neural-network - 如何从深度 CNN 输出创建不确定性彩色 map 图像？

Question

我正在设计一个用于城市特征检测的深度 CNN 分类器。大多数情况下，我的网络会正确地对建筑物进行分类和分割，但很多时候它会因光照/外观相似等与其他对象混淆。

我想创建一个颜色图以及可以表示特定分类器的分割图像？我已经将 softmaxwith loss 用于训练网络。

layer {
  name: "score"
  type: "Deconvolution"
  bottom: "pool_3"
  top: "score"
  convolution_param {
    num_output: 2
    bias_term: false
    pad:2
    kernel_size: 8
    stride: 4
  }
}

我期待与此彩色 map 图像类似的输出:



我的问题是

如何计算不确定性？

计算不确定性时如何处理负值？

注意:目前，我可以使用熵获取颜色图。

Answer 1

您可能想要执行遮挡敏感度实验，以构建最重要图像区域的热图。
来自 this answer on AI StackExchange :

Here's the idea. Suppose that a ConvNet classifies an image as a dog. How can we be certain that it’s actually picking up on the dog in the image as opposed to some contextual cues from the background or some other miscellaneous object?

One way of investigating which part of the image some classification prediction is coming from is by plotting the probability of the class of interest (e.g. dog class) as a function of the position of an occluder object. If we iterate over regions of the image, replace it with all zeros and check the classification result, we can build a 2-dimensional heat map of what's most important for the network on a particular image.