Open AccessIMAGE BASED WEATHER AND TEMPERATURE ANALYSIS SYSTEM
Author(s) -
Ashlin Francis,
S Vineetha
Publication year - 2019
Publication title -
international journal of engineering applied science and technology
Language(s) - English
Resource type - Journals
ISSN - 2455-2143
DOI - 10.33564/ijeast.2019.v04i02.010
Subject(s) - image (mathematics) , meteorology , computer science , environmental science , remote sensing , computer vision , artificial intelligence , geology , geography
Weather prediction is always a hot topic, as it has great importance in many fields. Recent issues related to climate change shows the importance of weather in man’s daily life. Weather information has a big social impact. There are a lot of methods to predict weather conditions but most of them use additional hardware. For an example, weather prediction using separate weather satellites. These methods are very expensive. This is the place where image based weather recognition is important. Image based weather recognition is an emerging topic. Images carry a lot of information to understand weather. The main advantage of image based weather recognition is large quantity of data, compare to other methods it doesn’t need additional hardware, comparatively less cost and images are easily accessible. Image based weather recognition has large applications like driving assistance, robotics etc. Convolutional neural network, a deep learning method is used to identify the weather information. Weather is divided in to six classes: cloudy, haze, rainy, snowy, sunny, thunder. Images are considered as clusters and each cluster is considered for feature extraction. Temperature analysis is the other field in this work. Even though temperature is influenced by many factors like humidity, pressure, weather etc. Weather has a great influence on temperature. Temperature analysis is calculated using the scores obtained in each weather classes. Keywords—Convolutional Neural Network , VGG16, Euclidean Distance, Image based Weather Recognition.