Intraocular vision implants offer potential cure for several 100,000 blind people, including blind patients suffering from corneal opacity, accidents, fire, explosion, and alkali burn. Wireless, fully implantable, intraocular vision implant systems are capable of applying optical stimulation patterns to the photoreceptor layer of blind patients' retinas. The optical stimulation patterns are derived in real-time from a live video feed. This paper presents the essential component of these intraocular stimulation systems - the digital signal processing that emulates the healthy eye and replaces the defective parts of the blind patients' eyes. The implementation is based on the comprehensive mathematical description of the photoreceptor encoding process. During this process, the live video feed is transformed in real-time by the photoreceptor encoding function to implement the temporal and spatial properties of pigment bleaching, horizontal cell feedback, Michaelson-Menton transduction, cone response, and inhibitory horizontal cell feedback. To customize the stimulation data for transmission to the micro-display, the source coding process also performs spatial downsampling and luminance dynamics reduction.