CEDAR Signature Handwriting Identification Dataset

The CEDAR Signature dataset is suitable for handwriting identification and signature authentication. The dataset contains handwriting results of 55 volunteers, including 24 real signatures and 24 forged signatures, totaling 2,640 images. Each signature is scanned at 300 dpi grayscale and binarized using grayscale histogram. Salt pepper noise removal and slant normalization are two steps involved in image preprocessing.