Text Image Processing and Segmentation

I am working on a project and one of its functionalities is to segment text images into words. The following image contains English and Arabic words. Words are separated by a fixed space. So how can I segment this text image into sub- images based on the words. I mean each segmented image contains one word. In advanced, based on the segmented images can I identify the language of the text that included in the segmented image especially the English ones.