ECG signal is time varying in nature which is most common source used for the diagnosis and analysis of heart diseases present in the patient.ECG is recorded by placing electrodes at specified positions of human body. During recording, ECG is contaminated with artifacts and noises which always degrade its quality, and makes accurate and automatic interpretation more difficult. Power line interference, baseline wanders and muscle tremors are mostly noticed artifacts or noises. So for accurate delineation of characteristics points of ECG, a good quality of ECG is necessary. This paper is presenting a survey on various methods developed for de-noising, delineation of characteristics points and classification of diseases along with their respective advantages and disadvantages.