论文部分内容阅读
视觉研究中,“视觉”在大脑进行图象分析之前只是“录像机”的作用的传统概念被完全打破了。实际上图像在进入大脑前已进行了100亿次/秒的运算及分析识别。在这个过程中,各种“功能块”的综合效应,包括过去经验知识的参与,均有重要作用。这方面的研究已与数学结合起来,成了计算机(人工智能)的一个领域,意义重大,前景可观。本文对这方面的研究、进展、展望作了评述。另本刊前此也有过几篇有关的文章,可参阅。
In visual research, the traditional notion of “visual” as a “video recorder” before the brain conducts image analysis is completely broken. In fact, the image before the brain has been carried out 10 billion times / second operation and analysis of recognition. In this process, the comprehensive effect of various “function blocks”, including the participation of past empirical knowledge, plays an important role. Research in this area has been combined with mathematics and has become an area of computer (artificial intelligence) that is both significant and promising. This article reviews the research, progress and outlook in this field. Another article before this there have been a few articles, you can refer to.