# 1-MSDU-Net

解决什么问题？

A blur detection problem which aims to separate the blurred and clear regions of an image。

模糊检测问题，其目标是将图像中模糊和清晰的区域分开。

用的什么方法实现？

To improve the quality of the image separation,extracting features from various scales of images.

为了提高图像分离的质量，从不同尺度的图像中提取特征。

Inspired by the success of the U-net architecture for image segmentation, we propose a multi-scale dilated convolutional neural network called MSDU-net.

受成功的U-net图像架构的启发，提出了一种多尺度扩展卷积神经网络，称为MSDU-net。

存在的困难是什么？

1.the matter of how to extract blur features。

2.and fuse these features synchronously is still a big challenge。

1.如何提取模糊特征的问题。

2.同时融合这些功能仍然是一个很大的挑战。

网络的创新点是什么？

In this model, we design a group of multi-scale feature extractors with dilated convolutions to extract textual information at different scales at the same time.

在该模型中，我们设计了一组带有扩展卷积的多尺度特征提取器进行提取同一时间不同尺度的文本信息。

The U-shape architecture of the MSDU-netMSDU-net can fuse the different-scale texture features and generated semantic features to support the image segmentation task.

U型结构能否融合不同尺度的纹理特征并生成语义特征来支持图像分割的任务。

# Image-Adaptive

Image-Adaptive YOLO for Object Detection in Adverse Weather Conditions

图像自适应YOLO在恶劣天气条件下的目标检测

解决了什么问题？

it is still challenging to locate objects from the low-quality images captured in adverse weather conditions.

从恶劣天气条件下拍摄的低质量图像中定位目标仍然是一个挑战。

存在什么问题？

The existing methods either have difficulties in balancing the tasks of image enhancement and object detection, or often ignore the latent information beneficial for detection.

现有的方法要么难以平衡图像增强和目标检测的任务，要么经常忽略有利于检测的潜在信息。

如何解决这个问题？

To alleviate this problem, we propose a novel Image-Adaptive YOLO (IA-YOLO) framework, where each image can be adaptively enhanced for better detection performance. 

为了解决这个问题，我们提出了一种新的图像自适应YOLO (Image-Adaptive YOLO)框架，其中每个图像可以自适应增强更好的检测性能。

论文创新点是什么？

Specifically, a differentiable image processing (DIP) module is presented to take into account the adverse weather conditions for YOLO detector, whoseYOLO parameters are predicted by a small convolutional neural network (CNN-PP).

具体来说，一个可微的图像处理(DIP)模块提出了考虑探测器在恶劣的天气条件下参数由一个小型卷积神经网络(CNN-PP)预测。