Building Change Detection in High-Resolution Remote-Sensing Images Based on Deep Learning

Xing Han; Ling Han; Liangzhi Li; Huihui Li

doi:10.3788/LOP202259.1001003

Journals >Laser & Optoelectronics Progress >Volume 59 >Issue 10 >Page 1001003 > Article

Laser & Optoelectronics Progress
Vol. 59, Issue 10, 1001003 (2022)

Building Change Detection in High-Resolution Remote-Sensing Images Based on Deep Learning

Xing Han¹, Ling Han^2、3、*, Liangzhi Li¹, and Huihui Li¹

Author Affiliations

¹School of Geology Engineering and Geomatics, Chang’an University, Xi’an 710054, Shaanxi , China

²School of Land Engineering, Chang’an University, Xi’an 710054, Shaanxi , China

³Shaanxi Key Laboratory of Land Consolidation, Xi’an 710054, Shaanxi , China

show less

DOI: 10.3788/LOP202259.1001003 Cite this Article Set citation alerts

Xing Han, Ling Han, Liangzhi Li, Huihui Li. Building Change Detection in High-Resolution Remote-Sensing Images Based on Deep Learning[J]. Laser & Optoelectronics Progress, 2022, 59(10): 1001003 Copy Citation Text

show less

Fig. 1. CBAM structure

Download full size

Fig. 2. PPM structure

Download full size

Fig. 3. Proposed network structure

Download full size

Fig. 4. Multiscale feature fusion

Download full size

Fig. 5. Comparison of detection results of building changes with different scales. (a) First-phase remote sensing image; (b) second-phase remote sensing image; (c) ground truth; (d) UNet; (e) ChangeNet; (f) CSCDNet; (g) proposed method

Download full size

ResNet50	Size
7×7， 64， stride2	112×112
3×3， max pooling， stride2	56×56
$[\begin{matrix} \begin{array}{l} 1 \times 1 \\ 3 \times 3 \end{array} & \begin{array}{l} 64 \\ 64 \end{array} \\ 1 \times 1 & 256 \end{matrix}]$ ×3	56×56
$[\begin{matrix} \begin{array}{l} 1 \times 1 \\ 3 \times 3 \end{array} & \begin{array}{l} 128 \\ 128 \end{array} \\ 1 \times 1 & 514 \end{matrix}]$ ×4	28×28
$[\begin{matrix} \begin{array}{l} 1 \times 1 \\ 3 \times 3 \end{array} & \begin{array}{l} 256 \\ 256 \end{array} \\ 1 \times 1 & 1024 \end{matrix}]$ ×6	14×14
$[\begin{matrix} \begin{array}{l} 1 \times 1 \\ 3 \times 3 \end{array} & \begin{array}{l} 512 \\ 512 \end{array} \\ 1 \times 1 & 2048 \end{matrix}]$ ×3 （dilated convolution）	14×14

Table 1. Network structure of feature extraction stage

Lab environment	Configuration
CPU	6×Intel（R）Xeon（R）CPU E5-2678 v3@2.50 GHz
GPU	NVIDIA GeForce RTX 2080 Ti
Memory	62 GB
Operating system	Ubuntu 18.04
Deep learning framework	Pytorch1.6
Programming language	Python 3.7
GPU processing framework	CUDA 10.0， CUDNN 7.6

Table 2. Lab environment

Initial learning rate	Maximum number of iterations	F1/%
0.1	500	69.76
0.01		80.01
0.001		88.89
0.0001		90.38
0.00001		88.17
0.0001	50	85.64
	100	88.58
	200	90.17
	500	90.12

Table 3. Hyperparameters’ optimization of neural network

Method	Precision /%	Recall /%	F1 /%
Baseline	80.362	82.946	81.273
+CBAM	81.261	84.594	82.884
+PPM	80.567	83.339	81.460
Proposed method	82.508	85.232	83.324

Table 4. Ablation experiment

Method	Precision /%	Recall /%	F1 /%	Time /h
UNet	72.507	69.421	70.864	8.7
ChangeNet	74.659	70.215	72.337	9.4
CSCDNet	80.295	82.347	81.362	11.3
method Proposed	82.508	85.232	83.324	10.1

Table 5. Evaluation of change detection results of different methods

Xing Han, Ling Han, Liangzhi Li, Huihui Li. Building Change Detection in High-Resolution Remote-Sensing Images Based on Deep Learning[J]. Laser & Optoelectronics Progress, 2022, 59(10): 1001003

Download Citation

Set citation alerts for the article

Tools

Set citation alerts for the article

Save the article for my favorites

Paper Information