With the proliferation of digital cameras, images of crimes, such as child sexual abuse images, are increasing dramatically. Both verification and identification of criminals and victims in these images are highly difficult and often impossible for the current biometric technology because their faces, tattoos, and distinctive skin mark patterns are not always observable. Superficial blood vessels under skin are a potential solution to compensate the weaknesses of the traditional biometric traits. However, blood vessels were neglected by law enforcement agencies because they are generally invisible in color images. To use blood vessel patterns in forensic analysis, this paper proposes three computational models to uncover hidden patterns, two optimization schemes to handle illumination variations and prevent over-relying on biophysical parameters measured in ideal medical conditions, a matching algorithm to automatically extract and compare noisy patterns, and two fusion rules to combine patterns from the three models for performance enhancement. The experimental results on 1900 color images and 1900 infrared images from 490 forearms and 460 thighs show that the matching performance of the blood vessel patterns from the color images is comparable with that from the infrared images. The proposed models are also applied to hands, arms, thighs, chests, breasts, and abdomens of men, women, and children in indoor and outdoor images collected from the Internet. Though these images were taken in uncontrolled environments and the subjects had different poses, the proposed models can uncover blood vessels. These results indicate that the potential of using blood vessel patterns in forensic analysis was underestimated.