Statistical dependence of pixel intensities for pattern recognition

Smielik, Ievgen; Kuhnert, Klaus-Dieter

doi:10.1109/icit.2013.6505840

Cited by 5 publications

(3 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previous study also suggests that we may exploit only a subset of all image pixels for image classification [39]. This somewhat implies that different pixels play different roles in image classification.…”

Section: The Analysis Of the Proposed Algorithmmentioning

confidence: 97%

Multiple representations and sparse representation for image classification

Zhang

Zhong

2015

Pattern Recognition Letters

View full text Add to dashboard Cite

Section: The Analysis Of the Proposed Algorithmmentioning

confidence: 97%

Multiple representations and sparse representation for image classification

Zhang

Zhong

2015

Pattern Recognition Letters

View full text Add to dashboard Cite

“…According to Ref. [43], we can set different weights for different pixel values to perform image classification tasks since different pixel points play different roles in image classification. And our proposed algorithm achieves more attention to pixel values for medium intensity, which justifies our proposed image representation from the other side.…”

Section: Algorithm Analysismentioning

confidence: 99%

Sparse representation scheme with enhanced medium pixel intensity for face recognition

Zhang

Wang

et al. 2023

CAAI Trans on Intel Tech

View full text Add to dashboard Cite

Sparse representation is an effective data classification algorithm that depends on the known training samples to categorise the test sample. It has been widely used in various image classification tasks. Sparseness in sparse representation means that only a few of instances selected from all training samples can effectively convey the essential class‐specific information of the test sample, which is very important for classification. For deformable images such as human faces, pixels at the same location of different images of the same subject usually have different intensities. Therefore, extracting features and correctly classifying such deformable objects is very hard. Moreover, the lighting, attitude and occlusion cause more difficulty. Considering the problems and challenges listed above, a novel image representation and classification algorithm is proposed. First, the authors’ algorithm generates virtual samples by a non‐linear variation method. This method can effectively extract the low‐frequency information of space‐domain features of the original image, which is very useful for representing deformable objects. The combination of the original and virtual samples is more beneficial to improve the classification performance and robustness of the algorithm. Thereby, the authors’ algorithm calculates the expression coefficients of the original and virtual samples separately using the sparse representation principle and obtains the final score by a designed efficient score fusion scheme. The weighting coefficients in the score fusion scheme are set entirely automatically. Finally, the algorithm classifies the samples based on the final scores. The experimental results show that our method performs better classification than conventional sparse representation algorithms.

show abstract

“…The first one relies on computing statistical values and suffer from rotation irresistance, insufficiency of number of features and sensibility of image noise. Such methods are [1], [2], [3]. In contrast to this, the spectral group of methods effectively measure image energy, generates rotataion resistant image feature vectors which cannot be influenced by image noise.…”

Section: Introductionmentioning

confidence: 99%

Efficiency comparative analysis between two search algorithms using DT CWT for content-based image retrieval

Vetova

Ivanov

2015

2015 IEEE International Conference on Industrial Technology (ICIT)

View full text Add to dashboard Cite

The following paper presents a comparative analysis on the efficiency of two search algorithms for Content-Based Image Retrieval. Both algorithms are designed using the Dual-Tree Complex Wavelet Transform for image feature extraction and Hausdorff distance for similarity distance computation. The difference in the steps of the algorithms produces difference in the final result. To estimate the efficiency of the algorithms, we performed some experiments and compared their results. They clearly show that one of the algorithms provides much higher retrieval rate and system performance than the other.

show abstract

Statistical dependence of pixel intensities for pattern recognition

Cited by 5 publications

References 8 publications

Multiple representations and sparse representation for image classification

Multiple representations and sparse representation for image classification

Sparse representation scheme with enhanced medium pixel intensity for face recognition

Efficiency comparative analysis between two search algorithms using DT CWT for content-based image retrieval

Contact Info

Product

Resources

About