No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
OPTIMASI PENERAPAN ALGORITMA CONVOLUTION NEURAL NETWORK DALAM KLASIFIKASI TINGKAT KESEGARAN DAGING SAPI
Abstract
This research discusses optimizing the application of the Convolutional Neural Network (CNN) algorithm to overcome the problem of mixing fresh and non-fresh beef on the market. The focus of the research is classification of freshness levels in beef images using the CNN method with ADAM optimizer. Data collection involves open and private data. Image preprocessing is done with LabelEncoder and cv2. The research results show that this method is very effective in identifying and classifying the level of freshness in beef images. By determining optimal parameters, the model achieved the highest accuracy level of 98.50% at 10 epochs and a learning rate value of 0.001. Confusion matrix shows good results with a high number of True Positives. By applying CNN with the ADAM optimizer, it provides an effective solution to the problem of classifying beef freshness levels because this model is able to classify beef images well.
... Salah satu algoritma yang populer dalam Machine learning untuk pengolahan data visual berbentuk foto dan video adalah Convolutional Neural Network (CNN) [13]. CNN memiliki kemampuan untuk melakukan klasifikasi pada foto atau gambar dengan memfilter gambar atau foto tersebut dengan memecah bagianbagian foto jadi lebih kecil, serta memeriksa dan menilai beberapa bagian dari foto tersebut untuk mengidentifikasi sesuai dengan kategori yang dibutuhkan [14]. Pada kasus ini membuat sebuah pemodelan dari Machine learning untuk mengeksekusi tugas yang repetitif, yang dilakukan secara berulang, terhadap cacat pada cat, secara efisien [15]. ...
Quality control is an important process carried out at the last stage of the production process, this activity is carried out by checking a product. Painting defects on cars are a problem that must be considered in the car production process at car companies. The perfection of a product is important to increase the level of customer satisfaction. These checking activities are still carried out manually with human power, which can still cause defective products to be missed in a production process that occurs as a result of human error. The use of artificial intelligence can be used to detect image and video objects, used to overcome the problem of human error in carrying out checks. Convolutional Neural Networks (CNN) is an algorithm that can be used in product defect inspection, image recognition, and image classification. The study focuses on modeling the inspection and detection of painting defects in cars using CNN, emphasizing the importance of quality control in ensuring product quality. The CNN model is trained with image data of normal car paint and defective car paint, and evaluated using a confusion matrix for optimal parameters. The results show quite high accuracy in detecting car paint defects of 98% with the help of the ResNet50 transfer learning CNN architecture.
Halal and thayyib food encompasses all types of food that are permitted and beneficial for human health, one of which is meat consumption. The high demand for meat in Indonesia has led to an imbalance between supply and demand, resulting in fraudulent practices. Therefore, the classification process is crucial, particularly in distinguishing between beef and pork, which are commonly sold in the market. This study aimed to develop a pork detection system using a Convolutional Neural Network (CNN) model. A total of 224 images was divided into two classes: beef and pork. The data was split with a training-to-testing ratio of 70:30. The model testing results, using a confusion matrix, showed classification performance with an accuracy of 97.95%, precision of 97.67%, recall of 98.44%, and an F-1 score of 98.05%. These results showed that the model successed in identifying and classifying between beef and pork.
The implementation of deep learning has expanded into various fields, not confined solely to the field of education, particularly in computer science. It has also integrated technology into various other domains, including geospatial, remote sensing, and even the medical field. This development has made a significant contribution to reshaping the way humans understand and tackle challenges across different sectors. In this context, deep learning is employed for object detection and classification. Despite the considerable progress facilitated by the application of deep learning, object detection remains a challenge that is not entirely resolved. Constraints such as variations in lighting conditions, angles of view, and object diversity make achieving high-accuracy object detection a difficult task. Therefore, further research is required to comprehend and compare the performance of various deep learning models in addressing this issue. This research focuses on the comparison of two deep learning models, namely ResNet50 and Xception, in terms of loss metrics when detecting an object, in this case, a chair. The models are provided with input images of chairs and predict whether the chairs are empty or occupied. The results obtained from this research indicate that the ResNet50 model has a lower total loss value of 0.19422098, while the Xception model has a total loss value of 1.1822930. The lower the loss value, the better the model's performance. Based on the comparison results, the author has developed a web application simulator using Flask, utilizing the model with the lowest loss, which is the ResNet50 model.
Konsumsi daging di Indonesia didominasi oleh sapi, kerbau, dan ayam. Namun, beberapa pedagang nakal mencampur daging sapi dengan daging babi sehingga sulit dibedakan oleh masyarakat awam. Beberapa penelitian telah menggunakan metode Convolutional Neural Network (CNN) untuk mengklasifikasikan citra, namun kekurangan data menjadi tantangan. Oleh karena itu, penelitian ini menerapkan teknik augmentasi data pada model CNN Alexnet untuk mengklasifikasikan daging sapi, babi, dan daging oplosan. Penelitian ini menggunakan dua rasio pembagian data yang berbeda, yaitu 90:10 dan 80:20, dengan total 600 data non-augmentasi dan 3000 data augmentasi yang dibagi menjadi tiga kelas. Beberapa hyperparameter diuji untuk mengoptimalkan kinerja model seperti optimizer Adaptive Moment Estimation (Adam), Stochastic Gradient Descent (SGD) dan Propagasi Root Mean Square (RMSprop) serta learning rate 0.1, 0.01, 0.001 dan 0.0001. Hasil menunjukkan bahwa penggunaan data citra augmentasi dengan optimizer Adam dan learning rate 0,001 memberikan accuracy tertinggi sebesar 85,00%. Sementara itu, penggunaan data citra non-augmentasi dengan skenario optimizer RMSprop dan learning rate 0, 0001 menghasilkan performa yang sedikit lebih rendah, yaitu mendapatkan accuracy 80.00%. Keduanya menggunakan perbandingan data 80:20. Teknik augmentasi data berhasil meningkatkan kinerja model deep learning dengan menciptakan data baru dari data yang ada.
Indonesia is the fourth largest coffee producing country in the world. However, when compared to 3 other countries, Indonesia's coffee production is still relatively small. Many factors cause this to happen, including the number of farmers' coffee trees that are attacked by diseases. If the handling of this disease is slow, then the disease in one tree can be transmitted to other trees. This causes a decrease in Indonesian coffee productivity. In this study, the author implemented the Alexnet Convolutional Neural Network (CNN) architecture using the MATLAB programming platform for the identification of diseases in coffee plants through images. The total number of datasets used is 300 data which is divided into 3 classes, namely health, rust and red spider mite. The training process involving 260 training data resulted in an accuracy of 69.44-80.56%. The network testing process using 40 test data resulted in an accuracy of 81.6%. Based on the results of the study, it can be said that the Alexnet architecture is accurate for the classification of leaf pests on coffee plants
Deep Learning, especially the Convolutional Neural Network (CNN) has proven to be reliable in processing data from various programming language platforms by utilizing deep learning. In this study, we modified it by calculating the statistical variance. The modifications made are replacing calculations on the Pooling Layer which generally use two formulas, namely max pooling and average pooling. We use the standard deviation to change the reduced image intensity value. With the research experiments built, it is expected to be able to perform facial recognition as an indicator for testing modifications. The Layer Pooling experiment uses the Standard Deviation for Classifying Face Image Datasets with the CNN Method, including the type of dataset used is the Aberdeen dataset https://pics.stir.ac.uk/2D_face_sets.htm. From the results of the experiments conducted, it was found that the highest value was using the Elu activation function and the Adagrad optimizer worth 77.844% for max pooling and 79.844% for pooling with a standard deviation. The Cellu activation function and the RMSprop optimizer are 77.986% for max pooling and 75.986% for pooling with a standard deviation. The highest score with the Softplus activation function and the Sgd optimizer is 77.844% for max pooling usage and 76.344% for pooling with standard deviation. The Tanh activation function and the Adadelta optimizer are 87.844% for max pooling and 85.844% for pooling with a standard deviation. The Elu activation function and the Adam optimizer are 87.853% for the use of max pooling and 85.285% for pooling with a standard deviation. By using the Elu activation function and the Adamax optimizer, the value is 87.842% for max pooling and 86.242% for pooling with a standard deviation. The highest score is using the Elu activation function and the Nadam optimizer with a value of 87.845% for max pooling usage and 86.345% for using standard deviation calculations as pixel pooling. From all experiments it was stated that the use of pooling with the highest value technique or max pooling still gave a better value than using the standard deviation calculation with the best tuning results using the Elu activation function and Adam's Optimiser, which was 87.853%.
- session recognition is an interesting topic, where facial expressions in today's technological advances can support several fields such as health, business, and so on. Facial expression recognition can be done using the extraction of certain features. Meanwhile, Convolutional Neural Network (CNN) can recognize an object in the image through the features found by itself in the convolution process. By using CNN's advantages, this study aims to see CNN's performance in facial expressions of happiness and sadness in unideal data conditions. So based on this research, on the FER2013 dataset, CNN using the Adamax optimizer produced a fairly good performance where the value is given is 66% compared to Adam, N-Adam, and SGD.Keywords - CNN, Facial Expression, FER-2013 Abstrak – Pengenalan ekspresi merupakan topik penelitian yang menarik, dimana peran ekspresi wajah dalam kemajuan teknologi saat ini dapat mendukung beberapa bidang seperti kesehatan, bisnis, dan sebagainya. Pengenalan wajah dapat dilakukan dengan menggunakan ekstraksi fitur-fitur tertentu. Sementara itu, Convolutional Neural Network (CNN) dapat mengenali objek pada citra melalui fitur yang ditemukannya sendiri dalam proses konvolusinya. Dengan menggunakan keunggulan CNN, maka penelitian ini bertujuan untuk mengetahui performa CNN dalam mengenali ekspresi wajah bahagia (happy) dan sedih (sad) pada kondisi data tidak ideal. Maka berdasarkan hasil penelitian ini, pada dataset FER2013, CNN dengan menggunakan Adamax optimizer menghasilkan performa yang cukup baik dimana akurasi yang diberikan adalah sebesar 66% dibandingkan dengan Adam, N-Adam, dan SGD.Kata Kunci - CNN, Ekspresi Wajah, FER-2013.
Brain tumors' diagnoses occur mainly by Magnetic resonance imaging (MRI) images. The tissue analysis methods are used to define these tumors. Nevertheless, few factors like the quality of an MRI device and low image resolution may degrade the quality of MRI images. Also, the detection of tumors in low-resolution images is challenging. A super-resolution method helps overcome this caveat. This work suggests Artificial Intelligence (AI)-based classification of brain tumor using Convolution Neural Network (CNN) algorithms is proposed to classify brain tumors using open-access datasets. This paper hiders on a novel Discrete Cosine Transform-based image fusion combined with Convolution Neural Network as a super-resolution and classifier framework that can distinguish (aka, classify) tissue as tumor and no tumor using open-access datasets. The framework's performance is analyzed with and without super-resolution method and achieved 98.14% accuracy rate has been detected with super-resolution and ResNet50 architecture. The experiments performed on MRI images show that the proposed super-resolution framework relies on the Discrete Cosine Transform (DCT), CNN, and ResNet50 (aka DCT-CNN-ResNet50) and capable of improving classification accuracy.
Meat is a staple food for some Indonesian people, apart from the taste, meat also contains vitamins and minerals that are good for the human body, however, not all meat can be consumed by the Indonesian people. the texture and color of beef, pork and mutton have similarities and tend to be similar, therefore a system is needed to recognize the three types of meat. In this study, the authors use various types of Deep Learning architecture such as Resnet-50, VGG-16, VGG-19 and Densenet-121 with Hard Voting to improve the performance of Deep Learning in recognizing the three types of meat. The results show that Resnet-50 with Hard Voting can outperform Deep Learning Resnet-50, VGG-16, VGG-19 and Densenet-121- with f1 score 98.88%, precision 98.89% and recall 98.88%. in image classification of pork, beef and mutton.
Kondisi pandemi global Covid-19 yang muncul diakhir tahun 2019 telah menjadi permasalahan utama seluruh negara di dunia. Covid-19 merupakan virus yang menyerang organ paru-paru dan dapat mengakibatkan kematian. Pasien Covid-19 banyak yang telah dirawat di rumah sakit sehingga terdapat data citra chest X-ray paru-paru pasien yang terjangkit Covid-19. Saat ini sudah banyak peneltian yang melakukan klasifikasi citra chest X-ray menggunakan Convolutional Neural Network (CNN) untuk membedakan paru-paru sehat, terinfeksi covid-19, dan penyakit paru-paru lainnya, namun belum ada penelitian yang mencoba membandingkan performa algoritma CNN dan machine learning klasik seperti Support Vector Machine (SVM), dan K-Nearest Neighbor (KNN) untuk mengetahui gap performa dan waktu eksekusi yang dibutuhkan. Penelitian ini bertujuan untuk membandingkan performa dan waktu eksekusi algoritma klasifikasi K-Nearest Neighbors (KNN), Support Vector Machine (SVM), dan CNN untuk mendeteksi Covid-19 berdasarkan citra chest X-Ray. Berdasarkan hasil pengujian menggunakan 5 Cross Validation, CNN merupakan algoritma yang memiliki rata-rata performa terbaik yaitu akurasi 0,9591, precision 0,9592, recall 0,9591, dan F1 Score 0,959 dengan waktu eksekusi rata-rata sebesar 3102,562 detik.
Facial recognition work combined with the facial owner's position estimation feature can be utilized in various everyday applications such as face attendance with position detection. Based on this, this study offers a system testing experiment that can be run with facial recognition features and an Indoor Positioning System (IPS) to automatically check the location of the owner of the face. Recently, deep learning algorithms are the most popular method in the world of artificial intelligence. Currently, the Deep Learning algorithm toolbox has provided various programming language platforms. Departing from research findings related to deep learning, this study utilizes this method to perform facial recognition. The system we offer is also capable of checking the position or whereabouts of objects using Indoor Positioning System (IPS) technology. Facial recognition evaluation using CNN obtained a maximum value = 92.89% and an accuracy error value of 7.11%. Meanwhile, the average accuracy obtained is 91.86%. In the evaluation of the estimated position tested using DNN, the highest value of r2 score is 0.934, the lowest is 0.930 and an average is 0.932 and the highest value is MSE is 4.578, the lowest is 4.366 and the average is 4.475. This shows that the facial recognition process that is tested is able to produce good values but not the position estimation process. Keywords: Face Recognition, IPS, CNN, MSE, Accuraccy.
Daging sapi merupakan salah satu bahan makanan yang mengandung protein tinggi dan termasuk media yang baik bagi pertembuhan mikroorganisme seperti bakteri. Salah satu metode yang dapat digunakan untuk menghitung jumlah bakteri pada daging sapi yaitu metode hitung cawan. Beberapa cara yang dapat dilakukan pada metode hitung cawan yaitu metode pour plate, spread plate, dan drop plate. Penggunaan cara hitung cawan yang berbeda seta larutan pengencer yang berbeda menyebabkan hasil perhitungan yang berbeda juga. Tujuan dari penelitian ini yaitu untuk mengetahui perbedaan cara preparasi sampel serta larutan pengencer yang digunakan terhadap hasil hitung cawan bakteri pada sampel daging sapi. Metode hitung cawan yang digunakan pada penelitian ini yaitu pour plate, spread plate, dan drop plate dengan 3 jenis pelarut yaitu akuades steril, garam fisiologis 0,85% dan larutan Buffer Peptone Water (BPW) yang ditambahkan 1,25% Na Sitrat.Berdasarkan hasil penelitiandiketahui bahwa cara preparasi sampel daging sapi yang lebih efisien yaitu dengan cara dihaluskan ditunjukkan dengan hasil hitung cawan yaitu 106 CFU/gram dan hasil hitung tidak bervariasi dibandingkan daging sapi yang dipotong (105 CFU/gram) dengan hasil hitung yang bervariasi. Hasil analisis data menunjukkan bahwa penggunaan cara preparasi sampel yang berbeda berpengaruh pada hasil perhitungan jumlah koloni disetiap metode hitung cawan. Larutan buffer peptone water dan garam fisiologis 0,85% merupakan larutan pengencer yang lebih baik dibandingkan dengan larutan akuades. Hal ini ditunjukkan dengan hasil hitung yang tidak bervariasi disetiap metode hitung cawan yang digunakan. Berdasarkan hasil hitung yang didapatkan, larutan buffer peptone water menghasilkan hasil perhitungan jumlah koloni terbesar.
Kebakaran menimbulkan kerugian yang dapat membahayakan jiwa maupun harta benda. Lokasi dan waktu terjadinya kebakaran sulit untuk diprediksi. Citra dari kamera CCTV dapat digunakan untuk memantau terjadinya kebakaran pada suatu lokasi. Dengan pemanfaatan perkembangan teknologi saat ini, data citra tersebut dapat diolah sehingga memberikan informasi terjadinya kebakaran pada suatu lokasi dengan lebih cepat dibandingkan secara manual. Salah satu metode yang dapat digunakan dalam pengolahan citra kebakaran adalah deep learning Salah satu model deep learning yang memiliki kinerja terbaik dalam pengenalan dan klasifikasi citra yaitu Convolutional Neural Network (CNN). Penelitian terdiri dari beberapa tahap yaitu pengumpulan data, pra-pemrosesan data, pembagian data, perancangan arsitektur model CNN, pelatihan, pengujian, dan hasil. Jumlah data citra yang digunakan pada penelitian ini yaitu 950 citra yang terdiri dari 700 data pelatihan, 170 data validasi, dan 80 data pengujian. Arsitektur model CNN yang digunakan terdiri dari dua lapisan konvolusi dan dua lapiran pooling. Pada pelatihan, model dikenalkan dengan pola citra kebakaran dan non kebakaran kemudian divalidasi. Pada tahap pengujian, model mengklasifikasikan citra kebakaran dan non kebakaran. Tingkat akurasi model pada tahap pelatihan sebesar 98.8% dan pada tahap pengujian sebesar 90%.
- A N Rahmad
- F S Pribadi
A. N. Rahmad and F. S. Pribadi, "Edu
Komputika Journal," Edu Komputika J.,
vol. 5, no. 1, pp. 33-43, 2018.
Kajian analisis konsumsi daging sapi dan daging babi ditinjau dari kesehatan
- S Maiyena
- E R Mawarnis
S. Maiyena and E. R. Mawarnis, "Kajian
analisis konsumsi daging sapi dan daging
babi ditinjau dari kesehatan," J. Pendidik.
Tambusai, vol. 6, no. 1, pp. 3131-3136,
2022.
Pengaruh Lama Penyimpanan Dalam Freezer Terhadap Sifat Fisik dan Jumlah Bakteri pada Daging Sapi Bali Jantan
- A Baiq Annisa
- Sulistia
A. Baiq Annisa Sulistia, "Pengaruh Lama
Penyimpanan Dalam Freezer Terhadap
Sifat Fisik dan Jumlah Bakteri pada
Daging Sapi Bali Jantan." Universitas
Mataram, 2023.
The Effect Of Image Feature Selection On The Local Beef
- T Yulianti
- M Telaumbanua
- H D Septama
- H Fitriawan
T. Yulianti, M. Telaumbanua, H. D.
Septama, and H. Fitriawan, "The Effect
Of Image Feature Selection On The Local
Beef," J. Tek. Pertan. Lampung, vol. 10,
no. 1, pp. 85-95, 2021,
Implementasi Deep Learning Menggunakan Convolutional Neural Network (Cnn) Pada Ekspresi Manusia
- P A Nugroho
- I Fenriana
- R Arijanto
P. A. Nugroho, I. Fenriana, and R.
Arijanto, "Implementasi Deep Learning
Menggunakan Convolutional Neural
Network (Cnn) Pada Ekspresi Manusia,"
Algor, vol. 2, no. 1, pp. 12-20, 2020.
Optimalisasi Adaptive Karnel Convolution Neural Netwok Menggunakan Algoritma Adgrad
- J K Azhar
J. K. Azhar, "Optimalisasi Adaptive
Karnel Convolution Neural Netwok
Menggunakan
Algoritma
Adgrad."
Universitas Siliwangi, 2022.
Identifikasi Daging Segar Berdasarkan Citra menggunakan Convolutional Neural Network
- M F S D Cahyo
- D Udjulawa
M. F. S. D. Cahyo and D. Udjulawa,
"Identifikasi Daging Segar Berdasarkan
Citra menggunakan Convolutional Neural
Network," in MDP Student Conference,
2023, pp. 306-313.