手掌静脉成像是手掌静脉识别的难点,设计了一种在拍摄全手的情况下能够采集到较清晰手掌静脉图像的低成本多波段采集系统,并分别在760 nm,850 nm,890 nm和960 nm四种单波长及两两组合的近红外光源下采集手掌静脉,总样本数为2500幅手掌静脉图像,然后利用提取的有效区域中手掌静脉平均灰度与非手掌静脉平均灰度的比值为波长选择标准对图像质量进行比较,最后选择在760 nm和960 nm混合波长下采集手掌静脉。
The palm vein image acquisition is difficult in palm vein patteru recognition. A multi-wavelength palm vein image acquisition system is proposed in this paper,which is low cost and can image the whole palm. A total of 2500 palm vein images are acquired under the near-IR lighting source of 760 nm,850 nm,890 nm,960 nm and the combination of every two of them. The ratio of the average gray-scale value of palm vein areas and other areas in the region of interest (ROI) is taken as the criterion of wavelength choice. Palm vein image acquisition with the combination of 760 nm and 960 nm NIR is adopted at last.