Face recognition is an interesting and challenging problem and impacts important applications in authentication and personal identification among others. Extraction of these important elements from a picture, their useable representation, and classifications are the core concepts of automatic face recognition. Face recognition based on the geometric features of a face is probably the most instinctive approach for Human identification. The entire process may be broken down into three main parts, with the first step being the search for a reliable database of faces that includes numerous photographs for each person. The next phase is to find faces in the database photos so that the face recognizer can be trained on them. The last step is to test the face recognizer to see if it can still find the faces that it was trained on. Here implement an application for criminal detection, it helps forensic department for the accurate identification of criminal using his face image. The training face images are initially collected and stored on server. This system provides essential security to apartments and other control applications. During face capturing the face image will be match with registered images. An efficient classifier uses to classify the face images accurately. Criminal images are collected and stored by forensic department. During capturing process, face image will be classified with the criminal image database. If a match is made, we will be able to identify the offender and quickly make an arrest.