摘要
重点在于应用基于三阶相关量的人工神经网络方法解决目标图像的平移 (Translation)、尺度变换 (Scale)、旋转 (Rotationin plane)不变性 (Invariant)识别问题 (TSRI)。该方法通过计算“零片”三阶相关量 ,对数极坐标变换和计算Fourier变换幅度谱 ,得到TSRI特征 ,并作为模式分类器的输入信号 ;模式分类器采用单层感知机。理论分析和对 3类飞机目标的实验结果表明 ,该法具有抗噪声 ,高识别率 ,处理速度快 ,结构简单等特点 。
The emphasis of this paper is laid on solving translation, scale and rotation-invariant target image recognition problems(TSRI) using ANN method based on triple correlation. By calculating ‘0 slice’ triple correlation, log polar mapping, calculating Fourier transform amplitude module, the TSRI feature is retrieved and used as input signal of pattern classifier, the single layer perception is used as a classifier. Theory analyzing and results from experiment with three airplanes show that this TSRI recognition method is efficient, and has many advantages such as insensitive to noise, high recognition rate, high speed processing, simple in structure etc.
出处
《兵工学报》
EI
CAS
CSCD
北大核心
1999年第1期39-42,共4页
Acta Armamentarii
关键词
图像识别
不变性模式识别
TSRI
飞机
目标识别
image recognition, artificial neural network, automatical target recognition, invariant pattern recognition
