摘要
根据工程需求研究了基于投影特征的图像识别方法。根据生产线实际,不需要对图像进行预处理,直接对图像进行八个方向投影,提取图像特征压缩。采用神经网络对图像进行识别。对投影进行归一化特征提取,自适应共振型ART-1神经网络通过学习训练对模板图像特征具有记忆,选择能够代表模板图像特征的神经元I;当输入待识别图像特征时,网络选择能够代表它的神经元I_2,如果神经元I和I2相同,则认为待识别图像与模板图像相同;否则,认为待识别图像与模板图像不同。识别误差和警戒参数ρ的设置有关。较好地实现了对图像的识别。
According to engineering requirements, the image recognition method based on projection featuresis researched. Based on the actual production line without preprocessing, the image is projected to eight directionsdirectly and image feature compression is extracted. Neural network is adopted to recognize the image and the pro-jection is performed normalized feature extraction. Adaptive resonance ART-1 neural network can remember tem-plate image features through learning and training, so neuron I representing template image features is chosen.When the features of the image to be identified are input, neuron I_2 is represented by network choice. If neuron Iand I_2 are same, it is thought that the image to be identified is the same as the template image, and otherwise it isthought that the image to be identified is different from the template image. Recognition error is related with the set-ting of vigilance parameter ρ.And image recognition is better achieved in the way.
出处
《光电技术应用》
2015年第6期51-55,共5页
Electro-Optic Technology Application
基金
高等学校博士学科点专项科研基金项目(博导类)(20121420110006)
山西省回国留学人员科研资助项目(2013-083)
山西省高等学校优秀创新团队支持计划资助
关键词
投影
归一化
特征提取
神经网络
projection
normalization
feature extraction
neural network
