Multispectral Pedestrian Detection using Deep Fusion Convolutional Neural Networks （深度学习多光谱行人检测综述）

最新推荐文章于 2025-07-07 10:40:32 发布

ylin01234

最新推荐文章于 2025-07-07 10:40:32 发布

阅读量2.9k

点赞数

CC 4.0 BY-SA版权

分类专栏：论文

本文链接：https://blog.youkuaiyun.com/ylin01234/article/details/81106203

当前的行人检测方法大多依赖于光照良好的彩色图像，但在夜间或能见度差的环境中表现不佳，限制了其在自动驾驶和监控系统等全天候应用中的使用。通过将可见光相机与长波红外信息融合，可以利用颜色和热图像的互补视觉信息来改善这种情况。本文研究了如何通过深度神经网络融合颜色和热图像进行行人检测，并探讨了不同融合层次（像素级、特征级、决策级）的效果，发现中层卷积特征融合在保留语义信息的同时，也保持了视觉细节，从而提高了检测性能。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

Now, salient detection methods most of current pedestrian detectors explored color images of good lighting, and they are very likely to be stuck with images captured at night, due to bad visibility of object. Such defect would cut these approaches off from the around-the-clock applications, e.g., self-driven car and surveillance system. In some sense, color and thermal image channels provide complementary visual information, so it is helpful to fuse the information of a visible camera with the information provided by a long-wavelength infrared.

We want to take advantage of color and thermal to detect object by deep Neural networks. Now, we must answer the following two questions: when strong detectors getting by neural networks are involved, does color and thermal images still provide complementary information? To what extend the improvement should be expected by fusing them together? So, we can model the multisp