科研之友
微信
新浪微博
Facebook
分享链接
摘要
This paper describes a method for image classification and retrieval for natural and urban scenes. The proposed algorithm is based on hierarchical image content analysis. First image is classified as urban or natural according to color and edge distribution properties. Additionally scene is classified according to its conditions: illumination, weather, season and daytime based on contrast, saturation and color properties of the image. Then image content is analyzed in order to detect specific object classes: buildings, cars, trees, sky, road, etc. To do so, the image is recursively divided into rectangular blocks. For each block probabilities of membership in the specific class are computed. These probabilities are computed as a distance in a feature space defined by optimal feature subset selected on the training step. Blocks which cannot be assigned with a high confidence to any class using computed features are separated into 4 sub-blocks which are analyzed recursively. Process is stopped, then all blocks are classified or if the size of block is smaller then a predefined value is taken. Training process is used to select optimal feature subset for object classification. Training set contains images with manually labeled objects of different classes. Each image is additionally tagged with scene parameters (illumination, weather, etc.).
- 出版日期2013-9-20
