A hand-held direct digital image sensor device includes a connection handle having a grip portion, a bite portion, and a carrier portion provided with at least one male coupling unit a soft protective cover provided with at least one female coupling unit for engaging with the male coupling unit and a direct digital image sensor received in a receiving cavity defined in the protective cover. Through engagement of the male and female coupling units with one another, the soft protective cover and accordingly the direct digital image sensor received therein can be connected to an end of the connection handle and be extended into a patients oral cavity without bringing too much discomfort to the patient. The used protective cover is discarded or disinfected to meet sanitary requirements in medical treatment.