Grouped Collaborative Representation for Hyperspectral Image Classification Using a Two-Phase Strategy

Shen, Xiangfei; Bao, Wenxing; Liang, Hongbo; Zhang, Xiaowu; Ma, Xuan

doi:10.1109/lgrs.2021.3070074

Cited by 11 publications

(6 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…S7: Pre-classify the features by SVM to assign appropriate feature weights to each future. S8: Superpixel-guided RCR classification for each superpixel region k up to K 1) Solve the RCR problem from ( 16) to (20) and find coefficient matrices {X v k }. 2) Evaluate the residuals by (21).…”

Section: E Relaxed Collaborative Representationmentioning

confidence: 99%

“…Zhang et al indicated that collaborative representation of training samples is important as sparsity. Therefore, several CRC-based HSIC methods have been proposed [16]- [20]. [16] uses a distance-weighted Tikhonov regularization Manuscript received 14 Dec 2023.…”

Section: Introductionmentioning

confidence: 99%

“…[19] proposed weighted regularized collaborative representation optimized classifier (WRCROC) which combines the advantages of CRC, NRS and WJCR. Shen et al calculates class-specific group weights of train samples to reveal the impact of different classes and proposed grouped CRC (GCRC) [20] algorithm.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Guided Filter of Random Patches Network and Relaxed-Collaborative-Representation-Based Hyperspectral Image Classification

Dundar,

Ince

2024

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

View full text Add to dashboard Cite

Feature extraction and accurate classification are crucial tasks in land-cover classification of hyperspectral image (HSI). We propose guided filter (GF) of random patches network (RPNet) and relaxed collaborative representation (RCR) based HSI classification (HSIC) method called GRR. The shallow and deep features are extracted using RPNet that requires no pretraining stage. In addition to the obtained feature set, the original HSI and extracted features are then filtered by GF to preserve the edge details. After that, all the distinct feature sets are separately concatenated with the original HSI to keep the original structure of the data. The high-dimensional feature sets are then processed by linear discriminant analysis (LDA) to increase class separability and to select the most representative features. Since few train samples are available in HSIC task, efficiency of LDA is improved using superpixel segmentation to generate pseudosamples. As final stage, the reduced-dimension feature sets are classified by the use of superpixel-guided RCR which utilize the resemblance and discrimination of the feature sets efficiently. The extensive experiments on the real HSIs are carried out to validate the efficacy of the proposed method.

show abstract

Section: E Relaxed Collaborative Representationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Guided Filter of Random Patches Network and Relaxed-Collaborative-Representation-Based Hyperspectral Image Classification

Dundar,

Ince

2024

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

View full text Add to dashboard Cite

show abstract

“…e first part refers to supervised learning-based methods that require an independent training process and predict testing data using an already learned model. e representative supervised models include sparse/collaborative representation [8][9][10], support vector machine [11][12][13], ensemble learning [14][15][16][17], and so on. e second part refers to unsupervised learningbased models that do not demand training samples and determine entire classes by considering the correlations among samples.…”

Section: Introductionmentioning

confidence: 99%

Application of Decision Tree‐Based Classification Algorithm on Content Marketing

Yang

2022

Journal of Mathematics

View full text Add to dashboard Cite

Traditional content marketing methods resort grossly to market requirements but barely obtain relatively accurate marketing prediction under loads of requirements. Machine learning-based approaches nowadays are widely used in multiple fields as they involve a training process to deal with big data problems. In this paper, decision tree-based methods are introduced to the field of content marketing, and decision tree-based methods intrinsically follow the process of human decision making. Specifically, this paper considers a well-known method, called C4.5, which can deal well with continuous values. Based on four validation metrics, experimental results obtained from several machine learning-based methods indicate that the C4.5-based decision tree method has the ability to handle the content marketing dataset. The results show that the decision tree-based method can provide reasonable and accurate suggestions for content marketing.

show abstract

“…Unsupervised learning, also known as the clustering method, does not need training samples but only interprets the data by exploring the structure and correlation information between the input data, including K-means [16], ISODATA [17], DBSCAN [18], Fuzzy C-Means [19], etc. Supervised learning needs a group of training samples to train the model, which has obtained the optimal model parameters, and then transplants the trained model to the test samples to observe the behaviors of test samples, including sparse/collaborative representation [20], ensemble learning [21], support vector machine [22]. Unlike unsupervised and supervised learning, semisupervised learning introduces some unlabeled samples into the training process for the sake of improving the robustness of method.…”

Section: Introductionmentioning

confidence: 99%

Integration of SOM and PCA for Analyzing Sports Economic Data and Designing a Management System

Cao

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

Currently, sports economic data have attracted more attention because they normally exist with a high dimensionality manner that reflects the historical behavior and the potential decision trend of users or players. Traditional analysis techniques dealing with such kinds of data rely heavily on the empirical knowledge of the manager. With the development of data science, traditional experience-based knowledge barely meets the requirements of multiple features and high-dimensional data analysis. In this regard, machine learning-based data analysis techniques nowadays can give more importance to the process of extracting latent information hidden in chaotic data, which helps users make decisions and take appropriate actions. In this paper, we integrate principal component analysis (PCA) and a self-organizing map (SOM) to exploit the hidden features in the high-dimensional data. Specifically, PCA considers an orthogonal transformation operation to linearly transform the observed data to the low-dimensional one. SOM clusters the data by constructing a two-layered neural network without manual intervention and knowing the category in the training stage. The integration of PCA and SOM helps promote the research on pattern recognition and visualization of high-dimensional data. The experimental results obtained from economic data indicate the effectiveness of the combination of PCA and SOM.

show abstract

Grouped Collaborative Representation for Hyperspectral Image Classification Using a Two-Phase Strategy

Cited by 11 publications

References 14 publications

Guided Filter of Random Patches Network and Relaxed-Collaborative-Representation-Based Hyperspectral Image Classification

Guided Filter of Random Patches Network and Relaxed-Collaborative-Representation-Based Hyperspectral Image Classification

Application of Decision Tree‐Based Classification Algorithm on Content Marketing

Integration of SOM and PCA for Analyzing Sports Economic Data and Designing a Management System

Contact Info

Product

Resources

About