Aiming at the problem that knuckleprint and palmprint are not easy to be extracted and the extracting process is easily affected by illumination and fine lines on the palm

an improved algorithm for knuckleprint and palmprint recognition based on the top-hat transformation is proposed. The top-hat transformation can eliminate the influence of uneven illumination and extract palm main line well. Graying and the top hat transformation are carried out on collected palm image

binarization processing is carried out on palm image

isolate the palm area and find its position

and isolate the palm boundary with Sobel edge detector

determine the coordinates of the fingertip and finger valley

segment the palm area

find the knuckleprint and palmprint area in general

match them with the cross-correlation method. The experimental results show that this method can eliminate the influence of uneven illumination and uneven palm color

with fast speed and strong anti-noise capability

suitable for large-scale palm library to achieve coarse screening.