Abstract: Automatic detection of R-peak is changing into a prominent tool for automatizing the designation of significant cardiovascular components. R-wave peaks in electrocardiography is a major indicator of the cardiac dysfunction. Most of the R-wave peaks prospectors have great troubles owing to non-stationary attitude of the ECG signal. In our current research, a smart technique established on bases of median filter and an automated R-wave peaks detection technique has been proposed. First median filter approach is used to eliminate the base line from the ECG signal. Then an efficient signal reshaping with sequential steps was proposed to detect the QRS position. The efficiency of the proposed work has been tested on MIT-BIH arrhythmia database. The technique has also been applied to a real signal for patients with different heart diseases. The proposed technique showed better performance and faster detection compared to the latter techniques at the same time. Error rate detection is low 0.05%, Positive predictivity (P+) of 99.85% Sensitivity (Se) of 99.94% and an average F-score of 0.999 are accomplished for the suggested ECG detector which are superior to the former results.