Abstract: This study deals with the segmentation and recognition of unconstrained off-line Bangla handwritten numerals. The projection profile-based heuristic technique is used to segment handwritten numerals. The results, based on this approach are quite satisfactory for the next classifier stage. Following the segmentation, character matrices from each numeral are extracted, normalized and then passed through an Artificial Neural Network (ANN) classifier for classification. The proposed scheme is tested on data collected from different individuals of various background and foreground colors.