摘要
研究了一种基于支持向量机的飞机图像识别算法。采用基于神经网络的图像边缘检测方法,该方法首先基于邻域灰度极值提取边界候选图像,然后以边界候选象素及其邻域象素的二值模式作为样本集,输入边缘检测神经网络进行训练。提取具有RST不变性的轮廓特征向量,输入支持向量机进行训练和识别。将支持向量机与传统的人工神经网络的算法进行了对比实验,实验表明基于支持向量机的飞机图像识别算法具有更好的性能。
This paper studies a plane image recognition algorithm based on Support Vector Machine（SVM） . A method of edge detection, which based on neural network is adopted. The method uses a logical judgement algorithm to get edge candidate images, and then edge pixels and their neighbor pixels compose the binary samples of the BP neural network. Next extracting the outline feature vectors of the object, finally training a SVM to classify planes. An experiment showed that the algorithm based on SVM has better performance than an algorithm based on Artificial Neural Network （ANN）.
出处
《现代电子技术》
2007年第21期127-129,共3页
Modern Electronics Technique
关键词
飞机图像识别
支持向量机
特征向量
神经网络
plane image recognition
support vector machine
feature vector
neural network
