Situation-oriented clustering of sightseeing spot images using visual and tag information

Now a day, the tourists get used to take many photos in a journey and share these sightseeing spot images on album websites. The meaningful grouping of these images will become important and useful. In particular, sightseeing spot scenes are vary with different situations, such as weather conditions and seasons. Thus the categorization of different situations is expected to be beneficial for tourists to plan when to visit there. This paper proposes a hybrid approach which integrates content-based image clustering with filtering based on tag information of image. Content-based image clustering categorizes sightseeing spot images into night, sunrise/sunset, cloudy, and shine situations based on color feature extraction from ROI (region of interest). By using geotag information, collected images can be limited to a reasonable boundary to eliminate outliers. Furthermore, by using the timestamp of images, the four situation categories constructed by content-based image clustering are further verified to increase the accuracy. Experimental results show that the hybrid approach of content-based image clustering and tag-based filtering is effective for obtaining clusters with high precision and recall.