Hi all,
Am working on my final year project on image classification.The idea is to use the colour visual feature like color histograms to classify images.
But can anyone please give me an idea of what kinds of images the classification can be applied to. Since the classification is based on color, the images used should have predominant colours for the classification process. For example if we are to classify landscape images by colour, like forest, sunset, snow, bird flying images, the images can be distinguished by the colour itself. Forests have green colour, snow has white, sunset has a reddish orange colour, sky has blue colour.
Classifying landscape images are ok but it has been researched a lot in the past. And i dont know if it s ok to repeat work done in the past as a FYP.
So please give me some ideas of the type of images to use to classify them based on colour. I though of using rock images. But its really difficult to find the image datasets of different types of rocks with different colours.
Please help me on this.!!!
Thanks,
Vashini