Weather image classification using convolutional neural network with transfer learning

Weather condition is an important factor that considered for various decisions. In the industrial world, weather classification very useful, such as in development of self-driving cars, smart transportation systems, and outdoor vision systems. Manual by humans inconsistent takes a long time. forecast information obtained from internet not real time at specific location. image has unique characteristics because one type can be like another. Computer branch computer science to recognize or classify images assist classifying do depend on internet. This study aims using Convolutional Neural Network (CNN) with Transfer Learning. Four CNN architectures, MobileNetV2, VGG16, DenseNet201, Xception were used perform classification. learning was speed up process training models get better performance faster. The proposed method will applied which consists six classes, cloudy, rainy, shine, sunrise, snowy, foggy classified this study. experiment result 5-cross validation 50 epochs showed best average accuracy 90.21% 10,962 seconds MobileNetV2 fastest 2,438 83.51% accuracy.