A Study Of Visual Prosthetic Devices Modeling And Visual Information Processing

Induced by a variety of retinopathy, visual loss has become the most serious form of disability which influences the quality of human life. With the rapid development and crossing among the information science, microelectronics, material science and biomedical disciplines, the visual prosthesis makes the visual reparation possible, which blindness caused by retinitis pigmentosa, age-related macular degeneration, and other eye, retina, optic nerve and visual cortex lesions. However, because the complex structure of the human visual system, it is complicated to produce the corresponding phosphenes for blind patients by impulse current stimulation on the optic nerves with the finite density of micro-current stimulator electrode array. The conflict thus formed that the objective world having a complex scene image versus a limited array of micro-motor stimulation contradiction. In addition, because of the manufacture limit for microelectrode array, at this stage, the scope of visual prosthesis vision reparation by visual prosthesis is very limited. Hence, it is required that visual prosthetic devices can flexibly rotatie under the will of humans like the real human eyes.In order to solve these problems above, we start from the least visual information needs to carry out the visual prosthetic device modeling and the visual information research. We design the visual prosthetic eye movement devices as a biological, mechanical, electronic integrated model. And we construct the psychophysical experiment to assess our work. The major study includes as the following: (1) This dissertation imitates the human visual system structure and performance, and made an integrated design for visual prosthesis devices by studying the principles of human visual system. From biomimetic point of view, a simplified visual prosthesis eye movement device is established, to achieve the device movement according to the eye movement under the will of human. In addition, from the perspective of robotics, kinematics mathematical modeling for visual prosthesis eye moviement device is established, which reveals the kinematics law for the simplified visual prosthesis device. The graphics-based alignment of visual location information acquisition algorithm is also proposed in order to gain the visual location information for visual prosthesis.(2) As the microelectrode manufacture and operation restriction for visual prosthesis, the array of implanted microelectrode to stimulate the optic nerve is limited. Therefore, it is necessary to process the micro-camera image through minimal visual information requirement segmentation to extract the edges to map as little as possible but enough visual information to the micro-motor stimulation array. Hence, we propose an algorithem started from the minimum visual information to determine the number of clusters and clustering algorithm design to solve these image segmentation issues, in order to acquire the visual information.(3) This dissertation proposed an assessment method based on psychophysical experiments according to the the actual simulated prosthetic vision in pixelized representation. A psychophysical experiment platform is built and the results are evaluated. Psychophysical experiments can perform on normal or partial vision remaining volunteers to not only study their perceived degree of visual adaptation to the environment with visual prosthesis, but also can be used to describe the early period of the prosthesis users'perception of the phenomenon, in order to better evaluate our methods.(4) This dissertation built a semi-physical simulation system for visual information processing, showing the functions and simulated procedure of overall framework of information processing systems of and each subsystem module of the semi-physical system. In addition, the subsystem of simulated phosphene psychophysics experiments is also used in our psychophysical experimental study to assess the effct of simulated visual phosphenes information we have obtained on the people.Our work will set up the foundation of the visual prosthesis in the field of information processing and coding. In addition, it will provide innovative design ideas for the visual prosthesis design based on biological, mechanical and electronic integration.