Computer-aided detection (CADe) of pulmonary nodules is critical to assist radiologist in early detection of lung cancer from computed tomography (CT) scans. So in proposed system we use CADe system based on hierarchical vector quantization (VQ) scheme. On comparing with commonly-used simple thresholding approach, the high-level VQ yields accurate segmentation of lungs from chest volume and in identifying initial nodule candidates (INCs) within lungs, low-level VQ proves to be effective for INC detection and segmentation, as well as computationally efficient compared to existing approaches. This proposed system also reduces false positive detection. False positive reduction is conducted via the rule based filtering operation in combination with feature-based support vector machine classifier. This proposed system shows out performance and demonstrate its potential for early detection of pulmonary nodules via CT imaging.