Video Summarization Using Feature Vector Clustering

Color is the most widely used visual feature in content based image retrieval. The visual coherence color space, HSV, is adopted to represent image. Hue component is used to denote color. Hue difference statistic is proposed to extract color change information as supplement to color feature. The image is divided into sub images equally. Color and change information is extracted in each region. After feature vector clustering and coding, image content can be expressed as vector codes. The text based analysis technology is used for image retrieval. Experiments show that the proposed method can realize efficient retrieval for unconstrained scene images.

Download Full-text

Feature vector clustering molecular pairs in computer simulations

Journal of Computational Chemistry ◽

10.1002/jcc.26028 ◽

2019 ◽

Vol 40 (29) ◽

pp. 2539-2549 ◽

Cited By ~ 1

Author(s):

Han‐Wen Pei ◽

Aatto Laaksonen

Keyword(s):

Computer Simulations ◽

Feature Vector ◽

Vector Clustering

Download Full-text

A comparison of three feature vector clustering procedures in a speech recognition paradigm

10.1109/icassp.1983.1172085 ◽

2005 ◽

Cited By ~ 4

Author(s):

L. Niles ◽

H. Silverman ◽

N. Dixon

Keyword(s):

Speech Recognition ◽

Feature Vector ◽

Vector Clustering

Download Full-text

HYPERSPECTRAL IMAGE BAND SELECTION BASED ON SUBSPACE CLUSTERING

International Journal of Engineering Technologies and Management Research ◽

10.29121/ijetmr.v8.i8.2021.1014 ◽

2021 ◽

Vol 8 (8) ◽

pp. 42-51

Author(s):

Zhi-jun ZHENG ◽

Yan-bin PENG

Keyword(s):

Land Cover ◽

Hyperspectral Image ◽

Feature Vector ◽

Computation Time ◽

Subspace Clustering ◽

Small Sample ◽

Hyperspectral Data ◽

Band Selection ◽

Svm Classifier ◽

Vector Clustering

Aiming at the problems in hyperspectral image classification, such as high dimension, small sample and large computation time, this paper proposes a band selection method based on subspace clustering, and applies it to hyperspectral image land cover classification. This method considers each band image as a feature vector, clustering band images using subspace clustering method. After that, a representative band is selected from each cluster. Finally feature vector is formed on behalf of the representative bands, which completes the dimension reduction of hyperspectral data. SVM classifier is used to classify the new generated sample points. Experimental data show that compared with other methods, the new method effectively improves the accuracy of land cover recognition.

Download Full-text