基于神经网络的海面溢油图像分类算法

A marine spill oil classification method based on neural network

提出了一种基于Gabor变换、KPCA和神经网络的图像分类方法。首先对图像进行Gabor滤波,获得不同方向的特征参数;然后提取图像的KPCA作为图像的特征,最后利用神经网络进行分类。通过对实验分类结果的定量分析可知,该方法可以获得精度比最小分类模型方法以及最大似然分布模型方法高的分类结果。

A marine spill oil image classification method based on gabor transformation, KPCA and neural network is put forward. First of all, extract the characteristic parameters in different directions of the image after the Gabor filter, The KPCA is used to describe the feature of the image, make classification of the feature vector based on neural network. Through the quantitative analysis of the results, the proposed method can gain high accuracy constrast to the minimum classification model method and the maximum likelihood distribution model method.