Building extraction in pixel level from aerial imagery with a deep encoder-decoder network 基于编解码网络的航空影像像素级建筑物提取

Engineering

• Marginals 100%
• Pixel Level 100%
• Convolutional Neural Network 60%
• Saturates 40%
• Input Image 40%
• Croppings 40%
• Raw Image 40%
• Experimental Result 20%
• Context Information 20%
• Land Use 20%
• Classical Method 20%
• Urban Planning 20%

Keyphrases

• Deep Encoder-decoder Network 100%
• Extraction Accuracy 13%
• Professional Knowledge 6%
• Encoder-decoder Structure 6%
• Land Use Analysis 6%