Penerapan CNN Untuk Klasifikasi Tingkat Kematangan Berry Dengan Augmentasi Transformasi Dan Colorjitter Menggunakan Tensorflow
Abstract
With the increasing consumer demand for fresh and high-quality fruits, automated recognition of fruit ripeness becomes crucial to enhance the quality and efficiency in fruit production. This research focuses on the implementation of Convolutional Neural Network (CNN) with Transformation Augmentation and Color Jitter using TensorFlow to classify the ripeness level of berries through images. The dataset consists of images of berries that have undergone pre-processing steps, including resizing images to 224x224 pixels, horizontal flipping, a rotation range of 20 degrees, zoom range of 0.3, shear range of 0.3, height shift range of 0.3, width shift range of0.3, random hue of 0.08, random saturation of 0.6-1.6, random brightness of 0.05, and random contrast of 0.7-1.3. The network architecture comprises 4 convolution layers, 4 max-pooling layers, a global average pooling layer, and a dense layer. The results show that the CNN method with Transformation Augmentation and Color Jitter using TensorFlow can achieve good accuracy in classifying the ripeness of berries. After 500 epochs of training, the model achieved an accuracy of 93.75% on the training data and 90.33% on the testing data. The model also achieved 100% accuracy on 30 new data samples that were different from the training and testing data. Additionally, the use of Transformation Augmentation and Color Jitter in CNN has a positive impact on the accuracy of ripeness classification of berries, with the augmented model showing higher accuracy.
References
Anandhavalli Muniasamy and Areej Alasiry, “Deep Learning: The Impact on Future eLearning. International Journal of Emerging Technologies”, 2020, https://doi.org/10.3991/ijet.v15i01.11435
Dani, “Budidaya Blackberry, Blueberry dan Raspberry Untung hingga 65%”, 2019. https://berempat.com/bisnis/umkm/9893/budidaya-blackberry-blueberry-dan-raspberry-untung-hingga-65
Harshit Kumar, “Data Augmentation Techniques”, 2024, https://iq.opengenus.org/data-augmentation/
Laith Alzubaidi, Jinglan Zhang, Amjad J. Humaidi, Ayad Al-Dujaili, Ye Duan, Omran Al-Shamma, J. Santamaría, Mohammed A. Fadhel, Muthana Al-Amidie and Laith Farhan, “Review of deep learning: concepts, CNN architectures, challenges, applications, future directions”, 2021, https://doi.org/10.1186/s40537-021-00444-8
Sumit Saha, “A Comprehensive Guide to Convolutional Neural Networks”, 2018. https://towardsdatascience.com/a-comprehensive-guide-to-convolutional-neural-networks-the-eli5-way-3bd2b1164a53
Tran Xuan Truong, Viet-Ha Nhu, Doan Thi Nam Phuong, Le Thanh Nghi, Nguyen Nhu Hung, Pham Viet Hoa, and Dieu Tien Bui, “A New Approach Based on TensorFlow Deep Neural Networks with ADAM Optimizer and GIS for Spatial Prediction of Forest Fire Danger in Tropical Areas”, 2023, https://doi.org/10.3390/rs15143458
Tutut Furi Kusumaningrum, “Implementasi Convolution Neural Network (CNN) untuk Klasifikasi Jamur Konsumsi Di Indonesia Menggunakan Keras”, 2018.
Zaharchuk, Gong, Wintermark, Rubin, and Langlotz, “Deep Learning in Neuroradiology. American Journal of Neuroradiology”, 2018, https://doi.org/10.3174/ajnr.A5543