Abstract: To improve authenticity of persons accessing secured locations, information, services, and/or goods, random pictures (images) and/or portions of picture are placed on a document (hard copy—e.g. a check—or computer generated). The person requiring access selects a set of one or more of the images/pictures, e.g. by crossing them out. Often the selected images/picture with be familiar to the user. The document is screened, e.g. by a special access server over a network, to check on whether the subset was correct, i.e., matches a subset of images previously stored and associated with the accessor. This can be combined with printed explicit textual questions related to an owner personal database and several possible answers for each question. For further security, biometrics, e.g. from user handwritten answer prompts, can be added. Similar security provision with random visual images can be used when users interact with computers to get access to some services (without providing hard copy documents).