Photo Time-Stamp Recognition Based on Particle Swarm Optimization

Time-stamp in image shows the time when an image was taken, and which is important information for retrieval. An automatic photo time-stamp recognition approach is proposed in this paper. The proposed method consists of three steps: 1) A photo is roughly segmented to determine which area contains the photo time-stamp. Four detecting areas are chosen in this step, which probably include the photo-stamp, and then their R component and G component are margin-detected using Sobel operator to determine whether the photo time-stamp is in them; 2) The area which contains time-stamp of the photo is finely segmented to locate each character; 3) Particle Swarm Optimization (PSO) algorithm is used to do template matching and optimization recognition to complete the photo time-stamp recognition. This approach needs no prior knowledge and off-line learning processes. Experimental results demonstrate that the proposed approach has high recognition accuracy.