Feature-Level Change Detection Using Deep Representation and Feature Change Analysis for Multispectral Imagery

Zhang, Hui; Gong, Maoguo; Zhang, Puzhao; Su, Linzhi; Shi, Jiao

doi:10.1109/lgrs.2016.2601930

Cited by 134 publications

(57 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The triangular membership function is used for the output. The range of the output image is taken as Hui, et al [1]. The pixel is made white if it belongs to uniform region.…”

Section: Resultsmentioning

confidence: 99%

Unsupervised Change Detection of Multispectral Imagery Using Multi Level Fuzzy Based Deep Representation

Usha

Vasuki

2017

Journal of Asian Scientific Research

View full text Add to dashboard Cite

Article HistoryChange detection in remote sensing images provides useful information for various applications. This paper proposes a robust methodology for the analysis of multispectral imagery using Deep belief network (DBN) and Fuzzy interference system (FIS). Initially Euclidean distance and cosine angle distance features are extracted from the image. Deep learning is a robust machine learning method in which the extracted features are processed through set linear mapping and the changes are detected. However, the coarse spatial resolution indicating the intensity of modifications in class proportion instead of accounting for the change using discrete land covers classes is used in fuzzy image classification. Hence, the FIS is combined with DBN which allows defining our own rules to detect the changes accurately. It uses triangular membership function to plot the changes. The experimental results show that the proposed method enhanced the change detection by improving the performance parameters.

show abstract

“…The triangular membership function is used for the output. The range of the output image is taken as Hui, et al [1]. The pixel is made white if it belongs to uniform region.…”

Section: Resultsmentioning

confidence: 99%

Unsupervised Change Detection of Multispectral Imagery Using Multi Level Fuzzy Based Deep Representation

Usha

Vasuki

2017

Journal of Asian Scientific Research

View full text Add to dashboard Cite

show abstract

“…[23] proposed an ensemble system based on multiple classifiers, and achieved good classification results. Zhang et al (2017) [24] combined deep learning with feature change analysis for remote sensing image change detection, and the results confirmed that this new method is superior to the traditional methods. Despite the advantages of supervised classifiers in classification and change detection, they require training samples that are labeled beforehand.…”

mentioning

confidence: 90%

Object-Based Change Detection Using Multiple Classifiers and Multi-Scale Uncertainty Analysis

et al. 2019

View full text Add to dashboard Cite

The drawback of pixel-based change detection is that it neglects the spatial correlation with neighboring pixels and has a high commission ratio. In contrast, object-based change detection (OBCD) depends on the accuracy of the segmentation scale, which is of great significance in image analysis. Accordingly, an object-based approach for automatic change detection using multiple classifiers and multi-scale uncertainty analysis (OB-MMUA) in high-resolution (HR) remote sensing images is proposed in this paper. In this algorithm, the gray-level co-occurrence matrix (GLCM), morphological, and Gabor filter texture features are extracted to construct the input data, along with the spectral features, to utilize the respective advantages of the features and to compensate for the insufficient spectral information. In addition, random forest is used to select the features and determine the optimal feature vectors for the change detection. Change vector analysis (CVA) based on uncertainty analysis is then implemented to select the initial training samples. According to the diversity, support vector machine (SVM), k-nearest neighbor (KNN), and extra-trees (ExT) classifiers are then chosen as the base classifiers for Dempster-Shafer (D-S) evidence theory fusion, and unlabeled samples are selected using an active learning method with spatial information. Finally, multi-scale object-based D-S evidence theory fusion and uncertainty analysis is used to classify the difference image. To validate the proposed approach, we conducted experiments using multispectral images collected by the ZY-3 and GF-2 satellites. The experimental results confirmed the effectiveness and superiority of the proposed approach, which integrates the respective advantages of the pixel-based and object-based methods.Remote Sens. 2019, 11, 359 2 of 17 fields, such as urban development [6], environmental monitoring, vegetation coverage studies [7], and land-use monitoring [8,9].Change detection can be divided into pixel-based and object-based change detection (OBCD). Many scholars have proposed a variety of pixel-based change detection methods, including change vector analysis (CVA) [10,11], Markov random field (MRF)-based change detection [12], etc. With the development of machine learning, methods such as the extreme learning machine (ELM) algorithm [13], support vector machine (SVM) [14], random forest (RF) [15], k-nearest neighbor (KNN) [16], the multi-layer perceptron neural network (MLPNN) [17], and convolutional neural network (CNN) [18,19] have improved the accuracy of classification and change detection. However, a single classifier cannot detect all the change information in an image effectively. To address this issue, ensemble learning has been applied to the research and application of change detection and classification [20][21][22]. [23] proposed an ensemble system based on multiple classifiers, and achieved good classification results. Zhang et al. (2017) [24] combined deep learning with feature change analysis for remote sensing image change d...

show abstract

“…A DCNN can achieve a superior performance compared to conventional classification algorithms. A restricted Boltzmann machine (RBM) [19], a convolutional neural network (CNN) [20][21][22], and deep belief networks (DBNs) [23] have been proposed for use in change detection. Such change detection algorithms based on deep learning yield a relatively good performance in terms of the detection accuracy.…”

Section: Introductionmentioning

confidence: 99%

Fusion Network for Change Detection of High-Resolution Panchromatic Imagery

Wiratama

Sim

2019

Applied Sciences

View full text Add to dashboard Cite

This paper proposes a fusion network for detecting changes between two high-resolution panchromatic images. The proposed fusion network consists of front-and back-end neural network architectures to generate dual outputs for change detection. Two networks for change detection were applied to handle image-and high-level changes of information, respectively. The fusion network employs single-path and dual-path networks to accomplish low-level and high-level differential detection, respectively. Based on two dual outputs, a two-stage decision algorithm was proposed to efficiently yield the final change detection results. The dual outputs were incorporated into the two-stage decision by operating logical operations. The proposed algorithm was designed to incorporate not only dual network outputs but also neighboring information. In this paper, a new fused loss function was presented to estimate the errors and optimize the proposed network during the learning stage. Based on our experimental evaluation, the proposed method yields a better detection performance than conventional neural network algorithms, with an average area under the curve of 0.9709, percentage correct classification of 99%, and Kappa of 75 for many test datasets.

show abstract

Feature-Level Change Detection Using Deep Representation and Feature Change Analysis for Multispectral Imagery

Cited by 134 publications

References 12 publications

Unsupervised Change Detection of Multispectral Imagery Using Multi Level Fuzzy Based Deep Representation

Unsupervised Change Detection of Multispectral Imagery Using Multi Level Fuzzy Based Deep Representation

Object-Based Change Detection Using Multiple Classifiers and Multi-Scale Uncertainty Analysis

Fusion Network for Change Detection of High-Resolution Panchromatic Imagery

Contact Info

Product

Resources

About