Most of the reverse image searching is based on the image search as a whole. But it does not always catch the information regarding how inter-related two images are, based on their contents. A picture having a number of birds is far more related to a picture having a few birds and an animal than a third picture with all cars. So here a different approach has been used to capture this relation by matching the component of the images. A given composite image (with multiple objects) is searched among a set of other composite images and ordered based on how closely related it is with the images of the set. The top-most image in the ordering indicates the closest image to the given image. For component detection, selective search with fast non-maximal suppression has been used with ZCA normalization. The Convolutional neural network (CNN) have been used for the identification of the components. This can be used to find similarity among images which is difficult to find in conventional methods. Another application of this project is visual question answering it is asked to explain an image with only one of given four options. In that case based on the components of the image it can be mapped to the correct option.