[{"createTime":1735734952000,"id":1,"img":"hwy_ms_500_252.jpeg","link":"https://activity.huaweicloud.com/cps.html?fromacct=261f35b6-af54-4511-a2ca-910fa15905d1&utm_source=V1g3MDY4NTY=&utm_medium=cps&utm_campaign=201905","name":"华为云秒杀","status":9,"txt":"华为云38元秒杀","type":1,"updateTime":1735747411000,"userId":3},{"createTime":1736173885000,"id":2,"img":"txy_480_300.png","link":"https://cloud.tencent.com/act/cps/redirect?redirect=1077&cps_key=edb15096bfff75effaaa8c8bb66138bd&from=console","name":"腾讯云秒杀","status":9,"txt":"腾讯云限量秒杀","type":1,"updateTime":1736173885000,"userId":3},{"createTime":1736177492000,"id":3,"img":"aly_251_140.png","link":"https://www.aliyun.com/minisite/goods?userCode=pwp8kmv3","memo":"","name":"阿里云","status":9,"txt":"阿里云2折起","type":1,"updateTime":1736177492000,"userId":3},{"createTime":1735660800000,"id":4,"img":"vultr_560_300.png","link":"https://www.vultr.com/?ref=9603742-8H","name":"Vultr","status":9,"txt":"Vultr送$100","type":1,"updateTime":1735660800000,"userId":3},{"createTime":1735660800000,"id":5,"img":"jdy_663_320.jpg","link":"https://3.cn/2ay1-e5t","name":"京东云","status":9,"txt":"京东云特惠专区","type":1,"updateTime":1735660800000,"userId":3},{"createTime":1735660800000,"id":6,"img":"new_ads.png","link":"https://www.iodraw.com/ads","name":"发布广告","status":9,"txt":"发布广告","type":1,"updateTime":1735660800000,"userId":3},{"createTime":1735660800000,"id":7,"img":"yun_910_50.png","link":"https://activity.huaweicloud.com/discount_area_v5/index.html?fromacct=261f35b6-af54-4511-a2ca-910fa15905d1&utm_source=aXhpYW95YW5nOA===&utm_medium=cps&utm_campaign=201905","name":"底部","status":9,"txt":"高性能云服务器2折起","type":2,"updateTime":1735660800000,"userId":3}]
*
这是暗图增强领域一篇经典的传统方法论文,发表在TIP这个顶刊
*
文章基于的是这样一个公式:
L = R ⋅ T L=R\cdot T L=R⋅T
其中, L L L是暗图, R R R是反射分量, T T T是illumination
map,并且对于彩色图像来说,三通道都共享相同的illumination map。我们可以使用各种方法估计 T T T,又已知 L L L,则可以得到反射分量
R = L / T R=L/TR=L/T,并认为反射分量就是增强结果,即亮图。
*
但文章认为,直接用反射分量 R R R来作为增强结果不太合适,因为反射分量失去了形状信息。其实 R R R就相当于将 T T T修改为全1时的 L L L。当
T TT不为全一时对应的 L L L会是一个更好的增强结果,也即预测一个修改后的 T T T( T ^ \hat T T^),用这个 T ^ \hat T T
^来得到 R ^ = L / T ^ \hat R=L/\hat T R^=L/T^,这个 R ^ \hat R R^会是更好的增强结果。总之,就是找到一个好的
T TT,使得 L / T L/T L/T是一个好的增强结果。
*
采用的优化目标目标如下:
*
即首先不能离初始化的T太远,初始化的T即暗图各通道最大值;然后是T要平滑。W矩阵取决于先验策略。
*
上述优化问题的两项都有T,虽然可以通过梯度下降法等通用方法求局部最优解,但是本文用的不是这种方法。文章将上述优化问题用G代替 ∇ T \nabla T ∇T
,变成如下带约束的优化问题:
*
然后用拉格朗日方程转化为如下优化问题:
*
可以用ALM优化方法来求最优值,即迭代交替求 T , G , Z , μ T,G,Z,\mu T,G,Z,μ的最优值。具体每一步的公式推导这里就不展开了。
*
W W W矩阵可以是如下三种形式之一:
或
或
*
根据算法估计出T后,可以用 L / T L/T L/T
得到增强结果。但此时的增强结果太亮了,可以用gamma校正把分母的T增大,使得增强结果稍微暗一点,文章设的gamma值为0.8。
*
进一步的,为了去除噪声,可以用BM3D算法对增强结果去噪,文章先对RGB的增强结果转到YUV色彩模式,然后对Y通道做去噪再转回RGB,然后把去噪前后的结果利用T进行重组以避免强亮度区域过模糊:
其中 R d R_d Rd即为增强结果
*
从实验结果上看,LIME的处理速度还是很快的,只需要100次迭代即可收敛,0.78s一张图片。