当前位置: X-MOL 学术Pattern Anal. Applic. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Encoding features from multi-layer Gabor filtering for visual smoke recognition
Pattern Analysis and Applications ( IF 3.9 ) Pub Date : 2020-02-14 , DOI: 10.1007/s10044-020-00864-x
Feiniu Yuan , Gang Li , Xue Xia , Jinting Shi , Lin Zhang

It is a challenging task to accurately recognize smoke from visual scenes due to large variations in smoke shape, color and texture. To improve recognition accuracy, we propose a framework mainly with a robust local feature extraction module based on Gabor convolutional networks. We first propose a Gabor convolutional network, each layer of which mainly consists of Gabor convolution and feature fusion. To fuse feature maps generated by Gabor convolution, we present three Basic Grouping Methods, which divide the feature maps into several groups along orientation axis, scale axis and both of them. To avoid exponential growth of feature maps and preserve discriminative information simultaneously, we propose three element-wise aggregation functions, which are mean, min and max, to combine feature maps in each group. To further improve efficiency, we use local binary pattern to encode hidden and output maps of Gabor convolutional layers. In addition, we use a weight vector to optimize concatenation of histograms for further improvement. Experiments show that our method achieves very outstanding results on smoke, texture and material image datasets. Although the feature extraction step of our method is training-free, our framework consistently outperforms state-of-the-art methods on small smoke datasets, even including deep learning-based methods.

中文翻译:

多层Gabor过滤的编码功能可用于视觉烟雾识别

由于烟雾形状,颜色和纹理的巨大差异,从视觉场景中准确识别烟雾是一项艰巨的任务。为了提高识别精度,我们提出了一个框架,该框架主要基于基于Gabor卷积网络的鲁棒局部特征提取模块。我们首先提出一个Gabor卷积网络,其每一层主要由Gabor卷积和特征融合组成。为了融合由Gabor卷积生成的特征图,我们提出了三种基本分组方法,这些方法将特征图沿方向轴,比例轴和这两者分为几组。为了避免特征图的指数增长并同时保留判别式信息,我们提出了三个元素平均聚合函数(均值,最小值和最大值),以将特征图组合在每个组中。为了进一步提高效率,我们使用本地二进制模式对Gabor卷积层的隐藏和输出图进行编码。另外,我们使用权重向量来优化直方图的级联以进一步改善。实验表明,我们的方法在烟雾,纹理和材质图像数据集上取得了非常出色的结果。尽管我们方法的特征提取步骤无需培训,但我们的框架在小型烟雾数据集上始终优于最新方法,甚至包括基于深度学习的方法。
更新日期:2020-02-14
down
wechat
bug