An Efficient Lightweight Neural Network for Remote Sensing Image Change Detection

Remote sensing (RS) image change detection (CD) is a critical technique of detecting land surface changes in earth observation. Deep learning (DL)-based approaches have gained popularity and have made remarkable progress in change detection. The recent advances in DL-based methods mainly focus on enhancing the feature representation ability for performance improvement. However, deeper networks incorporated with attention-based or multiscale context-based modules involve a large number of network parameters and require more inference time. In this paper, we first proposed an effective network called 3M-CDNet that requires about 3.12 M parameters for accuracy improvement. Furthermore, a lightweight variant called 1M-CDNet, which only requires about 1.26 M parameters, was proposed for computation efficiency with the limitation of computing power. 3M-CDNet and 1M-CDNet have the same backbone network architecture but different classifiers. Specifically, the application of deformable convolutions (DConv) in the lightweight backbone made the model gain a good geometric transformation modeling capacity for change detection. The two-level feature fusion strategy was applied to improve the feature representation. In addition, the classifier that has a plain design to facilitate the inference speed applied dropout regularization to improve generalization ability. Online data augmentation (DA) was also applied to alleviate overfitting during model training. Extensive experiments have been conducted on several public datasets for performance evaluation. Ablation studies have proved the effectiveness of the core components. Experiment results demonstrate that the proposed networks achieved performance improvements compared with the state-of-the-art methods. Specifically, 3M-CDNet achieved the best F1-score on two datasets, i.e., LEVIR-CD (0.9161) and Season-Varying (0.9749). Compared with existing methods, 1M-CDNet achieved a higher F1-score, i.e., LEVIR-CD (0.9118) and Season-Varying (0.9680). In addition, the runtime of 1M-CDNet is superior to most, which exhibits a better trade-off between accuracy and efficiency.

Analysis of the computational strategy of a detailed laminar cortical microcircuit model for solving the image-change-detection task

The neocortex can be viewed as a tapestry consisting of variations of rather stereotypical local cortical microcircuits. Hence understanding how these microcircuits compute holds the key to understanding brain function. Intense research efforts over several decades have culminated in a detailed model of a generic cortical microcircuit in the primary visual cortex from the Allen Institute. We are presenting here methods and first results for understanding computational properties of this large-scale data-based model. We show that it can solve a standard image-change-detection task almost as well as the living brain. Furthermore, we unravel the computational strategy of the model and elucidate the computational role of diverse subtypes of neurons. Altogether this work demonstrates the feasibility and scientific potential of a methodology based on close interaction of detailed data and large-scale computer modelling for understanding brain function.

Change detection based on dimension reduction SLIC and image matching for remote sensing images

The detection accuracy of traditional change detection algorithms is seriously affected by the low accuracy and high rate of omission, the radiometric correction accuracy, and the classification threshold for difference image. A change detection method based on image segmentation and image matching was proposed for remote sensing images. In this method, super-pixel-based dimension reduction SLIC image segmentation algorithm and SURF algorithms were used. The homogeneous region was used as the segmentation standard, and the homogeneity method was proposed to suppress the impact of inconsistent image segmentation on the change detection results. The experimental results show that this method improves the accuracy of remote sensing image change detection, has good robustness to the problem of redundant data, significantly reduces the error detection rate of image change detection, and can effectively accelerate the speed of change detection.

SAR Images Change Detection based on Sparse Coding and NMF

There are many applications for SAR image change detection, from military and agriculture to detection and management. But in fact, there is the speckle noise in SAR images inevitably. Therefore, the difficulty to detect change is increased. For purpose of reducing the interference of noise, we propose an unsupervised feature learning method using the non-negative matrix factorization algorithm and an improved sparse coding algorithm. First, non-negative matrix factorization method is used to obtain a dictionary which contains spatial structure information. Then, in order to increase the discriminate ability, we extract features for each pixel and apply sparse coding. Finally, the result of SAR image change detection is generated by applying simple k-means clustering method to divide the learned features into two different clusters. The superior performance of the proposed method is verified on several real SAR image datasets through comparisons with several existing change detection techniques.

An Image Change Detection Algorithm Based on Multi-Feature Self-Attention Fusion Mechanism UNet Network

Various UNet architecture-based image change detection algorithms promote the development of image change detection, but there are still some defects. First, under the encoder–decoder framework, the low-level features are extracted many times in multiple dimensions, which generates redundant information; second, the relationship between each feature layer is not modeled so sufficiently that it cannot produce the optimal feature differentiation representation. This paper proposes a remote image change detection algorithm based on the multi-feature self-attention fusion mechanism UNet network, abbreviated as MFSAF UNet (multi-feature self-attention fusion UNet). We attempt to add multi-feature self-attention mechanism between the encoder and decoder of UNet to obtain richer context dependence and overcome the two above-mentioned restrictions. Since the capacity of convolution-based UNet network is directly proportional to network depth, and a deeper convolutional network means more training parameters, so the convolution of each layer of UNet is replaced as a separated convolution, which makes the entire network to be lighter and the model’s execution efficiency is slightly better than the traditional convolution operation. In addition to these, another innovation point of this paper is using preference to control loss function and meet the demands for different accuracies and recall rates. The simulation test results verify the validity and robustness of this approach.