Segnet is deep fully convolutional neural network architecture for semantic pixel-wise segmentation. This is implementation of http://arxiv.org/pdf/1511.00561v2.pdf (Except for the Upsampling layer where paper uses indices based upsampling which is not implemented in keras yet( I am working on it), but that shouldnt make a lot of difference). You can directly download the code from https://github.com/pradyu1993/segnet. This post is a explaination of what is happening in the code.
qianshangbushang/segnet
A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation
Jupyter Notebook