Abstract

Feature extraction is a key step in hyperspectral image change detection. However, many targets with great various sizes, such as narrow paths, wide rivers, and large tracts of cultivated land, can appear in a satellite remote sensing image at the same time, which will increase the difficulty of feature extraction. In addition, the phenomenon that the number of changed pixels is much less than unchanged pixels will lead to class imbalance and affect the accuracy of change detection. To address the above issues, based on the U-Net model, we propose an adaptive convolution kernel structure to replace the original convolution operations and design a weight loss function in the training stage. The adaptive convolution kernel contains two various kernel sizes and can automatically generate their corresponding weight feature map during training. Each output pixel obtains the corresponding convolution kernel combination according to the weight. This structure of automatically selecting the size of the convolution kernel can effectively adapt to different sizes of targets and extract multi-scale spatial features. The modified cross-entropy loss function solves the problem of class imbalance by increasing the weight of changed pixels. Study results on four datasets indicate that the proposed method performs better than most existing methods.

© 2023 Optica Publishing Group

Hyperspectral deep convolution anomaly detection based on weight adjustment strategy

Dan Chong, Bingliang Hu, Xiaohui Gao, Hao Gao, Pu Xia, and Yinhua Wu
Appl. Opt. 59(31) 9633-9642 (2020)

Multikernel positional embedding convolutional neural network for photoacoustic reconstruction with sparse data

Jiayi Li and Yi-Chao Meng
Appl. Opt. 62(32) 8506-8516 (2023)

Multi-kernel driven 3D convolutional neural network for automated detection of lung nodules in chest CT scans

Ruoyu Wu, Changyu Liang, Jiuquan Zhang, QiJuan Tan, and Hong Huang
Biomed. Opt. Express 15(2) 1195-1218 (2024)

Implementation of energy-efficient convolutional neural networks based on kernel-pruned silicon photonics

Duan Huang, YanTing Xiong, Zhuangzhuang Xing, and Qi Zhang
Opt. Express 31(16) 25865-25880 (2023)

An end-to-end laser-induced damage change detection approach for optical elements via siamese network and multi-layer perceptrons

Jingwei Kou, Tao Zhan, Li Wang, Yu Xie, Yihui Zhang, Deyun Zhou, and Maoguo Gong
Opt. Express 30(13) 24084-24102 (2022)