How to set a condition if a sample/test image does not match with the image categories from the Image Category Classification Using Deep Learning example

In the image category using deep learning example if I enter a test image that does not belong to any of the trained category the program classifies the test image into any of the trained image categories but actually the test image does not belong to any of the category .For example : if horse , dog and cat are the three image categories that my network is trained with and if I give an image of "table" then the program will classify it as horse or dog or cat .So please anyone can help by providing a solution or code to resolve this problem