Lung cancer has become one of the most widely spreading diseases in the world. Detection of lung nodules is the initial step in lung cancer detection. We propose an idea to locate the lung nodules based on its intensity, shape and size. Lung CT images are used for detecting the lung nodules. Initially, Variant Ant Colony Optimization algorithm is used to detect the edges. Variant ACO algorithm greatly helps to reduce the False Positives. Nodules centers are detected in the edge detected image based on the proposed black circular neighborhood algorithm. The intensity of the lung nodules are classified based on the input image using the positions of the lung nodule center. We use lung intensity identification algorithm. Finally the malignant lung nodules are identified from the input CT image based on three features – intensity, shape and size.