jenis bunga dengan akurasi tinggi di tengah variasi spektrum warna yang luas. Memilih arsitektur CNN yang sesuai dan mengoptimalkan hyperparameter model untuk mencapai kinerja yang optimal adalah proses yang kompleks. Perlu dilakukan eksplorasi yang cermat terhadap berbagai arsitektur dan teknik optimasi untuk meningkatkan akurasi identifikasi jenis bunga. Dataset yang digunakan dikumpulkan dari berbagai sumber repository. dataset yang mencakup gambar-gambar bunga diambil dalam berbagai kondisi pencahayaan yang berbeda, mewakili spektrum warna yang beragam. Pada penelitian ini melalui tahapan pra-pemrosesan data termasuk normalisasi spektrum warna, ekstraksi fitur, dan augmentasi data untuk meningkatkan keragaman dataset. Model CNN pada penelitian ini dioptimalkan melalui optimasi arsitektur jaringan. Evaluasi model dilakukan menggunakan metrik evaluasi kinerja standar seperti akurasi, presisi, dan recall. Dari haril pengujian dataset sebanyak 551 gambar (5 label data) dengan gambar yang distandarisasi untuk ukuran 150 x 150 px dengan iterasi (pouch) sebanyak 50 mendapakan hasil akurasi sebesar 96% dan hasil loss sebesar 0.05%. Dapat diketahui bahwa detail dari suatu gambar yang digunakan dalam penelitian sebagai dataset sangat mempengaruhi nilai akurasi model data. Penentuan epoch dan jumlah convolutional layer sangat mempengaruhi akurasi yang akan didapatkan karena gambar akan melalui pemeriksaan berulang sebanyak jumlah epoch sehingga detail gambar dapat terdeteksi dengan baik. Metode CNN yang terdiri dari proses yang berlapis-lapis menunjukkan bahwa penggunaan model ini pada klasifikasi jenis bunga berdasarkan warna memiliki tingkat akurasi yang baik.

Skin cancer is one of the most worrying diseases for humans. In Indonesia alone, skin cancer occupies the third position after cervical cancer and breast cancer. Currently, doctors still use the biopsy method to diagnose skin cancer. It is less effective because this method requires the performance of an experienced doctor, takes a long time, and is a painful process. Because of that, we need a way in which skin cancer can be classified using dermoscopic images to help doctors diagnose skin cancer earlier. Researchers proposed to classify skin cancer into seven classes, namely actinic keratoses, basal cell carcinoma, benign keratosis-like lesions, dermatofibroma, melanoma, melanocytic nevus, and vascular lesions. The method used in this study is a convolutional neural network (CNN) with the MobileNet V2 architecture. The dataset used is the HAM10000 dataset, with a total of 10015 images. In this study, a comparison was made between data augmentation, learning rate, epochs, and different amounts of data. Based on the test results, the highest accuracy results were obtained, namely 79%. The best model is implemented into a mobile application.

Studi penelitian ini dilaksanakan untuk mengembangkan sistem deteksi fonetik dialek Kambera dari bahasa lokal Sumba Timur berbasis framework TensorFlow yang akan diimplementasikan pada aplikasi mobile. Sebagai bagian dari inisiatif ini, penelitian ini telah mengompilasikan sebuah dataset representatif dari sampel fonetik dialek Kambera. Tujuan utamanya adalah untuk meningkatkan presisi dalam pengenalan fonetik. Menggunakan dialek Kambera sebagai studi kasus, data diekstraksi dan dilatih menggunakan layanan Teachable Machine berbasis open-source. Penelitian ini mengadopsi pendekatan berbasis convolutional neural network (CNN) yang dikombinasikan dengan metode Mel frequency cepstral coefficients (MFCC) untuk ekstraksi fitur yang lebih akurat. Setelah proses pengumpulan data, pelatihan model, pengujian, dan implementasi, model diintegrasikan ke dalam platform Android untuk kepentingan publik yang ingin memahami bahasa dialek Kambera dari Sumba Timur. Pengembangan dan pengujian sistem ini dirancang untuk mendeteksi dan menginterpretasikan fonetik bahasa lokal Sumba Timur dengan dialek Kambera, memberikan kontribusi signifikan dalam optimalisasi pengenalan fonetik, serta menyediakan dataset untuk kepentingan penelitian berkelanjutan. Sistem ini juga berfungsi sebagai alat edukasi linguistik yang dapat diakses dan mendukung inklusi serta diversifikasi linguistik dalam teknologi digital. Evaluasi empiris menunjukkan bahwa tingkat presisi deteksi dialek rata-rata secara keseluruhan mencapai 98,3% hingga 99,6%, dengan tingkat kepuasan pengujian kepada pengguna mencapai 99,33%. Hasil ini menegaskan bahwa sistem yang dikembangkan memiliki kapabilitas deteksi yang sangat efisien dan baik.

The fashion industry has changed a lot over the years, which makes it hard for people to compare different kinds of fashion. To make it easier, different styles of clothing are tried out to find the exact and precise look desired. So, we opted to employ the Convolutional Neural Network (CNN) method for fashion classification. This approach represents one of the methodologies employed to utilize computers for the purpose of recognizing and categorizing items. The goal of this research is to see how well the Convolutional Neural Network method classifies the Fashion-MNIST dataset compared to other methods, models, and classification processes used in previous research. The information in this dataset is about different types of clothes and accessories. These items are divided into 10 categories, which include ankle boots, bags, coats, dresses, pullovers, sandals, shirts, sneakers, t-shirts, and trousers. The new classification method worked better than before on the test dataset. It had an accuracy value of 95. 92%, which is higher than in previous research. This research also uses a method called image data generator to make the Fashion MNIST image better. This method helps prevent too much focus on certain details and makes the results more accurate.

Copyright of Sudo Jurnal Teknik Informatika is the property of Ilmu Bersama Center and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The recognition and classification of rice plant diseases require an accurate system to generate classification data. Types of rice diseases can be identified in several ways, one of which is leaf characterization. One method that has high accuracy in identifying plant disease types is Convolutional Neural Networks (CNN). However, the rice disease data used has unbalanced data which affects the performance of the method. Therefore, the purpose of this research was to apply data augmentation to handle unbalanced rice disease data to improve the performance of the Convolutional Neural Network (CNN) method for rice disease type detection based on leaf images. The method used in this research is the CNN method for detecting rice disease types based on leaf images. The result of this research was the CNN method with 100 epochs able to produce an accuracy of 99.7% in detecting rice diseases based on leaf images with a division of 80% training data (2438 data) and 20% testing data (608 data). The conclusion is that the CNN method with the augmentation process can be used in rice disease detection because it has very high accuracy.

Fresh fruits and vegetables contain many nutrients, such as minerals, vitamins, antioxidants, and beneficial fiber, superior to those found in rotten or almost rotten produce. On the other hand, fruits and vegetables that are nearly spoiled or already rotten have significantly lost their nutritional value. Rotten produce also harbors bacteria and fungi that can lead to infections and food poisoning when consumed. Convolutional Neural Network (CNN) offers a programmable solution for classifying fresh and rotten fruits and vegetables. Image processing using the TensorFlow library is employed in this classification process. During testing on the training data, the CNN achieved an accuracy of 90.42%. In comparison, the validation accuracy reached 94.21% when using the SGD optimizer, 20 epochs, a batch size 16, and a learning rate of 0.01. For the testing data, the accuracy obtained was 80.83%.

Indonesia is a tropical country with a diverse range of plants that ancient people used for traditional medicines. However, the similarity in shape of the leaves became an obstacle to distinguishing them. Therefore, technological advancements are expected to help identify the herbal leaves to use them right on target according to their efficacy. In this research, image classification of katuk (Sauropus Androgynus) and kelor (Moringa Oleifera) leaves is applied using 3 different algorithms i.e hybrid of Gray Level Co-Occurrence Matrix (GLCM) feature extraction and Support Vector Machine (SVM) implementing 4 kernels namely linear, RBF, polynomial, and sigmoid; hybrid of GLCM and Convolutional Neural Network (CNN); and pure CNN. A dataset of 480 images has been collected with 2 different scenarios, including bright and dark intensities. Based on the result, a hybrid of GLCM and SVM showed the highest accuracy of 96% in the dark intensity test using a linear kernel, while sigmoid obtained the lowest accuracy of 35%. On the other hand, it has been discovered that CNN obtained the highest performance in the bright intensity test with an accuracy of 98%. While in the dark intensity test, a hybrid of GLCM and CNN is superior, obtaining 96% accuracy. In conclusion, CNN is more powerful for image classification with bright intensity. For dark intensity images, both the hybrid of GLCM+SVM (linear) and the hybrid of GLCM+CNN are fairly recommended.

Concerns have been raised about the social consequences of fake news as it has spread rapidly on online platforms. It is critical to detect and mitigate the spread of fake news in order to maintain a healthy community conversation. There is a need to put more effort into the identification of fake news as more people use the internet, especially as more internet-enabled gadgets become more widely available and inexpensive. With the help of two Neural Network techniques: long-short-term memory (LSTM) and Convolutional Neural Network (CNN). This research proposes novel deep-learning methods for identifying fake news using two datasets. These methods were considered for this research because they had proven to be successful in earlier studies that had been looked at. Finding the best-performing optimal models is the goal of this study. HyperOpt Technique was used for Neural Network model. The performance of the optimized models was compared with the performance of the models without optimization. The results obtained showed that for both datasets, CNN and LSTM performed better when training the models with the optimal values with an average difference of 12.7% for Accuracy, 11.9% for Precision, 12.3% for Recall and 15.4% for F1-Score.

Komoditas jagung di Indonesia menjadi tanaman pangan terbesar kedua setelah padi sebagai sumber karbohidrat. Namun dikarenakan keterbatasan kemampuan petani dan faktor lingkungan menyebabkan upaya penanganan tanaman jagung akibat adanya serangan organisme pengganggu tanaman menjadi terhambat. Penelitian ini mengusulkan upaya deteksi secara dini terhadap jenis penyakit pada daun tanaman jagung menggunakan metode Convolutional Neural Network (CNN) yang dikenal sebagai algoritma pembelajaran mesin berkinerja tinggi dalam mengklasifikasikan jenis penyakit tanaman ke dalam beberapa kelas seperti Blight, Common Rust, Grey Leaf Spot, dan Healthy. Selain itu, transformasi warna citra dari RGB, HSV dan Grayscale, proses segmentasi dengan Region of Interest (ROI) serta dilengkapi dengan penerapan ektraksi fitur tekstur dengan menggunakan GLCM telah mampu menghasilkan tingkat akurasi sebesar 94% dan nilai loss rate yang relatif kecil yaitu 0.1742. Hasil penelitian ini menunjukkan bahwa penggunaan metode CNN terbukti secara efisien & efektif dalam melakukan identifikasi jenis penyakit tanaman.

The problem faced in identifying the quality of black soybeans is that the quality of the assessment is inconsistent and it takes a relatively long time. This study aims to determine the best convolutional neural network architecture by comparing the performance of Custom CNN, MobileNetV2, and ResNet-34 architectures in identifying the quality level (grade) of black soybeans. The quality of black soybean is split into 4 different classes based on physical characteristics (split, damaged, other colors, wrinkles, dirt) and moisture content test. The number of images used is 1300 images, with the ratio of training data, validation data, and testing data are 50:25:25, 60:25:15, and 70:20:10. The best model for identifying the quality based on the physical characteristics is the MobileNetV2 architecture with a ratio of 50:25:25 which produces an accuracy of 90.18%. Morover, the best model for identifying the quality based on the moisture content is the ResNet-34 architecture with a ratio of 70:20:10, which produces an accuracy of 78.12%. The best overall accuracy in identifying the quality based on both physical characteristics and moisture content is the ResNet-34 architecture, with a ratio of 70:20:10, with an average accuracy of testing data of 79.21%.

Bahasa isyarat merupakan hal yang sangat penting bagi suatu kelompok masyarakat, yaitu masyarakat bisu atau tuli. Untuk dapat berkomunikasi dengan masyarakat bisu atau tuli, orang yang tidak bisu atau tuli memerlukan bahasa isyarat tersebut untuk dapat mengerti maksud atau pikiran mereka yang bisu atau tuli. Sebagian besar percakapan pada bahasa isyarat dilakukan dengan menggunakan tangan, dimana tangan beserta jari-jarinya digunakan untuk membentuk pose atau bentuk yang unik, sehingga dapat dikenali sebagai maksud tertentu. Penulis mengusulkan dikembangkan sistem rekognisi citra digital untuk dapat mengenali bahasa isyarat tersebut. Dengan menggunakan metode Convolutional Neural Network (CNN) yang merupakan bagian dari Deep Learning atau Machine Learning, sistem akan mengenali pose atau bentuk dari citra bahasa isyarat yang dimasukkan, dan memberikan luaran yang sesuai dengan maksud dari pose atau bentuk dari citra bahasa isyarat tersebut. Penelitian ini dimulai dengan pengumpulan data, baik data sekunder dari internet maupun data pribadi yang diambil secara manual. Data kemudian melalui pemrosesan awal dan diklasifikasikan dengan CNN, lalu didapatkan hasil untuk dianalisis. Apabila hasil memuaskan, model akan diekspor untuk dimasukkan ke dalam aplikasi berbasis web untuk digunakan secara real-time. Berdasarkan hasil pengujian, model yang terbaik untuk arsitektur adalah model EfficientNet B4 dengan menggunakan Hyperparameter optimizer Adam dan learning rate 0.001 beserta scheduler. Digunakan pretrained weights untuk meningkatkan akurasi tersebut, dan ditambahkan Spatial transformer untuk mencoba membuat model menjadi lebih kokoh. Ditambah dengan pretrained weights, model diekspor untuk digunakan secara real-time. Hasil pengujian real-time menunjukkan bahwa model mampu mendeteksi setidaknya 23 dari 26 alfabet pada latar belakang yang abstrak. Apabila diuji pada latar belakang polos seperti hitam atau putih, model mampu mendeteksi seluruh 26 alfabet dengan probabilitas yang hampir sempurna. Hal ini menunjukkan bahwa metode yang digunakan sudah mampu mengatasi masalah yang disampaikan. Abstract Sign language is very important for a group of people, namely the deaf or dumb. To be able to communicate with people who are mute or deaf, people who are not mute or deaf require sign language to be able to understand the intentions or thoughts of those who are mute or deaf. Most conversations in sign language are carried out using the hands, where the hands and their fingers are used to form unique poses or shapes, so that they can be recognized as having certain meanings. The author proposes to develop a digital image recognition system to be able to recognize sign language. By using the Convolutional Neural Network (CNN) method which is part of Deep Learning or Machine Learning, the system will recognize the pose or shape of the entered sign language image, and provide output that matches the meaning of the pose or shape of the sign language image. This research began with data collection, both secondary data from the internet and personal data taken manually. The data then goes through initial processing and is classified with CNN, then results are obtained for analysis. If the results are satisfactory, the model will be exported to be included in a web-based application for use in real-time. Based on the test results, the best model for the architecture is the EfficientNet B4 model with the Hyperparameter consisting of optimizer Adam and learning rate 0.001 along with the scheduler. Pretrained weights were used to improve accuracy, and Spatial transformers were added to try to make the model more robust. Coupled with pretrained weights, the model is exported for use in real-time. Real-time test results show that the model is able to detect at least 23 of the 26 alphabets on an abstract background. When tested on a plain background such as black or white, the model was able to detect all 26 alphabets with almost perfect probability. This shows that the method used is able to overcome the problem presented.

Copyright of Techno.com is the property of Universitas Dian Nuswantoro, Fakultas Ilmu Komputer and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Copyright of Techno.com is the property of Universitas Dian Nuswantoro, Fakultas Ilmu Komputer and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Penyakit kulit merupakan jenis penyakit yang sering dialami oleh makhluk hidup terutama kucing. Kucing salah satu hewan yang paling banyak dipelihara oleh manusia. Berdasarkan hasil survei berupa kuesoner yang disebarkan ke orang-orang yang memiliki kucing atau penyayang kucing, terdapat sekitar 65% dari 20 responden mengatakan kalau kucing peliharaannya pernah mengalami penyakit kulit dan sekitar 65% dari 20 responses mengatakan kesulitan mengenali penyakit kulit pada kucing. Pemilik atau pecinta kucing melakukan identifikasi penyakit kulit pada kucing dengan cara manual menggunakan indra penglihatan yaitu mata. Karena gejala penyakit kulit pada kucing memiliki kesamaan antar jenis penyakit, hal itu membuat pemilik atau pecinta kucing ragu atau salah mengenali jenis penyakit kulit pada kucing. Hal ini menyebabkan pecinta kucing salah memberikan pertolongan pertama atau pengobatan. Kondisi ini bisa membuat kucing bukan menjadi lebih sehat melainkan bisa bertambah parah. Solusi untuk mengatasi permasalahan tersebut untuk mengidentifikasi penyakit kulit pada kucing, penulis akan mengimplementasikan Deep Learning dengan menggunakan algoritma Convolutional Neural Network (CNN) untuk melakukan deteksi penyakit kulit pada kucing. Untuk mempermudah dalam pemakaiannya, dibuatlah dalam bentuk aplikasi android. Berdasarkan hasil uji confusion matrix didapatkan akurasi sebesar 78,33%, presisi sebesar 76,96%, dan recall sebesar 74,94%.

Copyright of Jurnal Ilmu Komputer dan Agri-Informatika is the property of IPB University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Counterfeiting has been one of the crimes of the 21st century. One of the methods to overcome product counterfeiting is a copy detection pattern (CDP) stamped on the product. CDP is a copy-sensitive pattern that leads to quality degradation of the pattern after the print and scan process. The amount of information loss is used to distinguish between original and fake CDPs. This paper proposed a CDP estimation model based on the convolutional neural network (CNN), namely, CDP-CNN. The CDP-CNN addresses the spatial dependency of the image patch. Thus, it should be better than the state-of-the-art model that uses a multi-layer perceptron (MLP) architecture. The proposed model had an estimation bit error rate (BER) of 9.91% on the batch estimation method. The error rate was 9% lower than the previous method that used an autoencoder MLP model. The proposed model also had a lower number of parameters compared to the previous method. The effect of preprocessing, namely the use of an unsharp mask, was tested using a statistical testing method. The effect of preprocessing had no significant difference except in the batch estimation scheme where the unsharp mask filter reduced the error rate by at least 0.5%. In addition, the proposed model was also used for the authentication method. The authentication using the estimation model had a good separation distribution to distinguish the fake and original CDPs. Thus, the CDP can still be used as the authentication method with reliable performance. It helps anti-counterfeiting on product distribution and reduces negative impacts on various sectors of the economy.

Buah Tin merupakan buah yang berasal dari Timur Tengah. Namun ada beberapa varietas buah Tin yang dapat ditemukan di Indonesia, seperti Buah Tin Green Jordan Fig. Banyak masyarakat menyukai buah tini karena rasanya yang manis dan memiliki banyak khasiat untuk digunakan sebagai obat. Klasifikasi kematangan buah Tin pada umumnya menggunakan penglihatan manusia sehingga dapat menyebabkan kesalahan dikarenakan ciri kematangan buahnya tidak jauh berbeda saat matang dan belum matang. Ketika matang berwarna hijau muda agak kekuningan, sedangkan saat belum matang berwarna hijau. Oleh karena itu didapatkan ide untuk membuat aplikasi yang dapat melakukan klasifikasi kematangan buah Tin melalui gambar berbasis Android menggunakan algoritma Convolutional Neural Network. Pada penelitian ini akurasi yang dicapai sebesar 94%.

Masalah Organisme Pengganggu Tumbuhan (OPT) terutama berkaitan dengan penyakit selalu menjadi isu dalam pengelolaan perkebunan kelapa sawit. Kelapa sawit memiliki penyakit yang disebabkan oleh hama dan lainnya yang dapat mempengaruhi pertumbuhan serta proses berbuahnya.Untuk itu, penelitian ini bertujuan untuk mengidentifikasi sehat dan tidaknya tanaman kelapa sawit melalui warna daunnya, sehingga akan memudahkan kinerja petani. Deep Learning (DL) merupakan bidang ilmu dari machine learning dengan melakukan pembelajaran lebih dalam untuk banyak lapisan. Convolutional Neural Network (CNN) adalah salah satu algoritma DL yang dirancang untuk mengolah data dalam bentuk dua dimensi misalnya gambar. Oleh karena itu, pada penelitian ini akan diterapkan metode CNN untuk mengklasifikasikan sehat atau tdiaknya tanaman kelapa sawit berdasarkan warna daunnya. Data yang digunakan berjumlah 3000 data dengan skenario pengujian untuk data training dan data testingnya adalah 90%:10%, 80%:20%, 70%:30% dan 65%:35%. Berdasarkan dari 4 skenario pengujian tersebut akurasi terbaik didapatkan adalah 99.90% untuk skenario 65% data training dan 35% data testing. Sedangkan tingkat akurasi yang paling rendah adalah 99,50% untuk skenario 90% data training dan 10% data testing.

Traditional Indonesian music is an Indonesian cultural heritage that is often forgotten by modern society. Many people do not know which area the traditional music came from. This is a problem because of the large amount of traditional music that loses its identity. Deep Learning technology can be a solution to this traditional music classification problem. The topic of traditional music classification was chosen because there has been no research using this topic before. This research will classify traditional music based on the area of origin using data from Youtube with the extraction method of the Mel-Frequency Cepstral Coefficients (MFCC) feature and the Convolutional Neural Network (CNN) classification model. There are 7 provinces that will be used as classification labels, namely Riau, Papua, Special Capital District of Jakarta, Special Region of Yogyakarta , North Sumatra, West Java, and South Sulawesi. The classification system produced in this study produced good classification accuracy with a value of 74.03%.

Penelitian ini bertujuan untuk mendeteksi ekspresi wajah pada pengendara mobil. Ekspresi yang dapat terdeteksi dalam penelitian ini terdapat tiga ekspresi yaitu, ekspresi sedih, marah, dan bahagia. Pada penelitian ini merancang sistem pengenalan ekspresi wajah menggunakan algoritma Convolutional Neural Network (CNN). Output yang dihasilkan dalam alat ini yaitu berupa suara yang keluar dari speaker sebagai peringatan kepada pengendara agar lebih berhati-hati saat berkendara dalam keadaan sedih, marah, dan bahagia.alat komponen yang digunakan dalam alat ini yaitu, Raspberry pi 4, webcam, dan speaker. Software sendiri mengunakan bahasa pemrograman python yang langsung di program pada raspberry Pi sebagai controller utama. Pemrograman dilakukan melalui laptop menggunakan software remote VNC berbasis IP dan local network. Hasil penelitian menunjukkan bahwa alat pendeteksi ekspresi wajah pada pengendara ini sangat efektif untuk di gunakan di dalam mobil saat berkendara karena dengan alat ini pengendara bisa lebih waspada dan berhati-hati dalam berkendara ketika sedang sedih, marah dan bahagia. Sehingga alat ini dapat mengurangi resiko kecelakaan lalu lintas yang sering terjadi akibat suasana hati yang sedang tidak terkendali.

The image of chronic wounds on human skin tissue has the similar look in shape, color and size to each other even though they are caused by different diseases. Diabetic ulcer is a condition where peripheral arterial blood vessels are disrupted due to hyperglycemia in people with diabetes mellitus. This research was aimed to analyze the accuracy of the Convolutional Neural Network algorithm in classifying diabetic ulcer disease with a transfer learning approach based on the appearance of the image of the wound on the sole in people with diabetes mellitus. By applying the transfer learning approach, the results showed that the Resnet152V2 model achieved the best accuracy value of 0.993 (99%), precision of 1.00, recall of 0.986, F1-Score of 0.993 and Support of 72. Therefore, the ResNet152V2 model was highly considered for classifying diabetic ulcer in patients with diabetes melitus.

Cough is one of the most common symptoms of diseases, especially respiratory diseases. Quick cough detection can be the key to the current pandemic of COVID-19. Good cough recognition is the one that uses non-intrusive tools such as a mobile phone microphone that does not disable human activities like stick sensors. To do sound-only detection, Deep Learning current best method Convolutional Neural Network (CNN) is used. However, CNN needs image input while sound input differs (one dimension rather than two). An extra process is needed, converting sound data to image data using a spectrogram. When building a spectrogram, there is a question about the best size. This research will compare the spectrogram's size, called Spectrogram Window, by the performance. The result is that windows with 4 seconds have the highest F1-score performance at 92.9%. Therefore, a window of around 4 seconds will perform better for sound recognition problems.

Copyright of Jurnal Ilmu Komputer dan Agri-Informatika is the property of IPB University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Tool wear is one of the cost drivers in the manufacturing industry because it directly affects the quality of the manufactured workpiece and production efficiency. Identifying the right time to replace the cutting tool is a challenge. If the tool is replaced too soon, the production time can be disrupted, causing unscheduled downtime. Conversely, if it is replaced too late, there will be an additional cost to replace raw materials damaged by broken tools. Therefore, researchers continue to develop tool condition monitoring (TCM) methods to analyze tool wear. A recent popular method is machine vision with convolutional neural networks (CNN). The present research aims to develop classification models that can categorize the image data of milling and turning inserts into GO (suitable for use) and NO GO (not suitable for use). Two approaches are selected for the modeling process, custom learning and transfer learning, with image data input from smartphones and microscope cameras. The experimental results show that the best model is the transfer learning approach using Inception-V3 architecture with a smartphone image. The model reaches 92.2% accuracy, hence demonstrating a relatively good performance in determining whether the tool is suitable for use or not.

Aksara Sunda becomes one of the cultures of sundanese land that needs to be preserved. Currently, not all people know Aksara Sunda because of the shift in cultural values and there is a presumption that Aksara Sunda is difficult to learn because it has a unique and complicated shape. The use of deep learning has been widely used, especially in the field of computer vision to classify images, one of the commonly used algorithms is the Convolutional Neural Network (CNN). The application of The Convolutional Neural Network (CNN) algorithm on sundanese handwriting imagery can make it easier for people to learn Sundanese script, this study aims to find out how accurate the neural network convolutional algorithm is in classifying Aksara Sunda imagery. Data collection techniques are done by distributing questionnaires to respondents. System testing using accuracy tests, testing on CNN models using data testing get 97.5% accuracy and model testing using applications get 98% accuracy. So based on the results of the trial, the implementation of deep learning methods using neural network convolution algorithms was able to classify the handwriting image of Aksara Sunda well.

Incung script is a legacy of the Kerinci tribe located in Kerinci Regency, Jambi Province. On October 17, 2014, the Incung script was designated by the Ministry of Education and Culture as an intangible heritage property owned by Jambi Province. But in reality, the Incung script is almost extinct in society. This study aims to identify the characters of the Incung (Kerinci) script with the output in the form of Latin characters from the Incung script. The classification method used is the Convolutional Neural Network (CNN) method. The dataset used as many as 1400 incung character images divided into 28 classes. In this study, an experiment was conducted to obtain the most optimal model. Showing the results using the CNN method during the training process that the accuracy of the training data reaches 99% and the accuracy of the testing data reaches 91% by using the optimal hyperparameters from the tests that have been done, namely batch size 32, epoch 100, and Adam's optimizer. It evaluates the CNN model using 80 images in words (a combination of several characters) with 4 test scenarios. It shows that the model can recognize image data from scanning printed books, digital writing test data, test data with images containing more than two characters, and check images with different font sizes

Kepuasan pelanggan menunjukan seberapa baik produk atau layanan suatu organisasi secara keseluruhan dalam memenuhi harapan pelanggan. Ekspresi wajah pelanggan dapat menunjukan kepuasan mereka terhadap layanan yang diberikan. Convolution Neural Network (CNN) adalah jenis algoritma neural network yang dapat digunakan untuk mengenali suatu objek di dalam sebuah gambar. CNN memanfaatkan proses konvolusi untuk menentukan dan membedakan suatu objek di dalam gambar dengan objek lainnya seperti untuk mengenali berbagai ekspresi wajah. Penelitian ini bertujuan untuk mengukur kepuasan pelanggan dengan memanfaatkan model CNN dengan mengenali setiap perubahan ekspresi wajah. Dari hasil pelatihan model CNN diperoleh akurasi sebesar 90,57%. Selanjutnya model yang terbentuk diimplementasikan ke dalam sebuah sistem berbasis web yang melakukan perekaman ekspresi wajah dan melakukan klasifikasi (puas atau tidak puas) terhadap setiap perubahan wajah yang terdeteksi. Ekspresi yang paling dominan merupakan hasil dari pengukuran kepuasan pelanggan.

Copyright of Jurnal Ilmu dan Teknologi Kelautan Tropis is the property of IPB University and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Not all mushrooms are edible because some are poisonous. The edible or poisonous mushrooms can be identified by paying attention to the morphological characteristics of mushrooms, such as shape, color, and texture. There is an issue: some poisonous mushrooms have morphological features that are very similar to edible mushrooms. It can lead to the misidentification of mushrooms. This work aims to recognize edible or poisonous mushrooms using a Deep Learning approach, typically Convolutional Neural Networks. Because the training process will take a long time, Transfer Learning was applied to accelerate the learning process. Transfer learning uses an existing model as a base model in our neural network by transferring information from the related domain. There are Four base models are used, namely MobileNets, MobileNetV2, ResNet50, and VGG19. Each base model will be subjected to several experimental scenarios, such as setting the different learning rate values for pre-training and fine-tuning. The results show that the Convolutional Neural Network with transfer learning method can recognize edible or poisonous mushrooms with more than 86% accuracy. Moreover, the best accuracy result is 92.19% obtained from the base model of MobileNetsV2 with a learning rate of 0,00001 at the pre-training stage and 0,0001 at the fine-tuning stage.

Arrhythmia is a condition in which the rhythm of heartbeat becomes irregular. This condition in extreme cases can lead to fatal heart attack accidents. In order to reduce heart attack risk, appropriate early treatments should be conducted right after getting results of Arrhythmia condition, which is generated by electrocardiography ECG tools. However, reading ECG results should be done by qualified medical staff in order to diagnose the existence of arrhythmia accurately. This paper proposes a deep learning algorithm method to classify and detect the existence of arrhythmia from ECG reading. Our proposed method relies on Convolutional Neural Network (CNN) to extract feature from a single lead ECG signal and also Gradient Boosting algorithm to predict the final outcome of single lead ECG reading. This method achieved the accuracy of 96.18% and minimized the number of parameters used in CNN Layer.

Ekspresi wajah dapat memperkuat informasi yang disampaikan dalam sebuah komunikasi interaktif. Dalam bidang pengembangan karakter virtual khusus karakter wajah, ekspresi wajah diperlukan untuk menghidupkan karakter virtual agar terlihat natural dan alami seperti manusia. Salah satu jenis ekspresi emosi adalah ekspresi emosi majemuk (kompleks), yang merupakan gabungan dari dua atau lebih emosi dasar. Misalnya, ekspresi emosi wajah kecewa merupakan gabungan dari emosi marah dan sedih. Ekspresi wajah dapat muncul oleh adanya rangsangan emosi, salah satunya dari makna kalimat. Makalah ini bertujuan untuk mengekstraksi data emosi dari kalimat bahasa Indonesia dengan menggunakan proses klasifikasi multi-label model Convolution Neural Network (CNN) sehingga menghasilkan ekspresi wajah majemuk yang diterapkan dalam animasi karakter virtual. Adapun kelas-kelas emosi yang digunakan dalam proses klasifikasi teks adalah enam kelas emosi dasar yang meliputi marah, jijik, takut, senang, sedih, dan terkejut. Berdasarkan hasil eksperimen, model CNN dapat menghasilkan akurasi sebesar 94,5% dengan komposisi data latih dan data uji adalah 8:2. Hasil dari proses klasifikasi tersebut menunjukkan bahwa setiap kalimat dapat menghasilkan lebih dari satu kelas emosi dasar yang membentuk emosi majemuk. Tiap kelas emosi majemuk tersebut divisualisasikan menggunakan karakter wajah virtual untuk membentuk ekspresi wajah yang dapat merepresentasikan emosi majemuk.

Sentiment analysis is an analysis with an objective to identify like, dislike, comments, opinion, or feedback on certain content which will be categorized into positive, negative, or neutral. In general selection, sentiment analysis widely known to be used to predict the winner on election process. This method tries to dig the people sentiment on their governor candidates during election, whether it’s positive, negative, or neutral opinion. The output of the positive sentiment is related to people acceptance towards one of the election nominee. That statement usually applied as a base reference for determining the result of the election process. In sentiment analysis, the importance of its fuzzy logics must be considered. Each of the people statement is assumed to have the level of positive, negative, or neutral percentage. The concept of fuzzy logic is developed and applied on one of this text mining method. This research is focusing on comparison analysis and fuzzy logic application in sentiment analysis method. Two method which discussed in this research are Fuzzy Naïve Bayes and Sentiment Fuzzy with convolutional neural network. This research is applied on PILKADA of Solo and Medan district case study. The data of the people opinion are acquired from twitter and collected on September 2020 to December 2020. The two methods which mentioned before are implemented on the acquired data and the output of these method application then compared. The conclusion of this research suggest that different approach will resulting in different output.

ABSTRAK Telur ayam negeri merupakan salah satu sumber protein yang banyak dikonsumsi masyarakat Indonesia. Untuk menjaga kualitas telur ayam negeri yang beredar di Indonesia, diperlukan sistem yang mampu mengidentifikasi grade telur ayam dan mudah digunakan oleh masyarakat. Penelitian sebelumnya telah mengembangkan sistem pengklasifikasian grade telur ayam negeri secara invasive dengan tingkat akurasi 80%, namun sistem ini membutuhkan sampel telur yang dipecahkan sehingga setiap sampel telur tersebut tidak dapat disimpan dalam waktu lama. Oleh karena itu, penelitian ini mengembangkan sistem klasifikasi grade telur ayam tanpa perlu memecahkan sampel telur ayam (non-invasive). Dengan menggunakan metode Convolutional Neural Network (CNN), sistem mampu mengidentifikasi grade telur ayam negeri pada tingkat akurasi 85,86% dengan arsitektur LeNet-5, optimizer Adam, learning rate 0,001, dan epoch 50. Kata kunci: telur ayam negeri, non-invasive, convolutional neural network, LeNet-5 ABSTRACT Local Chicken egg are one of the sources of protein that is widely consumed by the people of Indonesia. To maintain the quality of local chicken egg in the market, a system that can identified chicken egg’s grade and easy to use is needed. Previous research has developed an invasive chicken egg’s grade classification system with 80% accuracy. However, the system required egg sample to be cracked so the egg sample can’t be stored for too long. This research develop a non-invasive chicken egg’s grade classification system, which doesn’t require egg sample to be cracked. By using Convolutional Neural Network (CNN), system can identified chicken egg’s grade at 85,86% accuracy with LeNet-5 architecture, Adam optimizer, learning rate 0,001, and epoch 50. Keywords: local chicken egg, non-invasive, convolutional neural network, LeNet-5

Salah satu warisan budaya Indonesia yang diakui dunia adalah kain batik. Beragamnya motif batik di Indonesia membuat masyarakat awam sulit membedakan motif-motif yang ada. Penelitian ini menggunakan convolutional neural network (CNN) dalam melakukan klasifikasi multi-label citra motif batik. CNN merupakan salah satu algoritma deep learning pengembangan multi-layer perceptron (MLP) yang telah banyak digunakan dalam klasifikasi data, khususnya klasifikasi citra. Hasil penelitian menunjukkan akurasi penggunaan arsitektur CNN dalam melakukan klasifikasi multi-label pada 15 motif batik mencapai 91.41% dengan penggunaan epoch 100.

Cataract are the highest cause of blindness that there are 32.4 million people experiencing blindness and as many as 191 million people experiencing visual disabilities in 2010 in the world. On the other hand, the longer a patient suffers from cataracts or late treatment. The development of cataract identification using a traditional algorithm based on feature representation is highly dependent on the classification process carried out by an eye specialist so that the method is prone to misclassification of a person detected or not. However, at this time there is a deep learning, convolutional neural network (CNN) which is used for pattern recognition which can help automate image classification. This research was conducted to increase the accuracy value and minimize data loss in the process of cataract identification by performing an experience namely the manipulation process was carried out by changing epochs. The results of this study indicate that the addition of epochs affects accuracy and loss data from CNN. By comparing variety of epoch values it can be ignored that the higher the age values used, the higher the value of the model. In this study, using the epoch 50 value reached the highest value with a value of 95%. Based on the model that has been made it has also been successful to receive images according to the specified class. After testing accurately, 10 images achieved an average accuracy of 88%.