Nonparametric classification based on local mean and class statistics

Zeng, Yong; Yang, Yupu; Zhao, Liang

doi:10.1016/j.eswa.2008.10.041

Cited by 27 publications

(6 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Typical examples of the local transformation method include the local LDA method for multimodal data projection [13], the approach to use the local dependencies of samples for classification [14], and et al [15][16][17][18][19]. It is clear that the local transformation methods can reduce the computational load, however, the local PCA was also found more efficient than the global PCA in feature extraction [20].…”

Section: Introductionmentioning

confidence: 99%

M-nearest neighbor selection for two-phase test sample representation in face recognition

Liang³

2012

Proceedings of the 10th World Congress on Intelligent Control and Automation

View full text Add to dashboard Cite

As a powerful algorithm for face recognition, the proposed Two-Phase Test Sample Representation (TPTSR) increases the classification rate by dividing the recognition task into two steps. The first step intends to find the M most possible candidate training samples from the whole training set to match with the testing input, and the second phase classifies the testing sample to the class with the most representative linear combination by the selected training samples in the first phase. However, the linear representation criterion for selecting the M nearest neighbors in the first phase is too computational demanding, especially when the training set as well as the number of classes is large. Therefore, more straight forward and simplified criterions for the nearest neighbor selection are considered, such as the Euclidean distance and the City-block distance. The experimental results show that the TPTSR method with the Euclidean distance and the City-block distance criterions can achieve almost the same classification performance as the linear representation; they are much more efficient in reducing the computation time.

show abstract

Section: Introductionmentioning

confidence: 99%

M-nearest neighbor selection for two-phase test sample representation in face recognition

Liang³

2012

Proceedings of the 10th World Congress on Intelligent Control and Automation

View full text Add to dashboard Cite

show abstract

“…Hence, KNN has been widely studied. Many KNN-based classification methods, such as [2,[4][5][6], have been proposed.…”

Section: Introductionmentioning

confidence: 99%

A pre-averaged pseudo nearest neighbor classifier

2024

Preprint

View full text Add to dashboard Cite

K nearest neighbor rule is a powerful classification method. However, its classification performance will be affected in the situation of small-size samples with existing outliers. To address this issue, a pre-averaged pseudo nearest neighbor classifier (PAPNN) is proposed to improve classification performance. In the PAPNN rule, pre-averaged categorical vectors are calculated by taking the average of any two points of the training sets in each class, and then k pseudo nearest neighbors are chosen from the preprocessed vectors of every class to determine the category of a query point. The pre-averaged vectors can reduce the negative impact of outliers in some degree. Extensive experiments are carried out on nineteen numerical real data sets and three artificial data sets by comparing PAPNN to other five KNN-based methods. The experimental results demonstrate that the proposed PAPNN rule is effective for classification task in the case of small-size samples with existing outliers.

show abstract

“…Feature land cover classification algorithms are challenged by the complex spectral properties and high-dimensional features of hyperspectral data. Programs such as Spectral Angle Mapper [12], Support Vector Machines [13,14], Decision Tree [15], and K-Nearest Neighbours [16] identified HSI in the first stage of the process based solely on the similarity or difference between their spectra. In situations when there is a considerable amount of spectral overlap or similarity, techniques that solely consider spectral features have not worked effectively.…”

Section: Introductionmentioning

confidence: 99%

“…Many researchers have successfully combined Remote Sens. 2024, 16, 287 3 of 23 transformers with other components of CNNs, achieving state-of-the-art results in image classification tasks [47][48][49][50][51][52].…”

Section: Introductionmentioning

confidence: 99%

Dilated Spectral–Spatial Gaussian Transformer Net for Hyperspectral Image Classification

Zhang,

Wang,

Zhang

2024

Remote Sensing

View full text Add to dashboard Cite

In recent years, deep learning-based classification methods for hyperspectral images (HSIs) have gained widespread popularity in fields such as agriculture, environmental monitoring, and geological exploration. This is owing to their ability to automatically extract features and deliver outstanding performance. This study provides a new Dilated Spectral–Spatial Gaussian Transformer Net (DSSGT) model. The DSSGT model incorporates dilated convolutions as shallow feature extraction units, which allows for an expanded receptive field while maintaining computational efficiency. We integrated transformer architecture to effectively capture feature relationships and generate deep fusion features, thereby enhancing classification accuracy. We used consecutive dilated convolutional layers to extract joint low-level spectral–spatial features. We then introduced Gaussian Weighted Pixel Embedding blocks, which leverage Gaussian weight matrices to transform the joint features into pixel-level vectors. By combining the features of each pixel with its neighbouring pixels, we obtained pixel-level representations that are more expressive and context-aware. The transformed vector matrix was fed into the transformer encoder module, enabling the capture of global dependencies within the input data and generating higher-level fusion features with improved expressiveness and discriminability. We evaluated the proposed DSSGT model using five hyperspectral image datasets through comparative experiments. Our results demonstrate the superior performance of our approach compared to those of current state-of-the-art methods, providing compelling evidence of the DSSGT model’s effectiveness.

show abstract

Nonparametric classification based on local mean and class statistics

Cited by 27 publications

References 8 publications

M-nearest neighbor selection for two-phase test sample representation in face recognition

M-nearest neighbor selection for two-phase test sample representation in face recognition

A pre-averaged pseudo nearest neighbor classifier

Dilated Spectral–Spatial Gaussian Transformer Net for Hyperspectral Image Classification

Contact Info

Product

Resources

About