可应用于动物实验的人工视觉系统

摘要

为满足人工视觉在动物临床实验上的特定需求，设计以 DM6437为硬件平台的人工视觉一体化系统。为解决视觉信息量大和有限的植入体电极数目之间的矛盾，提出基于块分割和灰度分级的降像素化视觉图像处理方法，建立灰度等级和刺激电流幅度相关联的电极刺激策略。进行系统空域响应测试，测试结果表明，一对一的电极刺激准确率达到95％，在功耗一定的条件下优化后的系统图像处理速度可达30 fps，满足视觉中枢系统的实时性要求。%To meet specific needs of the artificial vision clinical experiments on animal,a DM6437-based integrated system was designed.To solve the conflict between the huge visual information and a limited array of implant electrode,a reducing visual image pixel method based on block partition and graded gray was presented.Meanwhile,an electrode stimulation strategy which associated gray levels with stimulus current amplitude was constructed.Finally,the results of test on space responses show that one to one electrode stimulation accuracy reaches 95% and when the system at certain power consumption,it achieves 30 fps pro-cessing speed after optimization,which meets real-time processing requirements of visual central system.