BoVSG: bag of visual SubGraphs for remote sensing scene classification

Amiri, Khitem; Farah, Mohamed; Leloglu, Ugur Murat

doi:10.1080/01431161.2019.1681602

Cited by 5 publications

(3 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Further Kernel Collaborative Representation-based Classification (KCRC) is used on extracted global and local features where the aerial images are annotated as per minimum approximation residual after the fusion step. The method of Bag of Visuals Subgraphs (BoVSG) is proposed in [5] where segmentation of the image is done into superpixels containing only relevant information, such type of superpixels collected from each image clustered as per there texture and color features and assigned with a land usage label. Superpixels belonging to the identical cluster should have the identical land usage label.…”

Section: Literature Reviewmentioning

confidence: 99%

Identifying Land Usage from Aerial Image using Feature Fusion of Thepade’s Sorted n-ary Block Truncation Coding and Bernsen Thresholding with Ensemble Methods

Thepade,

Chaudhari,

Das

2020

IJEAT

View full text Add to dashboard Cite

Automatic Land Usage Identification is one of the most demanded research areas in Remote Sensing. One of the primitive sources for Land Usage Identification is Aerial images. Automatic Land Usage Identification is implemented by exploring different feature extraction methods whereas, these features are categorized into local and global content description of image. Fusion of local and global features may be a potential approach for land usage identification. Accordingly, the major contribution of work presented here is fusion of global color features extracted using TSBTC n-ary method (applied on entire image) and local features extracted using Bernsen thresholding method applied on 3*3 windows of image for land usage identification. Consideration of more than one machine learning classifiers as an ensemble has shown better results than that of individual machine learning classifiers. In proposed work here, Thepade’s Sorted n-ary Block Truncation Coding (TSBTC n-ary) is explored in aerial image feature extraction with nine variations from TSBTC 2-ary till TSBTC 10-ary. The performance appraise of proposed Land Usage Identification technique is done using UC Merced Dataset having 2100 images categorized into 21 land usage types. In consideration performance measures like Accuracy, F Measure and Matthews Correlation Coefficient (MCC); the TSBTC 10-ary global features extraction method has given better land usage identification as compare to Bernsen thresholding local feature extraction method. The proposed method enhances the identification of land usage through feature level fusion of TSBTC 10-ary global features and Bernsen thresholding local features. Along with nine individual machine learning algorithms, ensembles of varied machine learning algorithms are used for further performance improvement of the proposed land usage identification technique.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Identifying Land Usage from Aerial Image using Feature Fusion of Thepade’s Sorted n-ary Block Truncation Coding and Bernsen Thresholding with Ensemble Methods

Thepade,

Chaudhari,

Das

2020

IJEAT

View full text Add to dashboard Cite

show abstract

“…For scene-level classification, the performance of the classifier largely depends on the feature extraction ability for remote sensing images [23]. For example, the histogram of an image can be used as a low-level summary of its features, but the classification accuracies obtained based on these low-level features are hardly satisfactory [24].…”

Section: Introductionmentioning

confidence: 99%

Identifying Urban Wetlands through Remote Sensing Scene Classification Using Deep Learning: A Case Study of Shenzhen, China

Yang

Luo

Ren

et al. 2022

IJGI

View full text Add to dashboard Cite

Urban wetlands provide cities with unique and valuable ecosystem services but are under great degradation pressure. Correctly identifying urban wetlands from remote sensing images is fundamental for developing appropriate management and protection plans. To overcome the semantic limitations of traditional pixel-level urban wetland classification techniques, we proposed an urban wetland identification framework based on an advanced scene-level classification scheme. First, the Sentinel-2 high-resolution multispectral image of Shenzhen was segmented into 320 m × 320 m square patches to generate sample datasets for classification. Next, twelve typical convolutional neural network (CNN) models were transformed for the comparison experiments. Finally, the model with the best performance was used to classify the wetland scenes in Shenzhen, and pattern and composition analyses were also implemented in the classification results. We found that the DenseNet121 model performed best in classifying urban wetland scenes, with overall accuracy (OA) and kappa values reaching 0.89 and 0.86, respectively. The analysis results revealed that the wetland scene in Shenzhen is generally balanced in the east–west direction. Among the wetland scenes, coastal open waters accounted for a relatively high proportion and showed an obvious southward pattern. The remaining swamp, marsh, tidal flat, and pond areas were scattered, accounting for only 4.64% of the total area of Shenzhen. For scattered and dynamic urban wetlands, we are the first to achieve scene-level classification with satisfactory results, thus providing a clearer and easier-to-understand reference for management and protection, which is of great significance for promoting harmony between humanity and ecosystems in cities.

show abstract

“…Subsequently, a sparse principal component analysis was harnessed to assimilate category-specific information and facilitate comprehensive classification. The widely recognized bag-of-visual-words (BoVW) paradigm found application in remote sensing scene classification assignments, driven by its straightforwardness and efficacy [46]- [48]. Cheriyadat et al [14] employed sparse coding to acquire a suite of bias functions from images.…”

mentioning

confidence: 99%

Large Kernel Separable Mixed ConvNet for Remote Sensing Scene Classification

Zhang,

Cui,

et al. 2024

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Among tasks related to intelligent interpretation of remote sensing data, scene classification mainly focuses on the holistic information of the entire scene. Compared to pixel-level or object-based tasks, it involves a richer semantic context, making it more challenging. With the rapid advancement of deep learning, convolutional neural networks (CNNs) have found widespread applications across various domains, and some work has introduced them into scene classification tasks. However, traditional convolution operations involve sliding small convolutional kernels across an image, primarily focusing on local details within a small receptive field. To achieve better modeling of the entire image, the smaller receptive field limits the ability of convolution operation to capture features over a broader range. To this end, we introduce a large kernel CNNs into the scene classification task to expand the receptive field of the mode, which allows us to capture comprehensive non-local information while still acquiring rich local details. However, in addition to encoding spatial association, the effective information within the feature maps is also strongly channel-related. Therefore, to fully model this channel dependency, a novel channel separation and mixing module has been designed to realize feature correlation in the channel dimension. The combination of them forms a Large Kernel Separable Mixed Convnet (LSMNet), enabling the model to capture effective dependencies of feature maps in both spatial and channel dimensions, thus achieving enhanced feature expression. Extensive experiments conducted on three datasets have also validated the effectiveness of the proposed method.

show abstract

BoVSG: bag of visual SubGraphs for remote sensing scene classification

Cited by 5 publications

References 51 publications

Identifying Land Usage from Aerial Image using Feature Fusion of Thepade’s Sorted n-ary Block Truncation Coding and Bernsen Thresholding with Ensemble Methods

Identifying Land Usage from Aerial Image using Feature Fusion of Thepade’s Sorted n-ary Block Truncation Coding and Bernsen Thresholding with Ensemble Methods

Identifying Urban Wetlands through Remote Sensing Scene Classification Using Deep Learning: A Case Study of Shenzhen, China

Large Kernel Separable Mixed ConvNet for Remote Sensing Scene Classification

Contact Info

Product

Resources

About