Bill Money Classification for Intelligent OA Systems Using Evolutional Neural Networks

使用进化神经网络的智能 OA 系统的账单金额分类

• 批准号: 09555131
• 负责人: OMATU Sigeru
• 金额: $ 4.48万
• 依托单位: Osaka Prefecture University
• 依托单位国家: 日本
• 项目类别: Grant-in-Aid for Scientific Research (B)
• 财政年份: 1997
• 资助国家: 日本
• 起止时间: 1997 至 1998
• 项目状态: 已结题
• 来源: https://kaken.nii.ac.jp/grant/KAKENHI-PROJECT-09555131/
• 关键词: Neural networks; Neuro-classification; Evolutional computation; Bill money classification; Digital signal processing; Neuro-board; 真偽判定; 階層型ニューラルネットワーク; 汎化能力

In this project, we have proposed a new approach to realize the intelligent pattern classification of bill money to enforce the office automation based on neural networks with evolution mechanism. The present project is to synthesize an intelligent pattern classifier system. To complete the project study, we have adopted the following approach to synthesize these intelligent control systems :(1) Learning Ability of the Neural networkWe have formulated the learning ability of neural networks by using information measure of the communication channel.(2) Emergence of Evolution Mechanism by Genetic AlgorithmsTo improve the learning ability of neural networks considered (1), we have introduced the genetic algorithms and find the global minimum of the learning curves.(3) Construction of Image Understanding by Neural NetworksTo realize the intelligence we have used not only numerical data but also image information processing as a multi-media data processing.(4) Hardware Implementation of Neuro-ClassifierIn order to speed up the classification and make the cost lower, we have implemented the neuro- hardware for bill money classification.

在本计画中，我们提出了一种新的方法来实现票据货币的智能模式分类，以加强办公自动化的基础上，神经网络与进化机制。本计画旨在综合一个智慧型模式分类器系统。为了完成本课题的研究，我们采用了以下方法对这些智能控制系统进行综合：（1）神经网络的学习能力我们利用通信信道的信息测度，建立了神经网络的学习能力公式。(2)为了提高考虑（1）的神经网络的学习能力，我们引入了遗传算法并找到学习曲线的全局最小值。(3)用神经网络构建图像理解为了实现智能化，我们不仅使用数字数据，而且还使用图像信息处理作为多媒体数据处理。(4)神经分类器的硬件实现为了加快分类速度并降低成本，我们实现了用于票据货币分类的神经硬件。

项目成果

姜 東植: "紙幣音響データによる新旧紙幣のニューロ識別" 電気学会論文誌C分冊. 118-C・12. 1706-1711 (1998)

Dong-sik Kang：“使用钞票声学数据对新旧钞票进行神经识别”，日本电气工程师学会汇刊，卷 C. 118-C・12 1706-1711（1998 年）。

姜 東植: "二つの適応ディジタルフィルタを用いた信号検出" Journal of Signal Processing. 2・2. 157-165 (1998)

Dong-sik Kang：“使用两个自适应数字滤波器进行信号检测”信号处理杂志 2・2。

Minoru Fukumi: "Rotation-Invariant Neural Pattern Recognition System Estimating a Rotation Angle" IEEE Transactions on Neural Networks.8・3. 568-581 (1997)

Minoru Fukumi：“估计旋转角度的旋转不变神经模式识别系统”IEEE Transactions on Neural Networks.8・3（1997）。

Yosuke Ito: "Polarimetric SAR Data Classification Using Neural Networks" Journal of the Japan Society of Photogrammetry and Remote Sensing. Vol.36-No.3. 13-22 (1997)

Yosuke Ito：“使用神经网络进行极化 SAR 数据分类”日本摄影测量与遥感学会杂志。

Masaru Teranishi: "New and Used Bill Money Classification Using Spectral Information Based on Acoustic Data of Banking Machine" Transactions of IEE of Japan. Vol.117-C, No.11. 1677-1681 (1997)

Masaru Teranishi：“基于银行机声学数据的频谱信息对新旧钞票进行分类”日本 IEE 交易。