ECG feature extraction stage is a significant job in diagnosing most of the cardiac diseases after the preprocessing of the ECG signal. Features extracted from ECG are extremely useful in diagnosis. In the previous work to detect the QRS complex wavelet multi-resolution analysis, threshold consideration is used. There has been proposed a structure for detection of the QRS complexes of the ECG signals with the help of Virtual Instruments (VI) of LabVIEW for the standard MIT- BIH arrhythmia database. This structure detects the various features of QRS complex. This paper deals with a resourceful composite method which has been proposed for detrending, denoising and feature extraction of the ECG signals. The proposed structure first employed a wavelet-based detrending and denoising of the ECG signal. Then execute a novel ECG feature extractor. The proposed feature extractor consists of virtual instrument of LabVIEW like Read Biosignal VI, Extraction Portion of signal express VI, Waveform Max-Min VI etc. The Waveform Max-Min VI and Extraction Portion of signal express VI is the alternative of the Peak detector VI without any threshold calculation. Various features like QR level, RS level, QR slope and RS slope etc has been detected by proposed structure. LabVIEW 2013 version has been used here to design the feature extractor.