http://www.iaeme.com/IJCIET/index.asp 1124 editor@iaeme.com International Journal of Civil Engineering and Technology (IJCIET) Volume 8, Issue 9, September 2017, pp. 1124–1127, Article ID: IJCIET_08_09_126 Available online at http://http://www.iaeme.com/ijciet/issues.asp?JType=IJCIET&VType=8&IType=9 ISSN Print: 0976-6308 and ISSN Online: 0976-6316 © IAEME Publication Scopus Indexed A SURVEY ON FAULT DETECTION TECHNIQUES IN DIFFERENT MACHINES - AN IMAGE PROCESSING APPROACH U. Rahamathunnisa Assistant Professor (Sr), School of Information Technology, VIT University, Vellore, Tamilnadu, India Babu Chellappa Chetty Associate Professor, School of Mechanical Engineering, VIT University, Vellore, Tamilnadu, India A. Clement King Associate Professor, Department of Computer Science, College of Computer Science, King Khalid University, Abha, KSA ABSTRACT Fault identification in any domain is a major challenge. Fault has to be identified in advance to prevent production problems. There are many image processing techniques used for fault identification and analysis. This paper presents a survey on various image processing approach for fault detection in various machines irrespective of domains. These techniques are used to retain the reliability of a system and prevent time delay. Key words: Fault detection techniques, Failure, Reliability, Image processing. Cite this Article: U. Rahamathunnisa, Babu Chellappa Chetty, A. Clement King, A Survey on Fault Detection Techniques in Different Machines- An Image Processing Approach. International Journal of Civil Engineering and Technology, 8(9), 2017, pp. 1124–1127. http://www.iaeme.com/IJCIET/issues.asp?JType=IJCIET&VType=8&IType=9 1. INTRODUCTION Failure in system causes a fall in reliability. There is a need to analyse the fault in advance so that productivity is not affected. There are various techniques proposed in the literature for fault detection and analysis. Early detection of fault in machine reduces the risk. A survey on fault identification methods are studied in [1] and [2]. Faults are the unacceptable variations in one of the system’s feature. Faults occur in the system due to malfunction in any of the components and also caused by any interruptions in the functioning of the system. Fault detection methods include data and signal based models, process models and knowledge based models as given below