Applications of Artificial Neural Network to Opto Electric Measurements
-
摘要: 人工神经网络不仅被广泛用于人工智能、自动控制、机械人、统计学等领域的信息处理中,还越来越多地被应用于光电检测.综述了人工神经网络技术在光电检测中的最新应用进展,重点讨论了人工神经网络技术在谷物蛋白质含量近红外在线监测、结构光三维视觉检测、焊缝缺陷射线实时成像识别、表面粗糙度及球半径精密测量应用中的优点和局限性,以期进一步推动本领域的研究和发展.Abstract: Artificial neural network technology is getting wider and wider scale application. Following a brief introduction on neural network, its recent applications to optoelectric measurements were described in detail, including online monitoring grain protein based on NIR, structured light based 3D vision inspection, X ray real time detection of welding seam defect, and precision measurements of surface roughness and sphere radius. By summarizing its advantages and limitations in these applications, suggestions and comments were given for further research efforts in this field.
-
Key words:
- neural networks /
- defect inspection /
- surface roughness /
- protein content /
- vision inspection /
- sphere radius
-
[1] Metropolis N, Rosenbluth A, Rosenbluth M, et al. Equation of state calculations by fast computing machines[J]. Journal of Chemical Physics, 1953,21(6):1087~1092. [2] Kirkpatric S,Jr C D G, Vecchi M P. Optimization by simulated annealing. Science, 1983,220(5):671~679. [3] Dekkers A, Aarts E. Global optimization and simulated annealing[J]. Mathematical Programming, 1991,50:367~393. [4] 马颂德,张正友.计算机视觉[M].北京:科学出版社,1998. -
点击查看大图
计量
- 文章访问数: 2373
- HTML全文浏览量: 143
- PDF下载量: 795
- 被引次数: 0
下载:
