An algorithm for image segmentation of the secondary septa in the lungs using light microscopy is presented. After converting the original analog image to binary with proper thresholding, morphological operations are applied to individual lumen regions to extract the secondary septum regions that appear as cavelike regions in each lumen. To reduce the errors of misclassifying the primary septa from secondary septa because of the occasional leaks in the broken primary septum lines, a gap-closing procedure is designed before another separation procedure of the secondary septa is executed. The algorithm is applied to real lung images of both normal and cancerous lungs. The results show that the algorithm is robust and is able to identify and segment the secondary septa in lungs effectively.