一种改进的暗原色单幅图像去雾方法

暗原色先验方法可以较好地处理单幅图像去雾,但对理图像中的灰白区域处理效果不好。通过分析暗原色先验原理,得出了暗通道图像和雾的透射分布率以及雾的浓度系数的关系,提出了一种结合峰值信噪比和暗原色优先法则的去雾方法。由实验结果分析,该方法能够更清晰的表现去雾后图像的细节,并且一定程度上克服了原方法在处理图像中灰白区域效果不佳的弱点。     

改进残差神经网络的图像去雾算法

针对图像去雾算法中存在因介质透射率估计不准确而造成色彩失真、去雾不完全的问题,提出了一种改进残差神经网络的图像去雾算法.首先采用并行多尺度卷积层提取雾图像特征.然后通过引入了深度可分离卷积层的残差网络学习介质透射率,并利用加权引导滤波细化介质透射率.最后根据大气散射模型反演得到无雾清晰图像.实验结果表明,该算法与其他去雾算法相比在峰值信噪比(peak signal to noise ratio, PSNR)和结构相似度(structural similarity, SSIM)指标上取得了一定的提高,并且去雾图像在主观视觉上也取得了较好表现.     

基于霾层学习的单幅图像去雾算法

针对传统去雾算法出现色彩失真、去雾不完全、出现光晕等现象,本文提出了一种基于霾层学习的卷积神经网络的单幅图像去雾算法.首先,依据大气散射物理模型进行理论推导,本文设计了一种能够直接学习和估计有雾图像和霾层图像之间的映射关系的网络模型.采用有雾图像作为输入,并输出有雾图像与无雾图像之间的残差图像,随后直接从有雾图像中去除此霾层图像,即可恢复出无雾图像.残差学习的引入,使得网络来直接估计初始霾层,利用相对大的学习率,减少计算量,加快收敛过程.再利用引导滤波进行细化,使得恢复出的无雾图像更接近真实场景.本文对不同雾浓度的有雾图片的去雾效果进行测试,并与当前主流深度学习去雾算法及其他经典算法进行对比.实验结果显示,本文设计的卷积神经网络模型在图像去雾的应用,不论在主观效果还是客观指标上,都有优势.     

基于半逆法和暗原色先验的单一图像去雾新方法

针对现有图像去雾方法的缺陷,依据雾天成像的理论模型,提出了一种基于半逆法和暗原色先验相结合的单一图像去雾新方法.对一幅待去雾图像,首先通过半逆法检测并标记出图像浓雾区域,结合暗原色先验,随机多次选取浓雾区域中具有较大暗原色值的候选像素点集,并通过计算原图中这些候选点集的像素点平均亮度值估算大气光常数.其次通过暗原色先验信息获取图像像素点的透射率分布.最后在得到大气光常数和透射率参数的基础上,根据雾天成像的光学理论模型反解出去雾后图像.通过图像去雾的无参考客观质量评价(NQA)体系对有效细节强度、色调还原程度和结构信息三项指标进行定量评估,评估结果表明,相比目前去雾效果最好的He算法,这一新方法平均能提高了15％.此外,该方法能减少去雾后图像的halo效应,更能满足人们的主观观测需求.     

单幅雾天图像复原

 本文提出了一种单幅雾天图像分割复原的方法,首先利用黑体原理估计粗糙的大气光传输图,然后使用拉普拉斯修补矩阵对分割之后的传输图进行修补,当场景目标和大气光很类似时,采用置信传播推断算法纠正传输图,经过修补和推断之后的传输图能够准确反映光线通过雾的传输过程,结合雾天图像光学模型,从雾的物理特性上去除雾对图像的影响.实验结果表明,本文提出的雾天图像分割复原算法能够有效恢复出清晰图像,并能获取相应雾天图像深度信息.     

基于深度学习的卫星图像识别分类方法

卫星遥感技术是一种非常重要的地球空间监测技术.卫星遥感图像经过处理后具有数据量大和数据类型复杂多样的特点,传统方法进行识别分类耗费大量人力物力.为了降低工作量,并为后续处理提供便利,本文将深度学习算法应用于卫星图像的识别分类中,设计了一种基于VGGNet的识别分类方法,利用除雾算法对训练数据进行数据增强处理,并添加岭回归正则化层,利用标签之间的相关性进行预测,使得方法达到90%以上的F2 score,并在实验部分进行了对比验证.最后利用此方法搭建了一个基于Django的在线识别分类展示系统.     

Single image visibility restoration using optical compensation and pixel-by-pixel transmission estimation

Aiming at image degradation in hazy and sandstorm weather,an optical compensation color restoration and pixel-by-pixel transmissvity estimation algorithm was proposed.The blue light was absorbed by sandstorm particles.The color shift phenomenon could be eliminated by optical compensate method,which convert the sandstorm images into hazy images.Then the ratio relationship between the minimum channel and its Gaussian function as the transmissivity,and median filter was used to eliminate its texture effects.The depth of the restored transmissivity alternated obviously and the edge was well preserved,which did not need the time-consuming postprocessing operativity.Finally,the image was restored by the atmospheric scattering model.The experimental results show that recovered sandstorm image treatment is better,and the saturation of the haze image is appropriate,the bright area is more nature,and running time is faster.     

AIDEDNet: anti-interference and detail enhancement dehazing network for real-world scenes

The haze phenomenon seriously interferes the image acquisition and reduces image quality. Due to many uncertain factors, dehazing is typically a challenge in image processing. The most existing deep learning-based dehazing approaches apply the atmospheric scattering model (ASM) or a similar physical model, which originally comes from traditional dehazing methods. However, the data set trained in deep learning does not match well this model for three reasons. Firstly, the atmospheric illumination in ASM is obtained from prior experience, which is not accurate for dehazing real-scene. Secondly, it is difficult to get the depth of outdoor scenes for ASM. Thirdly, the haze is a complex natural phenomenon, and it is difficult to find an accurate physical model and related parameters to describe this phenomenon. In this paper, we propose a black box method, in which the haze is considered an image quality problem without using any physical model such as ASM. Analytically, we propose a novel dehazing equation to combine two mechanisms: interference item and detail enhancement item. The interference item estimates the haze information for dehazing the image, and then the detail enhancement item can repair and enhance the details of the dehazed image. Based on the new equation, we design an anti-interference and detail enhancement dehazing network (AIDEDNet), which is dramatically different from existing dehazing networks in that our network is fed into the haze-free images for training. Specifically, we propose a new way to construct a haze patch on the flight of network training. The patch is randomly selected from the input images and the thickness of haze is also randomly set. Numerous experiment results show that AIDEDNet outperforms the state-of-the-art methods on both synthetic haze scenes and real-world haze scenes.     

海天背景图像去雾方法研究

海天背景图像存在大面积天空区域,且在远距离平视时目标一定出现在海天线附近,现有的去雾算法对天空区域的改善都是弱化天空区域的处理,这样势必会造成海天线附近去雾效果的减弱,不利于后续的目标检测.针对该问题,提出一种基于大气散射模型的图像复原去雾算法.首先,利用海天背景图像的特点,采用边缘检测算子将图像划分为天空和非天空区域,并结合大气光的物理意义,取天空区域最大的值作为大气光的估计值;其次,针对有雾图像对比度很低而无雾图像对比度较高这一先验信息设计代价函数,并通过SLIC超像素分割进行分块,通过求解每个小块内该函数的最小值,估计出粗透射率,再用引导滤波对粗透射率进行细化从而消除块效应;最后,利用大气散射模型,代入前两步求得的参数便可以得到恢复的无雾图像.实验结果与分析结果表明,本文能对海天背景的图像取得较好的去雾效果.     

基于视觉物理模型的去雾算法

针对现有图像去雾算法不能有效增强复杂大气环境下退化图像的问题,结合单色 大气散射模型、大气传输函数(ATF)以及Retinex 提出了一种基于视觉物理模型(VPM)的图像去 雾算法。新模型可同时描述非均匀光照退化、雾霾退化以及噪声退化等复杂大气环境下的图像 退化。模型求解过程首先使用变分法消除环境光退化,然后引入马尔科夫随机场将场景反射率 求解问题转换为了最大后验概率问题,最后利用对比度抑制自适应直方图均衡来校正场景反射 率亮度,从而实现图像去雾。实验结果表明VPM 能够指复杂大气环境下退化图像的增强,使 其物理保真度和视觉愉悦性得到有效改善。