Satellite Cloud Image Retrieval by using Different Methods.

      Аннотация: Satellite captures thousand of images everyday but all of them are not useful to us and only some of the images are utilized for our need. The challenging task in this project is to retrieve the useful image from the satellite image database which is done by various techniques. Content based image retrieval is the important tool to search the necessary image from the satellite database which is similar to the query image using 2 feature techniques namely shape features and text features respectively. Generally, different cloud images having different shapes based on the weather condition. In chapter 3, the rainfall is estimated using shape features in CBIR which is based on the shape of the cloud image captured from the satellite. In chapter 4, it describes the retrieval of satellite images from the satellite database. It allows the user to search the image from the database based on the texture feature. Texture features are estimated using Gray level Co-occurrence Matrix and the Euclidean distance metric is estimated for the similarity between the satellite images. The system performance is estimated by analyzing the retrieval result using precision.

Ключевые слова: image retrieval, CBIR, Shape feature, Texture Feature, gray level co-occurrence matrix, euclidean distance.