There is an increased tendency to copy another person's content blatantly from the information available in the Internet. The act of plagiarism is using someone else's idea and written work without their knowledge or acknowledgment; this is intellectual theft, and is a crime. To address this issue, it is necessary to maintain a plagiarism reporting system to keep track of text recycling. The objective of this paper, is to develop an application where texts are compared to detect plagiarism, even if the text is uploaded in image format. The text from images is extracted with the help of Optical Character Recognition (OCR). Similarity, analysis is calculated through machine learning techniques of word to vector conversion and cosine similarity. Dataset for comparison is taken from text scripts from the internet and manuscripts of journals or essays of students. The key concept behind this paper is to discourage academic plagiarism among the student community and to stimulate the practice of writing originally.