Identification of degraded traffic sign symbols by a generative learning method
Identification of degraded traffic sign symbols by a generative learning method
复制标题
DOI：
10.1109/icpr.2006.674
复制
发表时间：
2006-08
期刊：
影响因子：
--
通讯作者：
H. Ishida;Tomokazu Takahashi;I. Ide;Y. Mekada;H. Murase
中科院分区：
文献类型：
--
作者：
H. Ishida;Tomokazu Takahashi;I. Ide;Y. Mekada;H. Murase
我们提出了一种新的训练方法识别交通标志符号经历图像退化。为了科普退化，期望使用类似退化的图像作为训练数据。我们的方法根据实际的退化从原始图像人为地生成这些数据。实验结果表明，该方法在基于摄像头的交通标志识别中具有良好的实用性
We present a novel training method for recognizing traffic sign symbols undergoing image degradations. In order to cope with the degradations, it is desirable to use similarly degraded images as training data. Our method artificially generates these data from an original image in accordance with the actual degradations. We experimentally confirmed the usefulness of our method for the camera-based traffic sign recognition